Shelter Projects2011/2012

SHELTER PROJECTS
2011-2012
Introduction Shelter Projects 2011–2012
Foreword
SHELTER PROJECTS
2011–2012
www.ShelterCaseStudies.org i
Foreword Introduction
Shelter Projects 2011–2012
Published 2013
Available online from www.ShelterCaseStudies.org
Copyright for this book is retained by IFRC, UN-HABITAT and UNHCR.

Reproduction for non-profitable objectives is encouraged.
The copyright for the photographs and images remains with the photographers whose names are indicated on
each photograph.
DISCLAIMER
The maps contained in this publication are for illustrative purposes only and should not be considered
authoritative.
Whilst every effort has been made to ensure the accuracy and completeness of the content of this booklet, no
liability can be accepted for any errors or omissions contained within it.
The designations employed and the presentation of the material in this publication do not imply the expression
of any opinion whatsoever on the part of the Secretariat of the United Nations concerning the legal status of
any country, territory, city or area, or of its authorities, or concerning delimitation of its frontiers or boundaries,
or regarding its economic system or degree of development. The analysis, conclusions and recommendations
of the report do not necessarily reflect the views of the United Nations Human Settlements Programme, the
Governing Council of the United Nations Human Settlements Programme or its Member States.
Approximate prices are given in US Dollars (US$), based on exchange rates at the time of the project.
Front Cover: A transitional shelter in the Philippines being relocated by 20 people.

© Charisse Mae Borja, CRS.
Back cover left to right:

© Daniel N’dri Yao / Solidarités International, Heritage Foundation of Pakistan, Christian Jepsen, Richard Evans,
Thanchanitch Suttichote/IOM Thailand, Mikel Flamm, Arnaud Fratani, Sandra D’Urzo/IFRC, CRS.
ii
Introduction Shelter Projects 2011–2012 Foreword
Foreword
The fourth edition of ‘Shelter Projects’, is launched at a time when shelter is more relevant than ever as an
instrument of humanitarian response. The case studies in this edition reflect the on-going challenges posed by
responses to complex emergencies such as Haiti and Pakistan as well as new challenges derived from unprec-
edented level of population displacement in Africa, Asia and in the Middle East. While the increase of shelter needs
prompt larger mobilisation of resources, shelter programmes need to explore improved models of delivery as well
as innovative, cost-effective solutions which incorporate best practice and position the persons of concern at the
forefront of our interventions.
Where people live largely determines their ability to meet their basic needs. It is of paramount importance
that shelter solutions look beyond the physical structure and consider the environment within which the shelter is
placed. Nowadays, large displacement of population due to humanitarian crisis, mostly affects urban areas where
people expect to find easier access to opportunities be it of social or economic nature. Nevertheless, this trend
further exacerbates the existing urbanisation phenomenon by placing additional strain on already vulnerable areas.
It is therefore increasingly evident that new concepts for sheltering people have to incorporate a more holistic
approach which includes the shelter and its surrounding context. Shelter is an integral part of settlement planning,
which guides spatial allocation of functions maintaining equilibrium between population needs, availability and al-
location of resources, economic dynamics, amelioration of living conditions, provision of services, communication
transportation networks as well as recreational spaces.
The case studies contained in the fourth edition of Shelter Projects are a reminder once again that every crisis is
unique. There is no ‘silver bullet’ for shelter response. The main objective should be to operate in accordance with
recognized shelter best practice while enabling those displaced to return to their homes or equivalent living space
in a timely manner encouraging community recovery and building resilience to possible future shocks. Participation
and promoting ownership is the key to achieving successful projects.
As well as being an important reference point for shelter facilitators this publication also acts as a learning tool
allowing the successes and challenges of completed shelter projects to be replicated and improved on. The case
studies address common issues emerging in shelter response, outline different approaches to addressing shelter
needs and assist in evaluating the impact on affected communities. The shelter projects case studies provide an
excellent resource against which to gauge proposed shelter interventions and access possible outcomes. Let’s col-
lectively try to avoid ‘re-inventing the wheel’.
This edition also contains a new section comprising relevant thematic topics of interest compiled by technical
experts. The issues addressed are: a background to the indicator for covered living space ‘the 3.5m2 principle’, cash
transfers as a tool in shelter response, sheltering of livestock and the importance of settlements.
In keeping with developments in on-line information and social media, greater emphasis is being placed on
electronic dissemination, and in this regard the shelter projects website www.sheltercasestudies.org is identified
for reference on each page of the document. We welcome your feedback and hope you will utilize the website in
this regard.
A special thanks to those who contributed with case studies and the Technical Advisory Committee for the
articles of interest, without your support we would have no stories to tell. We trust that the reader will find this
edition of ‘Shelter projects’ relevant and thought provoking leading to improved shelter solutions for affected
communities.
Esteban Leon
Head Shelter and Rehabilitation Unit
Risk Reduction and Rehabilitation Branch Graham Saunders
UN-HABITAT Head
Shelter & Settlements Department
International Federation of Red Cross
Monica Noro and Red Crescent Societies
Chief of UNHCR Shelter and Settlement Section
United Nations High Commissioner for Refugees
www.ShelterCaseStudies.org iii
Acknowledgements Acknowledgements Introduction
Acknowledgements
Project coordinated by: Esteban Leon (UN–HABITAT), Miguel Urquia (UNHCR), Sandra D’Urzo (IFRC) and Joseph
Ashmore.
The Shelter Projects publication is overseen by a technical advisory group, including those mentioned above and
Eddie Argeñal (USAID), Jake Zarins (NRC), Jim Kennedy (Independent), Luca Pupulin (ACTED), Nuno Nunes (IOM),
and Seki Hirano (CRS).
Compiled and edited by: Joseph Ashmore, with additional editorial support from Jon Fowler, Jim Kennedy and
Wan Sophonpanich.
These 32 case studies have been provided from the programmes of the following 18 organisations:
ACTED
British Red Cross Society
CRAterre
Colombian Red Cross Society
CRS - Catholoc Relief Services
Habitat for Humanity
Heritage Foundation of Pakistan
IFRC- International Federation of the Red Cross and Red Crescent Societies
IOM - International Organisation for Migration
Netherlands Red Cross Society
NRC- Norwegian Refugee Council
Oxfam GB
Première Urgence
Shelterbox
Solidarités
UNHCR - United Nations High Commissioner for Refugees
UN-Habitat - United Nations Human Settlements Programme
USAID/OFDA
The editors would like to express their gratitude to the following individuals who wrote and reviewed the case
studies in this book:
Alexandre Douline, Alexis Bricogne, Alison Ashlin, Amanda Rashid, Amelia Rule, Andrew Schaefer, Andy Buchanan,
Anicet Adjahossou, Anna Pont, Annalisa Caïmi, Antoine Vollet, Arnaldo Arcadio, Arnaud Fratani, Bettina
Morgenstern, Brian Kelly, Carlo Gerhardi, Cesar Augusto Uruena Pulido, Charles Campbell, Charles Parrack,
Charles Setchell, Christina Belinga, Corinne Treherne, Damien Jusselme, Ed Cook, Edwin Pinto Ladino, Eef Marije
Krol, Elias Jourdy, Elizabeth Palmer, Elizabeth Troman, Elsa Cauderay, Fernando Murillo, Florence Foster, Garrett
Paolo Nolasco, Ghada Ajami, Giovanni Cassani, Habibullah Mujaddidi, Hilary DuBose, Hovig Etyemezian, Javier
Gonzales, Jamie Mok, Jeanne Ella Andrianambinina, Jennifer Ankrom, Jeremiah Atho Ougo, Jockny Martinez,
Joseph Curry, Julia Macro, Julien Mulliez, Karl Baker, Kifle Tesfa, Ko Takeuchi, Laura Dills, LeGrand Lee Malany,
Leonilo Escalada, Manuel Moniz Pereira, Maria Moita, Mario Flores, Mariyam Nizam, Mark Hopley, Martino Costa,
Michiko Ito, Neil Brighton, Niek de Goeij, Olivier Moles, Ophélie Goulias, Patrick Mutai, Phil Duloy, Philippe Garnier,
Pornphun Sutthiprapha, Rekha Das, Richard Evans, Rick Bauer, Salve Pacatang, Shintaro Yamamoto, Subhash
Jadhav, Thanchanitch Suttichote, Timothy Mutunga, Tom Nickalls, William Schmitt, Yasmeen Lari and Zabihullah
Afghani.
We would like to thank those who contributed to Shelter Projects 2008, Shelter Projects 2009, and Shelter Projects
2010 whose work is reflected in this document.
We also would like to thank the many hundreds of people who have implemented the projects that are mentioned
in this book, but who have not been individually credited.
Photographs are reproduced by kind permission of those whose names appear next to them in the text.
iv
Introduction Shelter Projects 2011–2012 Contents
Contents
Contents
Foreword iii
Acknowledgements iv
Contents v
Introduction vii
Section A – Case Studies

A.1 Afghanistan – 2012 – Conflict Returns 3
A.2 Burkina Faso – 2012 – Conflict 6
A.3 Colombia – 2010–2011 – Floods 9
A.4 Côte d’Ivoire – 2010–2011 – Post-electoral Crisis 12
A.5 Côte d’Ivoire – 2010–2011 – Post–Electoral Crisis 15
A.6 Côte d’Ivoire – 2010–2011 – Post–Electoral Crisis 18
A.7 Democratic Republic of Congo – 2002 – Volcano 21
A.8 Ethiopia – 2011 – Sudanese Conflict 24
A.9 Ethiopia – 2012 – Conflict and Drought 27
A.10 Haiti – 2010 – Earthquake 31
A.14 Japan – 2011 – Earthquake and Tsunami 44
A.15 Kenya (Dadaab) – 2011 – Famine / Conflict in Somalia 48
A.16 Lebanon – 2007 – Conflict 53
A.17 Lebanon – 2011 – Conflict 56
A.18 Madagascar – 2012 – Tropical storm 59
A.19 Nicaragua – 2007 – Hurricane 62
A.20 Pakistan – 2010 – Floods 65
A.24 Peru – 2012 – Flooding and land slides 78
A.25 Philippines – 2011 – Cyclone 81
A.28 Somalia – 2011 – Famine / Conflict 89
A.29 Republic of South Sudan – 2011 – Conflict 93
A.30 Thailand – 2011 – Floods 96
A.31 Tunisia – 2011 – Conflict in Libya 99
A.32 USA (Chicago) – 1871 – Fire 103
Section B – Opinions
B.1 The History of Three Point Five Square Metres 109
B.2 Bankers and Builders 112
B.3 Livestock Sheltering in Humanitarian Situations 115
B.4 A Reflection on the Importance of Settlements in Humanitarian
Shelter Assistance 118
Annexes
Annex 1 – Index – by country 123
Annex 2 – Conversion Tables 126
Annex 2 – Further reading 127
www.ShelterCaseStudies.org v
Contents Contents Introduction
vi
Introduction Shelter Projects 2011–2012 Introduction
Introduction
This book contains summaries of Selection of case studies meet shelter need, as providing
shelter projects that have been im- The case studies were selected shelter is more than simply
plemented in response to conflicts, using the following criteria: designing architecturally
complex emergencies, and natural impressive structures.
disasters (Section A). It also contains Projects must be wholly or
largely complete by the end In compiling the case studies
a section entitled Opinions (Section for this edition, special efforts were
B). These are summaries of signifi- of 2012. This is to allow solid
learnings to be gained. made to include projects which
cant issues in humanitarian shelter were not restricted to construc-
provision, written by shelter prac- Given the scale of emergency
shelter need every year, case tion of an agreed shelter design.
titioners with specific interests and As a result readers will find projects
experiences. studies must have had large
scale impacts. Discontinued which include issues such as rental
A full list of case studies trials or design concepts were support, (e.g. A.10 and A.11 Haiti –
contained in all four editions: not included. 2010), settlement issues (e.g. A.31
Shelter Projects 2008, Shelter The majority of the project – Tunisia – 2011), site planning (e.g.
Projects 2009, Shelter Projects 2010 must be implemented within A.15 – Kenya – 2011) and coordi-
and Shelter Projects 2011–2012 the first years following a nation (e.g. A.20 – Pakistan – 2010
can be found in Annex 1. natural disaster. For conflict and A.28 – Somalia).
The case studies in this book affected populations, chronic In the case studies, we include
were implemented by many emergencies and return some findings from a 10 year evalu-
different organisations, a full list of processes, longer time scales ation of a transitional shelter project
which can be found in the acknowl- can be considered. (A.7 – Democratic Republic of
edgements section (page iv). In Accurate project information is Congo – 2002). We also include a
order to allow weaknesses as well available from staff involved in case study from 1871 that illustrates
as strengths of programmes to be the project implementation. the long history of shelter projects (
openly shared, the case studies are The case studies should illustrate A.32 – USA (Chicago) – 1871),
not directly attributed to individual a diversity of approaches to and contains an early design for a
organisations. Host government t-shelter / core house.
projects are not included.
Including a case study in this

book does not necessarily
mean that it represents best
practice...
As a result of the projects being

implemented in diverse and often
challenging conditions, they illus-
trate both good and bad practices.
From every case study there are
lessons that can be learnt, and
aspects that may be repeated or
need to be avoided.
Warning
Each project must take into con-
sideration the local contexts and
needs of the affected population,
which will differ from case to case.
Projects should therefore not be
directly copied or there will inevi-
tably be programmatic weaknesses
and failures. This edition of Shelter Projects focuses on a broader range of projects than
previous editions - such as A.15, Kenya (Dadaab), which includes site planning.
Photo: Joseph Ashmore
www.ShelterCaseStudies.org vii
Introduction Introduction Introduction
Global shelter need Natural disasters Countries have differing

The data presented in the 2011/2012 capacities to cope with the
table below indicates that over In 2011 there were 336 affects of such disasters. For
seventy million people were recorded natural disasters affecting example millions of people in
displaced or remained displaced approximately 209 million people1. China are displaced every year
as a result of conflict, natural Although this gives an idea of the by natural disasters, but little
disasters and economic develop- scale of disaster impacts it cannot humanitarian aid is requested.
ment in 2011/2012. However, how be directly linked to shelter needs.
Conflicts in 2011/2012
these people settled and the total However, analysis of the data does
It is estimated that 60 per cent
number of people who required give an idea of where the greatest
of all forced migrants are displaced
shelter support is not known. needs may lie.
by conflict and violence. All of them
Although most of these seventy The overwhelming majority of required new shelter in their dis-
million people were displaced people affected by natural disasters placement locations. There were
before 2011, all have required new live in Asia and in countries with additional shelter and land needs in
shelter solutions at some stage. medium or low Human Develop- locations of eventual return.
Many found their own solutions, ment Index scores.
Countries with conflicts
whilst many more were provided causing significant displacement in
The data available for 2012
with external assistance. 2011/2012 included:
reflects the pattern that floods,
droughts and storms affect the
Syrian Arab Republic
greatest number of people. Major
Total number of refugees and IDPs by Central African Republic
floods in 2012 in China, Pakistan,
category (in millions). Democratic Republic of Congo
the Philippines and India dominate
2010 2011 Sudan / Rupublic of South Sudan
the statistics for numbers of people
UNHCR refugees 10.55 10.40 affected by natural disasters. As in previous years, the total
Droughts in 2012 are estimated refugee numbers remain fairly
Asylum seekers 0.85 0.87
to have affected 11 million people static. More than half of the world’s
Palestinians (care of 5.00 5.10 refugees came from three countries
UNRWA) in the Horn of Africa and 3 million
people in North Korea. in 2011: Iraq, Somalia and Afghani-
REFUGEES TOTAL 16.40 16.37 stan.
IDPs (conflict and Other disasters also had signifi-
generalized violence- 27.50 26.40 cant impacts. The 2011 earthquake Around three-quarters of the
induced) and tsunami in Japan caused sig- refugee population remain in a
Natural-hazard nificant loss of life and destroyed situation of “protracted displace-
42.30 14.90
disaster-induced 128,000 houses. It was also the ment” with the international
IDPs TOTAL 69.80 41.30 most expensive disaster in history. community unable to produce
durable solutions as a result of
Development induced The limitations of these figures
(e.g displaced by 15.00 15.00 ongoing disagreements over land
dams) in terms of assessing shelter needs rights and political instability.
TOTAL 101.20 72.67 is limited due to the following
factors: It is estimated that half a million
Source: Table 1.1 IFRC World Disaster refugees voluntarily repatriated in
Report 2012, p. 15 Accurate numbers of people 2011 and nearly 2.5 million internal-
displaced are not always ly displaced people returned home.
available. This is an improvement on 2010,
which had one of the lowest return
Total number of people reported affected by natural disasters
rates in 20 years. Projects from Af-
(in thousands.) ghanistan (A.1), Cote d’Ivoire (A.5-
A.7), and Sudan (A.29) in this book
Asia Africa Americas Europe Oceania relate to return and resettlement
2007 190,885 1,253 9,119 1,651 172 programmes.
2008 182,754 22,653 20,314 268 105 The deterioration of the security
2009 174,056 42,636 7,046 141 77 situation in Syria led to over half a
2010 292,534 3,724 12,744 834 549 million people seeking protection
in Turkey, Jordan and Lebanon by
2011 176,453 19,092 13,457 79 484
the end of 2012. A further two
Total 1,016,682 89,358 62,680 2,973 1,387 million people were thought to be
Adapted from Annex (Table 3) IFRC, World Disasters Report 2012, p. 258 displaced within Syria, numbers that
continued to rise.
1
Figures for disaster-affected populations in 2012 are incomplete. The main statistical resource quoted here for disaster infor-
mation (EM-DAT international disaster database) is under constant revision. Final figures for 2012 are not yet available.
viii
Where to find different types of response in the case studies
Settlement option
Short-term land, house
Urban neighbourhoods
Planned and managed
or apartment tenant
camps / relocation
Collective centres
Unplanned camps
Non-displaced /
Dispersed self-
Resettlement
settlement
Case study
Hosting
returns
sites
A.1 – Afghanistan – 2012 x x
A.2 – Burkina Faso – 2012 x
A.3 – Colombia – 2012 x
A.5 – Cote d’Ivoire – 2012 x x
A.6 – Cote d’Ivoire – 2012 x x
A.7 – DRC – 2002 x x
A.8 – Ethiopia – 2011 x
A.9 – Ethiopia – 2012 x
A.10 – Haiti – 2010 x x x x
A.11 – Haiti – 2010 x x x x x
A.12 – Haiti – 2010 x x x
A.13 – Haiti – 2010 x
A.14 – Japan – 2011 x x x
A.15 – Kenya – 2011 x
A.16 – Lebanon – 2007 x x x
A.18 – Madagascar – 2012 x
A.19 – Nicaragua – 2007 x
A.20 – Pakistan – 2010 x x x x x
A.21 – Pakistan – 2010 x
A.24 – Peru – 2012 x
A.26 – Philippines – 2012 x x x
A.27 – Philippines – 2012 x
A.28 – Somalia – 2012 x
A.29 – Sudan – 2012 x x
A.30 – Thailand – 2011 x x x x
A.31 – Tunisia – 2011 x x
A.32 – USA – 1871 x x x
www.ShelterCaseStudies.org ix
Table illustrating which forms of support were provided in each case study
Guidelines / materials /mass

Construction materials
Structural assessment
Transitional shelter /
progressive shelter
Housing repair and
Emergency shelter
Household items
Advocacy / legal
communications
Cash / vouchers
Core housing /
Rental support
Infrastructure
Site planning
host families
Support for
retrofitting
T-shelter
Training
Loans*
Tools
Case study
A.1 – Afghanistan – 2012 x x x x x x

A.2 – Burkina Faso – 2012 x x x x x
A.3 – Colombia – 2012 x x x x
A.5 – Cote d’Ivoire – 2012 x x x x x x
A.6 – Cote d’Ivoire – 2012 x x x x
A.7 – DRC – 2002 x x x x
A.8 – Ethiopia – 2011 x x x x x x x
A.9 – Ethiopia – 2012 x x
A.10 – Haiti – 2010 x
A.11 – Haiti – 2010 x x x x
A.12 – Haiti – 2010 x x x x x
A.13 – Haiti – 2010 x x x x
A.14 – Japan – 2011 x x x
A.15 – Kenya – 2011 x x x x x
A.17 – Lebanon – 2011 x x x x x
A.18 – Madagascar – 2012 x x x x
A.19 – Nicaragua – 2007 x x x x
A.20 – Pakistan – 2010 x x x
A.22 – Pakistan – 2011 x x x x
A.24 – Peru – 2012 x x x x
A.26 – Philippines – 2012 x x x
A.27 – Philippines – 2012 x x x x x x x
A.28 – Somalia – 2012 x x x x x
A.29 – Sudan – 2012 x x x x x
A.30 – Thailand – 2011 x
A.31 – Tunisia – 2011 x x
A.32 – USA – 1871 x x x x x x
* Although there are no examples in this edition of Shelter Explanation of columns:

Projects, "Loans" is included as a category of assistance in Household items - tents / blankets and other non-food items
Construction materials - were provided for construction/repair.
this table as there were examples in previous editions. (e.g.
Emergency shelter / transitional shelter, T-shelter, temporary
A.29 Tajikistan, 2010 Shelter Projects 2010)
shelter, semi-permanent shelter, core housing / progressive
shelter. Terminology is used according to the wording used in the
response.
x
Recurring themes Scale using the cluster approach to

Affected people are the first re- Disasters and displacements vary coordination. This was proposed as a
sponders massively in scale, and as a result so way of addressing gaps and ensuring
The first and main response in do responses. Many are also dealt responses were more effective
most of the case studies in this book with in country (see A.30 – Thailand (see humanitarianresponse.info
is by the affected people them- – 2011) and sheltercluster.org for more
selves. Of the case studies in this background).
In many responses there is simply
book, the more effective projects not sufficient funding or capacity In this book, we include two
were set up with assessments that for organisations to provide the projects as examples of cluster
led to a clear understanding of the support that is required. coordination, whilst many others
needs, and with two way commu- highlight coordination components
nication between the implement- In the light of resource con-
of the project. In A.20 – Pakistan
ing organisations and the affected straints, organisations often have to
– 2010, we include a case study of
people. make tough decisions as to whether
coordination in a very large scale
to provide a high level of longer
The fact that disaster and response and the need to ensure
term support for a limited number
conflict affected people are usually that coordination takes place at
of households, or a lower level of
highly proactive in finding solutions the village level as well as at the
support for a larger number of
to their own shelter needs is recog- national level. In A.28 – Somalia –
households. In the case of displace-
nized in many of the case studies in 2011, we look at some of the issues
ments over borders due to conflict,
this book. For example in A.4 – Cote in ensuring that multi-sectoral
there is often little choice and some
d’Ivoire, organisations assessed responses are coordinated in the
kind of support must be provided
the “self recovery rate”, and made complex urban environment of
to all displaced people (see A.15 –
follow-on planning assumptions for Mogadishu.
Kenya (Dadaab) – 2011).
the support that was needed. Disaster Risk Reduction (DRR)
Many project are set up to work
Sphere standards1 and indicators In many responses, particu-
at a small scale with the hope that
provide common standards on par- larly to natural disasters, there is a
the project can allow the organisa-
ticipation, initial assessment, moni- need to support people to “build
tions to provide a larger scale of
toring and evaluation. back safer” and enhance the resil-
support through advocacy and by
ience of people to withstand future
Types of response providing a replicable model (see
disasters. In this edition of Shelter
The previous two tables highlight A.3 – Colombia – 2012). Often
Projects we have tried to highlight
some themes that recur between by implementing even small scale
the DRR components of projects.
case studies and in which case study projects, organisations engage
This ranges from structural and
they can be found. The first of the in the practical realities of shelter
engineering support (A.1 – Af-
two tables identifies the kinds of and reconstruction and can be in a
ghanistan – 2012) to projects firmly
settlement options supported by much stronger position to advocate
based in community-based disaster
the project and the second provides from.
risk reduction principles (A.21 and
more detail on the type of assis- In some circumstances the A.23 – Pakistan – 2011).
tance that was provided. scale changes rapidly and the
Settlements, land and planning
A quick glance at these tables programmes must adapt to the
In this edition of shelter projects
shows that: changing scale of needs (A.17 –
we have tried to look at shelter in
Lebanon – 2011)
Camps are not the only the broader sense and to include
settlement option supported. Selecting an area to intervene in issues relating to settlements. In the
There are a diversity of types Selecting the area of interven- “Opinion pieces” (Section B), we
of shelter that can be built by tion (province/district/village) is include a piece on this subject (B.4
affected people or by supporting the first step in selecting who a – Reflection on the Importance of
organisations. project will support. It has very far Settlements in Humanitarian Shelter
There is a diversity of ways of reaching implications as to whether Assistance).
supporting people to improve a project will meet the needs of the
Some case studies include a
shelter. These range from most vulnerable people. In many
discussion of some of the issues
direct support in construction, projects this decision is taken rela-
encountered surrounding land in
to offering legal support, to tively rapidly and it is made using
recovery operations (A.26 – Philip-
improving communication with less detailed criteria than are used
pines – 2012) whilst others look at
disaster affected people so that for selecting individuals.
the issues surrounding site planning
they can make more informed Coordination and clusters the context of conflict induced dis-
choices. Since 2005, many of the larger placement. (A.15 – Kenya (Dadaab)
responses have been managed – 2011 and A.31 – Tunisia – 2011).
1
Sphere Project, Sphere: Humanitarian charter and minimum standards in humanitarian response, 2011
www.ShelterCaseStudies.org xi
Minimum and maximum materials and labour costs per household for each project in this book
(Note that these figures are not necessarily representative of the entire responses)
6000
Minimum materials and labour cost

Inflation adjusted cost (US$)
5000 Beneficiary contribution

Maximum materials and labour cost
Beneficiary contribution
4000
3000
2000
1000
A.18 – Madagascar – 2012

A.5 – Cote d’Ivoire – 2012
A.6 – Cote d’Ivoire – 2012
A.2 – Burkina Faso – 2012

A.1 – Afghanistan – 2012
A.27 – Philippines – 2012
A.26 – Philippines – 2012

A.19 – Nicaragua – 2007
A.17 – Lebanon – 2011*
A.7 – DRC Goma – 2002

A.16 – Lebanon – 2007
A.3 – Colombia – 2012
A.20 – Pakistan – 2010


A.28 – Somalia – 2012
A.8 – Ethiopia – 2011
A.9 – Ethiopia – 2012

A.29 – Sudan – 2012
A.14 – Japan – 2011
A.33 – USA – 1871

A.13 – Haiti – 2010
A.11 – Haiti – 2010

A.12 – Haiti – 2010
A.24 – Peru – 2012

Case study and chapter number
Costing shelter projects Terminology Interpret and contribute

The stated costs of the various There has been a lot of academic In reading this book, or
projects in this book are shown in and practical debate surrounding browsing relevant case studies, it is
the graphic above. However the terminology used in shelter. Addi- hoped that readers will be able to
costs do not all measure the same tional confusions have been added draw their own lessons and identify
things and should not be used as by language translation issues. useful techniques and approaches.
indicators of value for money or of
Issues of the definition of words Readers are encouraged to send
project success.
have been particularly great sur- in their own projects for future
Each project was conducted in rounding the language used for editions. In this way, the humanitar-
very different circumstances with different phases of assistance. As ian community can compile good
very different markets, local con- an example the terms “transitional and bad practices and hopefully
struction methodologies, materials, shelter”2, "T-shelter", "temporary implement increasingly effective
skills availability and logistics con- shelter", "semi-permanent shelter" shelter projects in the future.
straints. Projects also varied greatly and "incremental shelter" have all
in the type of assistance provided, Contribute at:
been used in responses to define
from provision of materials, both the types of shelters and the www.ShelterCaseStudies.org
to projects with much higher processes used.
advocacy, training or mobilisation
In this book we use the
components.
terminology that was used in country
In reporting the overall project for each response. Although there
costs different organisations have can be some confusions, practical
used different approaches making response specific understandings
direct comparison difficult. Some are usually developed surrounding
have divided the entire project the use of these terms. In some
budget by the number of shelters cases, flexibility in terminology
built, whilst other projects have has helped projects to take place
multiple sources of funding or sooner.
work in multiple sectors, making
overall shelter project costs harder
to calculate.
xii
“[the sites were] ... filled by little towns of tents and huts; so
also the Estrela district in the west and the Campo de Santa
Clara on the east side of the town were full of squatters... It
was estimated that about nine thousand wooden buildings
were put up during the first six months after the quake, a
fine achievement, for wood was very scarce indeed in Lisbon
and much of it had to be brought to the city for this special
purpose.
The general desire was to get out of buildings into tents or

huts, and to sleep in the garden rather than indoors, even if
one’s home still stood safe and sound, and for this reason
the great camps on the high and open places round the city
were for a long time crowded communities, in spite of the
initial discomfort and squalor of the miserable bivouacs of
matting, planks, and sail-cloth under which many of the
squatters spent their first few nights.
The most remarkable concourse of these campers was that in

and around the quinta of the Oratory in Cotovia... in a little
while an ordered settlement of wooden huts was established,
and some of these, built for nobles and high officials, began
to be quite luxurious bungalows with glass windows and
tapestry hangings and good domestic offices. ”
Shelter in response to the Lisbon earthquake of 1755.

Source: T. D. Kendrick the Lisbon Earthquake,1956
1755 German copperplate image, The Ruins of Lisbon:

In the left of illustration is a tented camp in the suburbs of Lisbon following the fire of
1755. On the right damage that is probably related to the 1531 earthquake.
Source: Wikimedia commons
www.ShelterCaseStudies.org xiii
xiv
Case Studies Shelter Projects 2010 Section A
SECTION A
Case Studies
This section contains case studies of projects from both conflicts and natural disasters. It also contains
one update from a project (A.7) that was included in Shelter Projects 2008.
See “Annex 1 - Index - by country” for case studies that are in previous editions.
A.1 Afghanistan – 2012 – Conflict Returns A.17 Lebanon – 2011 – Conflict

A.2 Burkina Faso – 2012 – Conflict A.18 Madagascar – 2012 – Tropical Storm
A.3 Colombia – 2010 – Floods A.19 Nicaragua – 2007 – Hurricane
A.4 Côte d’Ivoire – 2010–2011 – Post-electoral Crisis A.20 Pakistan – 2010 – Floods
A.5 Côte d’Ivoire – 2010–2011 – Post-Electoral Crisis A.21 Pakistan – 2010 – Floods
A.6 Côte d’Ivoire – 2010–2011 – Post-Electoral Crisis A.22 Pakistan – 2011 – Floods
A.7 Democratic Republic of Congo – 2002 – Volcano A.23 Pakistan – 2011 – Floods
A.8 Ethiopia – 2011 – Sudanese Conflict A.24 Peru – 2012 – Flooding and Land Slides
A.9 Ethiopia – 2012 – Conflict and Drought A.25 Philippines – 2011 – Cyclone
A.10 Haiti – 2010 – Earthquake A.26 Philippines – 2012 – Cyclone
A.11 Haiti – 2010 – Earthquake A.27 Philippines – 2012 – Cyclone
A.12 Haiti – 2010 – Earthquake A.28 Somalia – 2011 – Famine / Conflict
A.13 Haiti – 2010 – Earthquake A.29 Republic of South Sudan – 2011 – Conflict
A.14 Japan – 2011 – Earthquake and Tsunami A.30 Thailand – 2011 – Bangkok Floods
A.15 Kenya (Dadaab) 2011 – Famine / Conflict A.31 Tunisia – 2011 – Conflict in Libya
A.16 Lebanon – 2007 – Conflict A.32 USA (Chicago) – 1871 – Fire
www.ShelterCaseStudies.org 1
Section A Case Studies
2
Conflict / Complex Shelter Projects 2011–2012 A.1
A.1 Afghanistan – 2012 – Conflict Returns

Case Study: Keywords: Returns, Urban neighbourhoods, Construction materials, Core housing construction,
Cash / vouchers, Infrastructure, Training.
Country: Project timeline

Afghanistan
12 months - – Project completion
Project location:
Kabul, Herat and Jalalabad
Disaster:
Conflict returns 12 months – – Monitoring and
quality check
Date:
2002 onwards
Number of houses damaged:
More than 130,000 houses in
project areas (within Kabul)
Number of people returned: 8 months – – Hand over and
Over 5 million people since 2002 assessment of oc-
cupancy rate
150,000 families in Kabul
Project target population: 5 months – – Hygiene promotion
Pilot 295 households assessments
(Expanded to 2,075 households)
Project outputs:
295 shelters with hygiene activities 4 months – – Shelter construction
Shelter size: Herat Kabul
One-room shelter: average 18m2 Jalalabad
Two-room shelter: average 30m2 4 months – – Beneficiary selec-
Afghanistan
Materials cost per household: tion and community
Two-room shelter US$ 1,700 mobilisation
(household contributes US$ 500) 1 month – – Project planning
One room shelter US$ 800
(household contributes US$ 200) 1st January
Project cost per household: 2012 – – Project start
Two-room shelter, including
indirect cost US$ 2,286
Project description
This project addressed the poor living conditions of recent refugee-returnees, IDPs and host families through
the construction of 295 semi-permanent shelters with household latrines and hygiene promotion. Cash grants
gave beneficiaries an active role in the project and allowed the organisation’s staff to spend more time with the
community rather than managing contractors. The pilot phase of the project was successful and was scaled up
to target a further 2,075 households.
Strengths and weaknesses 8 It was challenging to identify the most vulnerable

9 The beneficiaries took control of the construction families. The urban context made this more difficult.
process, and adapted the design of the shelters 8 The project did not address wider community
according to their own needs. planning issues of community sanitation and drainage,
9 Groups of five beneficiary households worked or community-level disaster risk reduction (DRR).
together to manage the construction process, 8 It was not anticipated that some construction
promoting community cohesion. techniques, which returnees had brought back, were
9 Freed from construction management tasks, field not earthquake-resistant, leading to weaker buildings.
teams focused on discussing specific DRR measures - There is ongoing discussion about whether smaller,
with each household. single-room core shelters provide enough space.
9 The cash-grant project resulted in three times - Allowing households control over design required
the number of shelters being built compared to the greater technical support from the organisation.
previous year's direct-procurement method. - Separating chronic needs from returnees needs in
8 A gender balance amongst beneficiaries was not urban Kabul was challenging.
acheived, despite using a vulnerability list.
A.1 Conflict / Complex
Groups of five households were provided phased cash payments to allow them to build a one- or two-room shelters with
sanitation facilities. People were given flexibility to build what they needed
Photo: Left: Jim Kennedy, Right: Jake Zarins
Before the conflict build shelter, but many households ciary selection committees. Land
In the 1970s the population of lacked the labour and materials. ownership documents were
Afghanistan’s capital, Kabul, was checked by the Maliks, who were
By the end of 2011, more than
500,000. Despite the fact that a able to resolve local and non-writ-
200,000 shelters had been provided
range of different ethnic groups ten issues surrounding tenure.
for returning refugees and Internal-
lived together in relative peace, ly displaced people (IDPs) by various The pilot phase targeted 295
some groups were discriminated different organisations under one households, prioritising recent
against, with differing access to national programme. However, returnees from Pakistan and Iran.
resources, property and services. As there remained a national gap of These were followed by IDPs,
a result, the Hazara minority were 50,000 shelters. and finally, host communities.
living on the outskirts of the city In addition to these main target
whilst Pushtoon and Tajik groups The government set a target of
groups, the organisation prioritized
occupied more central areas. the end of 2010 for the complete
according to primarily landless, and
rehabilitation and integration of all
The 1978 revolution was then land-owning returnees who
displaced people. Two years later
followed by civil war and Soviet had been displaced or returned
housing and landlessness remained
invasion. This led to a significant since 2008.
significant obstacles.
growth in Kabul’s population The final section was based on
as many people were displaced The lack of available shelter
the following criteria:
from rural to urban areas. The or land in Afghanistan is the
city’s Hazara population increased primary reason for many refugees female headed households
tenfold , establishing new settle- remaining in Pakistan and Iran. The disabled headed households
ments in the western part of the Ministry of Refugees and Repatria- child headed housholds
city. tion (MoRR) launched a land alloca- elderly headed of households
tion scheme at the end of 2005 to victims of Gender Based
The collapse of Afghanistan’s deal with this issue. The scheme has Violence (GBV)
communist regime in 1992 led to so far provided 42,000 families with large families
an intensification of conflict, killing temporary land ownership deeds. very low income families with
tens of thousands of people in just no regular income.
four years. During this period many Selection of beneficiaries
city residents (mainly Hazaras) had The most vulnerable families
In 2011, districts 13 and 16 in
fled to Pakistan, Iran and other were given additional financial and
the western part of Kabul were
parts of Afghanistan. technical assistance.
identified as the neediest areas of
During this period all the houses urban Kabul for shelter assistance.
in western Kabul were destroyed. The organisation worked with “The Community Driven
beneficiary selection committees Method (CDP) allowed me to
Conflict returnees established in each community. purchase the material for my
Since 2002 and the fall of the Each committee consisted of two
Taliban regime, over five million shelter according to my own
staff from the organisation (one choice. The design of my
people have returned to Afghani- male and one female), a represent-
stan. shelter unit was finalised in
ative from the government, and the
Most of the returnees found ‘Gozar’s Malik’ (religious leader). consultation with my family
that their own houses had been members.”
Beneficiary selection forms
totally destroyed and rented shelter and the guidelines and criteria
or stayed with host families. Many for filling them in were explained
had land that they could use to Abdullah - Shelter beneficiary
during workshops with benefi-
4
Implementation had urgent sheltering needs and

The organisation had been who could not afford rent, did not
building shelters in Afghanistan for wait until the handover to move
a number of years, but had previ- into their new shelters.
ously directly provided construction
materials. This was the first time Technical issues
that a cash-based, owner-driven Key to the success of the project,
approach had been attempted by the returnee population had the
the organisation in Afghanistan. skills to build their own houses.
The pilot project was implemented Some people had learnt these in the
in 2011. construction industry in Iran.
After signing a memorandum of Instead of providing fixed Households were expected to

understanding with the provincial designs, the project provided a make contributions of money and
generic bill of quantity and technical labour to the construction of their
authorities, the selection of benefi- shelter.
ciaries began. advice to individual households to
Photo: Jim Kennedy
address disaster risks. This included
The project established benefi- advice on proper jointing for stone
ciary groups of 4 to 5 members to unaware of the greater seismic risks
and brick masonry, the proper
create community networks that that such materials carry. Encourag-
placement of lintels and roofing
would support vulnerable ben- ing families to use more traditional
beams, and proper roof drainage.
eficiaries (especially women and materials and methods was chal-
disabled people). The whole group The training provided by the lenging.
would not receive their grant instal- field teams was accompanied by il-
lustrated construction drawings. Logistics
ments if one of the group members In previous projects the countr
had not reached the agreed stage The decision to give homeown- programme had directly managed
of construction. This condition ers flexibility in what they could procurement and logistics, and
forced the group members to help build was based on learnings from this had led to many challenges.
each other and work together. previous projects where a single, In contrast, in the community-
Grants were paid out in hard standard shelter design was issued. driven approach, only timber for
currency (cash in envelopes) in Plot sizes in Kabul vary greatly and bracing, tool kits and hygiene kits
four instalments. The cash was flexible design allowed households were procured by the project and
to be used for purchasing shelter to adapt constructions to the space delivered to the beneficiaries.
materials. Mobile phone banking available.
The rest of the materials such
options were investigated but as lintels, roofing materials, doors,
rejected as being too complicated. Some families piled windows and latrine slabs were
In the original pilot project in sandbags around procured by the households them-
Kabul, 102 out of 295 families the foundations as a selves. Households made a personal
opted to construct a single room preventative measure to contribution of around one third
shelter. This was mainly because prevent erosion in case of of the costs of construction and
the shelter plots were not large materials.
flooding.
enough for the two-room shelters.
As the shelters were constructed
Project follow up
DRR components The pilot project team was made
by the affected households, the di-
Since Kabul has earthquake up of six people with mixed skills,
mensions of each shelter varied.
risks, timber braces were provided including engineers, community
Technical assistance to all households to be used at each mobilisers and people with data-col-
During the shelter construction, corner of each shelter. Families lection experience. In previous years
households received support from were also provided with technical this team had built 100 shelters per
the project technical staff. This training on disaster risk reduction. year using the direct procurement
included advice on the plot layout, method. The cash-based approach
ground clearing and foundation As the cash-based approach nearly tripled this figure.
digging, stone masonry, brick allowed team members to spend
more time with households, they The success of the pilot project
masonry, seismic risk reduction led to the implementation of the
measures and roofing design. were able to better explain seismic
mitigation measures compared to cash-based approach in other parts
Handover previous projects. of Afghanistan and by the end of
The houses were handed over 2012, 60 per cent of the planned
Many people were interested 2,075 shelters had been built in
to the households when they were in more modern materials and
completed and well dried. However Kabul, Jalalabad and Herat.
construction methods but were
some of the neediest people, who
A.2 Conflict
A.2 Burkina Faso – 2012 – Conflict

Keywords: Planned and managed camps, Construction materials, Emergency shelter, Transi-
Case Study: tional shelter / T-shelter, vouchers, Site planning.

Burkina Faso
Project location:
Férério Refugee Camp, Oudalan 31 August
Province 2012– – Project completion
Conflict:
Malian Refugee Crisis
Conflict date:
March 2012
Number of people displaced:
July 2012 (increased later in
2012): 18 June 2012 – – Distribution of shel-
ter materials started
IDPs in Mali: 150,000
Refugees - Burkina Faso: 100,000
Project target population:
Férério Refugee Camp: 30 June 2012 – – Shelter sample
constructed with
3,000 households May 2012 refugee community
4,000 households August 2012 Mali
Project outputs: 27 June 2012 – – Beneficiary assess-
1,000 shelters ment and selection
commenced
Occupancy rate on handover:
Férério
100 per cent 12 June 2012 – – Project start
camp
Shelter size:
21 m2 Burkina Faso
Materials cost per shelter:
US$ 240
Project cost per shelter:
Unknown March 2012 – – Conflict start
Project description
This project provided temporary shelters for nomadic Tuareg refugees displaced from northern Mali to the
Oudalan Province in Burkina Faso. Shelters were built through a self-help construction approach using traditional
construction materials. Participation in the selection of the type of shelter to be provided was crucial since the
refugees had already rejected other proposed solutions by other agencies. The project worked within the cultural
norms of a Tuareg population where women were the main constructors of tents, and families moved their
shelters according to nomadic traditions to increase spacing between shelters and tribal groups.
Strengths and weaknesses sanitation services did not change their layout, leading
9 Beneficiaries felt involved in the process right from to many latrines being either too far away or too close
the first discussion around shelter design. to other groups.
9 The project was implemented through existing 8 Coordination was hampered by the lack of a camp
community structures which facilitated beneficiary management focal point.
selection as well as shelter materials distribution. 8 Site selection, though beyond the influence of
9 Close involvement of the beneficiary communities this project, made accessing populations difficult as
guaranteed the security of both project staff and stored communication connections were poor.
materials. - Site planning at Férério camp had to adapt to
9 The host government representatives on site were the cultural norms and social structures of the camp
part of the coordination process. population. A traditional grid layout was inappropriate
9 Staging the distribution of materials worked as an and was rejected by the refugees who preferred to
incentive to complete the shelters. group their shelters according to tribal affiliations and
8 Coordination with some other agencies could have space them in a way that reflected their usual, nomadic
been strengthened. Despite the change in site layout in way of living.
terms of spacing between the shelters the providers of
6
Conflict Shelter Projects 2011–2012 A.2
Shelter structure under construction. Completed shelter structure. Traditional tanned skin roof cover.
Photo: Christian Jepsen Photo: Ghada Ajami Photo: Christian Jepsen
Before the conflict Beneficiary selection Implementation

The Tuareg population in The organisation received an Participation in all stages of the
northern Mali is made up of assessment report from another project was crucial so a sample
nomadic and semi-nomadic groups shelter actor that concluded that shelter was built following discus-
moving across sparsely populated 1,000 shelters were needed. These sions with community groups about
desert areas. Traditional Tuareg tent shelters were to fill gaps in support the design. The sample shelter was
shelters are made from wooden as some shelter solutions had then a focal point for suggested
supports covered with tanned ani- already been provided. modifications before the final
mal-skin roofs, and are designed to materials list was established.
Initially an agency proposed a
be easily dismantled. The organisation procured the
shelter design based on a standard
The semi-nomadic population box-style shelter with a gable roof materials. Triple-weave plastic
construct mud brick houses with to be arranged in a grid format, sheeting was procured (though
traditional tents erected close by. fairly close together. Some shelters not produced) in Burkina Faso and
Although land is mainly owned by were built by an external contrac- the quality was seen as better than
men, the Tuareg tent is built by tor. plastic sheeting that had previously
women and is the property of the been distributed in the camp.
Both the design and the site plan
family matriarch. To prevent damage to the local
were rejected by the refugees and
Mali is one of the poorest this agency was invited to provide environment by cutting down trees,
countries in the world, with a life an alternative solution, securing wooden poles were procured from
expectancy of just over 50 years extra funds to meet any additional sustainable Eucalyptus plantations
and a Human Development Index needs. in the Southern regions of Burkina
ranked 175 (out of 187). Faso.
Three criteria for selection were
shared with the beneficiary groups, All materials were first trans-
After the conflict ported by truck to a hub three-and-
A large number of the Tuareg the United Nations and the govern-
ment representative in the camp: a-half hours’ drive from the camp
population of northern Mali sought
and then to the camp itself.
safety in neighbouring countries. households with a lack of
The Tuareg population targeted adequate shelter The beneficiary communi-
by this project moved mainly to households with elderly ties were given responsibility for
the Sahel region of Burkina Faso occupants guarding the wood stored in an
but were asked by the Burkinabe vulnerable female-headed open-air, fenced-off area, while
government to move to managed households who have no access desirable items like plastic sheeting
camp sites. to adequate shelter. and mats were distributed immedi-
ately to reduce the risk of theft.
By March 2012, Férério camp The community groups were
contained over 2,000 households. organised by the refugees them- Each household was given
Four months later, in July, the figure selves and were based on tradi- a materials coupon. Structural
had risen to risen to 3,500. tional tribal structures. Leaders of materials were distributed first and,
these groups drew up a draft list when the structure was completed,
Initially other organisations
of potential beneficiaries. This was materials for covering the roof and
provided all-weather tents, but
used as the basis for an assessment walls were distributed.
people refused to occupy them.
Emergency tents were seen as too by the agency in coordination with Distribution was coordinated
flimsy to protect people from strong camp community leaders and the with the tribe leader who organised
winds and high temperatures. host government representative. the order in which families would
The final beneficiary list was fetch their materials. The whole
This project was established to
presented to the group leaders, community of each tribe assisted
involve beneficiaries in the develop-
who communicated the outcome in moving the materials to the con-
ment of a shelter solution.
to the other families. struction site.
A.2 Conflict
An overall view of one section of the camp.

Photo: Christian Jepsen
Field monitors checked the Agencies working on water and

“I am very, very happy.
structures during and after con- sanitation continued place latrine
struction. blocks according to the site plan Look around, here is much
instead of adapting it to the settle- more space”, says Fatima
As Tuareg women had a tradi-
ment patterns of the refugees. As the proud new homeowner
tional leading role in tent construc-
a result, a high percentage of the surrounded by her children.
tion, it was they who led the con-
refugee population did not use the “There is even enough space
struction groups. Each group would
latrines either because of the long
complete one shelter at a time. for the little ones to play
distance (sometimes up to 500m)
The community specifically or because some tribes refused to inside, and I have room for
said that they did not require share latrines with other tribes. visitors.”
the agency’s support in construc-
tion and, in the case of vulner- Technical solutions
able families, men helped to dig The shelter model chosen was Tanned animal skins took too
the pole-holes while the women similar to a traditional Tuareg tent. long to produce, and were not an
groups erected the shelter. Due It had a wooden-pole structure but option as a roofing material. To
to the communal organisation of instead of the traditional tanned replicate the thermal insulation
childcare and of many other often skins for the roof and walls plastic qualities of the skins, a set of nine
female-dominated activities it is sheeting was used. In some cases woven straw mats were placed
not thought that the women were families used the emergency tents under the two plastic sheets.
over-burdened by their construc- that had been provided earlier as
tion responsibilities. roofing material. The refugees paid a lot of
attention to detail in construction.
As a side-project, the agency Tuareg tents are suited to the en- The two plastic sheets provided
contributed to the reduction vironmental conditions: high wind were hand sewn together while the
of work carried out by children loads, high temperatures and sand 8mm rope connecting the plastic
through the provision of donkey storms. The shelter contained no sheets to the roof was skilfully
carts for the collection of water. concrete so did not worsen water secured in place by tying it to the
scarcity. The sides of the shelter corner poles of the shelters.
Site planning and WASH were made from mats which could
Families did not like living in be re-positioned in order to change Materials Quantity
close proximity to each other and the location of the doors depending Stage I - structure
traditionally lived spread out. on the direction of the wind. Eucalyptus Poles 16
Green wood. Length = 4m
The camp held more than 25 The shelter could be disas- 6cm diameter at mid length
tribal groups. The camp popula- sembled and relocated to another Eucalyptus Poles 18
tion re-organised itself according to location without any material Green wood. Length = 4m
4cm diameter at mid length
these groups. A standard camp grid wastage, and women knew how to String 0.3cm diamter 2x20m
plan could not be applied to this maintain them. Materials could be Machete 1
more “organic” spread of families taken with families when the camp Stage 2 - coverings
and if the refugees did not like closed. Rope 0.8cm diameter 30m
where they were sited they simply Plastic mats (1.2m x 2.5m) 8
Plastic sheeting (4mx5m) 2
dismantled their shelter and moved Straw mat (1m x 1.8m) 9
it somewhere else.
8
Natural Disaster Shelter Projects 2011–2012 A.3
A.3 Colombia – 2010–2011 – Floods

Case Study: Keywords: Non-displaced, Housing repair and retrofitting, Advocacy, Infrastructure, Training.

Colombia
16 months – – Project completion
Project location:
Department of Chocó
Disaster:
Floods
13 months – – Elevation of school
Disaster date: and trainings
2010 to 2011
Number of houses damaged /
destroyed: 10 months – – Staff based in the
Over 350,000 community
5.463 people in 5 communities 7 months – – Procurement con-
80 households in target village tracts in place
Project outputs:
80 elevated houses 5 months – – Project revision
1.1km footbridge 4 months – – First Trainings
Disaster risk reduction activities
for 5527 people Chocó 3 months – – First pilot house
2 months – – Technical assess-
100 per cent ment
Shelter size: 1 month – – Community selec-
70m2 Colombia tion
Materials cost per shelter: June 2011– – Project start
US$ 3000 12 months– – Floods ongoing
Project cost per shelter
US$ 5300 (including staffing,
June 2010 – – Floods
volunteers, and logistics)
Project description
The project used community participation to improve the overall living conditions of 80 families who were
struggling to survive following flooding. It supported a total of 5,527 people in surrounding villages with disaster
risk reduction (DRR) activities. Stilt construction was used to build 80 new houses and a 2.5m high, 1.1km long
footbridge. Disaster preparedness activities, first aid, hygiene promotion and safe construction trainings were also
provided. The project is now an example, both at regional and national level, of what can be done to support
riverside communities to mitigate the effects of recurrent floods.
Strengths and weaknesses have not been allocated for large-scale replication.
9 The project demonstrates (both locally and globally) 8 The project did not have either communication or
that there is an alternative to resettling people affected advocacy strategies.
by floods and that living with floods is possible. 8 Local government was involved late in the project.
9 Long-term, positive impact on the community's 8 Water and sanitation components of the project
resilience, disaster preparedness and social cohesion. were not resolved.
9 As logistics costs were high, a greater impact was - The government had limited capacity to provide
achieved by concentrating on a few communities. technical and financial support.
9 The disaster risk reduction (DRR) project included - High logistic costs demanded capacity from outside
housing improvements, infrastructure reconstruction, the village, staff from the organisation and local alliances.
food security, environmental education, hygiene - Risk management and DRR at local level is still solely
promotion, livelihoods and training on how to elevate focused on emergency response.
buildings. - Project timelines imposed by donors were very
9 The model is easily replicated for other flood-prone tight. The project needed to balance the timeframes
communities. and flexibility required for local construction practices,
8 The project was relatively small-scale and resources livelihoods and genuine participation against pressure to
complete the project.
A.3 Natural Disaster
The main carpenter and his assis-

tants received US$ 340 for each
completed house.
On-the-job training was
provided to carpenters to ensure
long-term knowledge transfer of
techniques such as wooden pole
treatment and replacement and the
principles of elevated construction.
Initially, damaged tools were
The project had a strong focus on training and disaster preparedness, replaced by the project. Later it
sometimes using simulations. was decided that each carpenter or
Photo: Sandra D’Urzo/IFRC woodcutter would pay for his tools
Before the floods Some people considered reset- and keep them at the end of the
Chocó is a department in north- tling closer to the main town but project.
western Colombia, on the Pacific the community was attached to the The project began by elevating
coast and is famed for its jungle and collectively owned land. A national an existing house and school
biodiversity. decree protects this ethnic group building. However, a technical
and other indigenous populations. review stated that new construc-
As most of Chocó is inaccessible
by road, rivers are traditionally the tion, although far more expensive,
Implementation would be more effective than
major transport routes. The project was implemented
elevating existing buildings.
The community of San José de with a focus on participation. Over
la Calle was displaced by conflict in the course of one year, the entire A pilot house, elevated by 2.5m,
the region in the early 1990s. Since community contributed to create was then built to demonstrate the
then, livelihoods have been based a village which serves as a model building technique. Families would
on timber exploitation and seasonal for other projects. The community need time to adjust to the new
fishing. The remote location council was the main decision-mak- design, especially in dry season,
hampers development of alterna- ing entity. but were keen to live “on the first
tive livelihoods and job creation, floor” in order to escape the effects
Lumberjacks from the village
while municipal services such as of flooding. A total of eighty new
worked together to cut timber and
electricity and water are scarce or houses was built.
decided its price. Women cooked
non-existant. collectively during the construction, The new footbridge design was
In 2002, there was a massacre and children helped to carry smaller based on a 3km long bridge built
in the nearby town of Bellavista. materials for the footbridge. in another community. The bridge
Since then, international aid organi- had shown to have a positive effect
Continuous dialogue with the
sations distributed relief and made on psycho-social wellbeing, as
main community representatives
water and sanitation improvements villagers could stay connected with
(the council, women’s groups,
in the area. San José de la Calle one another during the months of
craftsmen and the lumberjacks
benefited from a latrine-building flooding.
union) facilitated collective deci-
project, but unfortunately these sion-making. This was achieved A school, an elevated collective
were only usable in the dry season. during the donor’s timeline of 15 garden, a community centre and an
Until recent years, floods lasted months (one year of construction elevated children’s playground were
about one month, isolating house- activities). also built.
holds, and interrupting schools and At first, craftsmen were not There was no water and sani-
livelihoods. Families built mezzanine paid for the construction of their tation component to the project.
levels inside their homes to keep own houses, and only technical as- Existing, partially-damaged latrines
them and their possessions dry. sistance was provided. Later, food were dismantled.
for work and cash for work were
After the floods Selection of beneficiaries
provided to accelerate construc-
The 2010 floods lasted The entire community benefited
tion, though families still needed to
six months, during which the from the risk reduction aspects of
continue with existing livelihoods
community lost most of its this project. In the selected village
activities.
economic resources. The severity of all houses were reconstructed.
the flooding is expected to continue Skilled carpenters were hired
in future years primarily as a result from outside the community. Con- Coordination
of over-exploitation of the forests struction was managed in teams of The project was coordinated
leading to silt deposits in the Atrato three people who were paid daily. with government departments and
river. institutions. The government was
10
In one “model“ village, the organisation built eighty elevated houses and community infrastructure.
Photo: Sandra D’Urzo/IFRC
willing to provide extra funds to Several videos were produced The bridge was constructed
complete the newly-built houses during the project to showcase from a wooden frame with recycled
and helped to promote the project the DRR component as a model to wooden railings and paved with
elsewhere. other communities, and to increase recycled plastic slabs (using 1
the awareness of technical options million recycled plastic bottles). It
Unsuccessful attempts were
to improve flood resistance. was one third cheaper than using
made to coordinate with other or-
new timber. Using the recycled
ganisations to resolve the water In the targeted village:
materials also avoided using 2,800
and sanitation issues.
Carpenters wrere trained on the timber slabs, equivalent to cutting
care and maintenance of the 15 trees that would take up to 40
‘We are happy because
houses. 55 carpenters received years to grow back.
we are going to resist the a recognized training on safer
waters, when the river The recycled plastic slabs were
construction.
guaranteed for 20 years with
will come, we will be 480 household water filters
reduced maintenance, three times
here, ready, resisting the and 500 individual filters were
the duration of timber.
flooding” delivered.
A solid waste management plan Logistics
was established and a compost Construction involved the trans-
Beneficiary area organised. portation of 24,500 sawn boards by
Seeds were produced in the boat.
collective garden to support
Community-based DRR Eleven woodcutters and five
replanting of timber species
Five communities and schools lumberjacks participated in the con-
used for construction.
were supported to enhance their struction. The timber used was a
preparedness for recurrent floods. Technical solutions local species of tree sourced from
This support included: Several elevated footbridges collective land or land belonging to
with a total length of 1.1km were individual households.
risk management plans
built to connect the main dock with
community risk maps The timber was processed into
most homes, schools, community
emergency equipment planks in the forest and then trans-
buildings and the community
trainings on disaster prevention ported to villages by boat where it
garden.
for community councils and the was then distributed by hand.
local authorities
training of thirty teachers and
local authorities in school risk
management
risk awareness and self-
protection training for school
children
a first aid post inside the schools
two disaster simulations
involving 820 people. Elevated walkways were built to enable the community to remain
connected when the floods next came.
Photo: Sandra D’Urzo/IFRC
A.4 Conflict
A.4 Côte d’Ivoire – 2010–2011 – Post-election Crisis

Keywords: Returns, Household items, Construction materials, Core housing construction, Hous-
Overview: ing repair and retrofitting, Vouchers, Advocacy / legal, Training.
Summary
The November 2010 election in Côte d’Ivoire triggered violence that
lead to the displacement of up to a million people. The western part of
the country was particularly affected. Families were displaced both within
Côte d’Ivoire and over the border in to neighbouring Liberia.
Support for returnees by international organisations focused on
rebuilding communities as well as houses. About 30 per cent of the
24,000 households whose houses had been damaged or destroyed were
targeted by the coordinated interagency response. About one third of
those assisted were in spontaneous sites.
Organisations supported only the most vulnerable households,
assuming that most households had the capacity to rebuild on their own.
Background were afraid to return to their The rest of the shelter support,
Côte d’Ivoire is a lower-middle- villages of origin. Many of those in the form of support for repairs
income country experiencing sig- wanting to return cited damaged and reconstruction, was largely
nificant demographic changes. The houses as one of the main impedi- targeted at returning IDPs and re-
proportion of people living in cities ments to return. patriated refugees.
in Côte d’Ivoire has risen from 15 The fragile security situation
per cent in 1960 to 50 per cent in Early Recovery Strategy
continued well into 2012. Given the problems at the core
2010.
of the crisis, it wasn’t simply the
Despite long term efforts by the Emergency phase in 2011
houses that needed to be repaired
government to encourage housing Although the Coordination
and rebuilt, but also the com-
construction through the private mechanism for the response was
munities themselves. The goal
sector, there remains a shortfall of established in January 2011, a Co-
was to support vulnerable house-
around 400,000 houses. ordinator was not in place until
holds through a community-based
March 2011.
Côte d’Ivoire’s development has approach that would promote
been hindered by conflict in 2002, Between January and September positive relations within the
2008 and 2010-2011. 2011, organisations assisted 8,150 community and to reinforce existing
households with emergency shelter coping mechanisms. The following
The conflict support. About 35 per cent of this two case studies (sections A.5 and
The violence associated with assistance went to support the A.6) all adopted this principle in
the 2010-2011 post-election crisis displaced people in various sponta- slightly different ways, depending
was particularly destructive in the neous settlements in the west, such on the context.
west of Côte d’Ivoire, where ap- as the Catholic Mission in Duékoué
proximately 24,000 houses were camp, which housed around
damaged or destroyed. 27,000 IDPs at its peak.
At the height of the crisis in early
2011, up to a million people were
thought to be displaced, including
over 700,000 within or from
Abidjan. More than 200,000 people
fled to neighbouring countries.
Relations between some com-
munities had been strained due Roofing with corrugated irons. (Toa-Zéo)
to issues of immigration, ethnicity Photo: Daniel N’dri YAO
and access to agricultural land. The
violence further damaged relations
between the different communi-
ties.
Lack of physical security in the
west due to ongoing hostilities Non-food items were provided to returnees and those directly affected by the
crisis.
meant that thousands of families Photo: Neil Brighton
12
24,000 houses were damaged in Western Côte d’Ivoire. It was estimated that nearly 65 per cent of the population would be
able to rebuild their houses without external assistance.
Photo: Neil Brighton
After the emergency, the Only vulnerable households Of the 24,000 damaged or
focus was on supporting vulner- were targeted, as a significant pro- destroyed houses in the west, in-
able households to rebuild their portion of the population was both ternational organisations targeted
mud-brick or wattle and daub able and willing to rebuild them- 8,775 vulnerable households in
buildings. selves. 2012. Of these, 7,200 had earth-
brick or wattle and daub houses.
Self-recovery Although the government had
Before deciding on a target the primary responsibility to assist Some organisations worked on
for the number of households to those affected by the crisis, it lacked confined masonry buildings, but
support, an assessment was made resources to support the entire pop- this was a small proportion of the
of how many people would be able ulation and was not able to respond response. Return kits were also dis-
to conduct their repairs without quickly enough. tributed to displaced households
external assistance. returning home.
In Western Côte d’Ivoire, the

vast majority of households lived in
houses that are built with materials
found locally and were either con-
structed from mud-bricks or wattle
and daub. The roofs were thatched
or covered in corrugated iron.
Though the house walls were prone
to erosion from rain and wind, and
were relatively weak, they were
built by the households themselves
and contributed to a high self-re-
covery rate.
By mid-2012, the affected com-
munities themselves had rebuilt
approximately 50 per cent (11,500
houses) of the destroyed mud-brick
and wattle and daub houses. This
type of construction made up ap- Destroyed home - the majority of houses were made from
proximately 90 per cent of the wattle and daub or mud blocks.
damaged or destroyed buildings.
A.4 Conflict
Government response The coordination team Closing of the

The government made both The shelter coordination team coordination system
food and non-food item distribu- consisted of one coordinator and By mid-2012, the security
tions in the emergency phase. In three protection monitors from a situation in Côte d’Ivoire was
the summer of 2012 the govern- local organisation. It was in place beginning to stabilize and life was
ment announced that a permanent from January 2011 to December returning to normal. The decision
housing project would target 1,000 2012. The protection monitors was taken in August 2012 to close
households in the Department of assessed the damaged and the coordination system by the end
Duékoué. Though this capacity was destroyed houses, and assessed the of the year. The table below sum-
welcomed, the decision was taken capacity of communities to recon- marises the collective goals for the
unilaterally with little consultation struct without external assistance. response and the extent to which
with the humanitarian community those goals were met.
This team proved invaluable
who had been working in the target
for collecting critical baseline data,
area since mid-2011.
which informed the shelter strategy
in different organisations’ project
planning.
Goal for 2012 Result achieved

Support 90 per cent of vulnerable households (6,489 households) with damaged or destroyed earth houses
4,461 households
(mud-brick or wattle and daub) to rebuild by 31st December 2012.
Support 25 per cent of vulnerable households (1,425 households) with lightly damaged confined masonry
434 households
houses to rebuild by 31st December 2012
Support 10 per cent of households (1,150 households) that are building back their own house with some
200 households
material or technical assistance by 31st December 2012.
Support 90 per cent of affected households (37,455 households) that lost essential household items with
37,455 households
distributions of NFI Return Kits by 31st December 2012
Self recovery: A man rebuilds his wattle and daub house without external assistance.
14
A.5 Côte d’Ivoire – 2010–2011 – Post–Electoral Crisis

Keywords: Returns, Urban neighbourhoods, Construction materials, Core housing construction,
Case study: Housing repair and retrofitting, Vouchers, Advocacy / legal, Training.

Côte d’Ivoire
Project location:
Duékoué, Western Côte d’Ivoire 19 months – – 1130 houses con-
Conflict: structed
Post-electoral crisis
Conflict date:
2010 to 2011
24,000 in Western Côte d’Ivoire
1 million people nationwide
150,000 displaced in the West
1,465 households
7,325 people 2 months – – Start manufacturing
bricks
Project outputs:
1st project: 335 households
2nd project: 1,130 households Côte d’Ivoire
Occupancy rate on handover: 1 month – – Identification of
needs, selection of
Between 75 per cent villages
and 100 per cent Duékoué
Shelter size: May 2011 – – Crisis ends
28m2, 2 rooms. April 2011 – – Voluntary returns in
Duekoué area
US$ 580 Materials
US$ 80 Labour November – Post electoral crisis
2010 – and displacements
(Total project / number shelters):
US$1070
Project description
The lead organisation worked with three partners to provide houses for vulnerable returnees, whose house
was damaged by the post-electoral crisis. The project had the goal to sustainably improve the living conditions of
returned households by providing one shelter per household. At the end of the project over 1,130 houses were
built or rehabilitated by one of the three partners.
Strengths and weaknesses 8 Difficulties arose in validating beneficiary lists as

9 Every beneficiary helped to make mud-bricks for the some chefferies saw opportunity to recover influence
whole community. This led to strong involvement of over some beneficiaries and NGOs. Traditional
the population throughout the project. decision-making systems, through “chefferie” were
9 By supporting local technicians, the project injected undermined by the post electoral conflict.
cash within the communities. 8 In a context of rivalry between communities and a
9 Training sessions entitled “building back better” weakened social cohesion, the shelter project targeted
gave people the opportunity to share experiences and mainly people from one ethnic group.
construction methods and to discuss different related - There was an unforseen challenge of holes left
issues such as sanitation and hygiene. from brick prodcution. These were dangerous for small
9 Having access to shelter was a starting point for a children during the rainy season and encouraged poor
new life and a durable return. sanitation making mosquito breeding areas. Work was
8 The project found it challenging to ensure that the required to reduce this risk.
beneficiaries were the owners of the land and houses - The organisation provided sand to beneficiaries. This
because many people had lost their papers during the was so that they could spend time on agricultural work.
crisis. rather than collecting sand.
A.5 Conflict
Background
See A.4 Côte d’Ivoire –
2010–2011 for background.
After the conflict

As a result of improved security
in Côte d’Ivoire in the West of the
country, part of the population
displaced during the post-electoral
conflict had started to gradually
return to their places of origin.
However, there was significant
damage to society, the economy
and infrastructure.
In the communities of return,
Households participated throughout the construction, manufacturing bricks,
there were significant humanitarian
preparing mortar and conducting other tasks.
needs and serious risks of secondary Photo: Antoine Vollet
displacement.
led to the final selection list of ben- householders recruited builders,
According to assessments,
eficiary households. who were then paid with vouchers.
food and shelter were indicated by
returnees as overwhelming priori- Land deeds verification The organisation provided tools
ties, followed by education, health- Before the beginning of the and equipment that had to be given
care and water. construction work, the land deeds back at the end of the construction
Intercommunity tensions, land that households provided were Every step of construction or
disputes and lack of access to basic authenticated. If documents were rehabilitation was checked by a
services represented major protec- not available, the identification of technical supervisor and the team
tion threats to returnees. Without land ownership was made in coor- leader. A form with key points
resolving housing issues it would be dination with the local community. was completed to check whether
difficult to address social needs. In every case the signature of the or not the house was ready to be
village chief was required. occupied.
Selection of beneficiaries
In the countryside and the Each beneficiary participated in
The organisation assessed many
villages, the traditional informal the following activities:
issues, including the numbers of
system is predominant. There was
destroyed houses, ongoing dis- manufacture of mud bricks
no choice but follow the statements
placements, and returns, mainly preparation of the mortar
of the chief of lands and the village
in two locations. Households were involvement throughout the
chief. In some questionable cases,
selected based on criteria defined construction so that they could
the organisation also interviewed
by the organisation with the com- later upgrade their houses.
the neighbours. The land service
munities. Two non-negotiable
of the municipality was sometimes Training
criteria were that:
also able to help. Regular trainings and meetings
the household was affected by were organised by the organisation
There were some cases where
the post-electoral crisis in order to keep a high level of mo-
there were lacking title deeds, and
their house was either damaged tivation and involvement through-
conflict over the land. This was
or destroyed. out the project. Specific attention
often due to conflicts between
Other criteria, such as the siblings. was paid to the following aspects:
household social and economic
Eventually only 6 households In most communities, the
situation before/during/after the
were excluded on account of land population was not accustomed
crisis, were agreed to better assess
not being identified. to working together and every
the household’s vulnerability with
step of the project required a
respect to shelter security. Implementation meeting with all households.
Based on these criteria, a pre - All construction materials were Rehabilitations often require
selection list was written down by provided. Doors and windows were technical skills and as a result
each village committee, if it existed, constructed by local carpenters. are led by local masons
or the Village Chief. Metal sheets were given for the and carpenters. To ensure
roof. participation, beneficiaries were
People on this list were surveyed
asked to collectively produce
with around fifty questions to verify One mason and one carpenter
mud-bricks.
levels of vulnerability. The survey were paid to work on several
Some beneficiaries finished their
houses. In some remote villages
houses earlier than the others.
16
Roofing with corrugated irons. (Toa-Zéo) Drying mud bricks. (Niambly)

Photo: Daniel N’dri Yao Photo: Damien Laporte
They were required to continue Coordination also allowed or- Staffing

participating in the fabrication ganisations to exchange information The entire project was managed
of mud-bricks. on technical issues and challenges by a staff of 22 people: A project
The following trainings were faced as well as to share analysis manager assistant, a field logisti-
conducted: about socio-economic trends. cian, a mobilisation team leader,
7 mobilisation agents, a technical
Mobilisation and the role of Technical solutions team leader and 11 technical su-
the committee: The committee Two designs were proposed for pervisors. The team used 4 cars
was established to assist the the construction, and households (pick-ups and one 4x4)
organisation in the daily work chose the design that they wanted:
and to take project ownership. Classic: walls made of dried Logistics
Trainings about mobilisation mud-bricks joined by mortar In each village, with the support
were repeated every time they with a corrugated iron roof of the population, a storage area
were needed. supported by a wooden roof was identified for all construc-
How to improve the shelter: structure. tion material for every household.
Before the households choose Improved: This was a more rain This area was managed by a local
the shelter design, a training resistant shelter. The walls were storekeeper chosen by beneficiary
gave advice on improving the made of dried mud mixed with households and supervised and
resistance of a house against rain cement bricks, with cement trained by the organisation.
and humidity (simple principles mortar on the base and the first All supplies were purchased
of the reaction of mud-bricks to four rows, and with mud mortar from the nearest town of Duékoué.
humidity, and how protect the for the rest of the building. The
base from water). roof and its structure were the Materials list
Explaining what is expected from same as the traditional design. Materials Quantity
the committee members and Wall and base construction:
role allocation (e.g. president, Rehabilitations Sand 3m3
secretary, storekeeper). Where buildings were rehabili- Cement “A 32.5 N” 12 sacks
Mud-brick 1,200pcs.
There were occasional tated, repairs were based on an as- Red wood 15cm x 3cm x 400cm 1pc.
awareness raising activities sessment of needs and observed Doors and windows:
regarding cleaning the village. damage. Most of the time, they Plank 25cm x 4cm x 400cm 7pcs.
consisted in replacing or repairing Rafter 8cm x 6cm 4pcs.
The trainings were conducted Nail n°6 2kg
by the mobilisation team members. the roof. Nail n°8 1kg
Technical trainings were given by Every building was assessed Nail n°10 1kg
Crochet medium 2pcs.
the technical supervisors. by the technical supervisors who Pairs of split hinge 140 steel 4pcs.
completed a bill of quantities. This Paris of split hinge 110 ordinary 4pcs.
Handover was then checked by the technical Door handle 2pcs.
When the project was over in Lock 2pcs.
team leader and the programme Wood screw 1 packet
a village, the village committee manager. A random control took Carpentry:
initiated a key-giving ceremony. place in every village, led by the Rafter 8cm x 6cm 18pcs.
Program Manager and Technical Rafter 6cm x 4cm 12pcs.
Coordination Nail n°8 1 packet
Few organisations were working Team Leader. Wire 15m
in the same area, and coordina- Sometimes, the level of support Roof:
tion helped to avoid gaps and du- Corrugated iron (2m x 0.8m) 33
required was too high for the Nail n°6 2 packets
plications in areas targeted by the available budget. In these cases the Nail n°8 1 packet
different actors. beneficiary household was asked to Rubber band for washers 5pcs.
provide materials to fill the gap.
A.6 Conflict
A.6 Côte d’Ivoire – 2010–2011 – Post–Electoral Crisis

Keywords: Returns, Urban neighbourhoods, Construction materials, Core housing construction,
Case study: Advocacy / legal, Training.

Côte d’Ivoire
Project location:
Duékoué, Western Côte d’Ivoire
Conflict:
Post-electoral crisis
20 months – – 1,341 shelters
Conflict date: complete
2010–2011
Approximately 24,000 houses in
the west of the country
1 million people nationwide
150,000 displaced in the West
8,046 people 8 months – – 421 shelters
complete
Project outputs:
1341 shelters
Occupancy rate on handover: Côte d’Ivoire 5 months – – Project start date
99% of the first 421 shelters
occupied in July 2012 Montagnes
Shelter size:
36m² (3 rooms) for the house + May 2011 – – Post electoral crisis
2m² for the latrine. ends
US$ 585 (Material), US$ 70 (Labour) November
US$ 200 (Beneficiary contribution) 2010 – – Post electoral crisis
and displacements
US$ 886
Project description
This shelter intervention built 1,341 shelters, supporting participation at the household and community levels
through self-help groups and shelter committees. The shelter design used abundant local resources and promoted
a design well known by the beneficiary households and local builders. The goal of the project was to contribute
to the return process through shelter rehabilitation for the most vulnerable households.
Strengths and weaknesses electoral crisis, or as a result of a previous crisis.

9 More than half of the work was completed by the 8 The project staff found it challenging to resolve
beneficiaries through self-help groups. land tenure disputes. There was no formal system of
9 Maximises the use of local resources which provide land tenure security, and some disputes arose when
all the masonry material (bricks and mortar) while at shelters for migrant households were rehabilitated.
the same time limiting local environmental impact. Work continued into 2013 to solve the disputes.
9 By adapting the design, and ensuring strong 8 The second phase of the project began a few
community involvement and good quality of work, months before the start of the rainy season in March
capacity to build and to maintain shelters was improved. and ended two months after the rainy season in
9 By using mud blocks and mortar, the organisation December. This greatly affected the production of mud
built larger shelters with the same cost as shelter bricks as well as masonry works.
projects led by other organisations. 8 Despite an initial awareness campaign at the start
9 Given that one of the major concerns of the IDPs of the project, it was necessary to regularly re-explain
and refugees was the loss of their homes, shelter the beneficiary selection criteria, especially with newly
reconstruction supported durable return after the crisis. arrived returnees that could not be selected given the
8 It was sometimes difficult to verify whether the time and resource limitations of the project.
house was destroyed during the 2010–2011 post
18
By using mud and other local resources, the project was able to reduce costs and build more shelters.
Photo: Yao Albert Konan
Background For 40 families with land issues, Self help groups

See A.4 Côte d’Ivoire – solutions were found by working Self-help groups were created
2010–2011 for background. with the local administration. This with the aim of encouraging col-
was done with the assistance of a lective work, especially to ensure
Selection of beneficiaries legal assistance programme that bricks were available for weak,
The first project was imple-
the organisation was running. It old or disabled people. In practice,
mented in Duekoué and Bloléquin,
took about four months to agree it was almost impossible to mix
departments where the reported
on durable land for these families. people from different communi-
destruction was most severe. About
ties to work together, and it was
2,200 houses were destroyed in the Implementation difficult to stimulate a team-work
11 selected districts. 2,500 mud bricks were with 8 to 10 families to produce
The communities provided their produced per household (2,000 for enough bricks. The majority of
own list of households, which was the shelters and 500 for the latrine) households decided to work alone
confirmed by a door to door survey. through the work of the self-help or with family members.
group formed of 8 households.
The vulnerability criteria were Each self help group was provided At the end of a training session,
based on: gender of head household, with tools and brick moulds at the each self-help group received a
age, presence of disabled people start of the project. A community construction kit to share (spade,
in the family, household size, mobiliser and the shelter committee hoe, shovel, 1m3 water tank and
household economic resources, supported the beneficiaries jerrycan).
food security scoring, ownership or throughout the process.
access to land, and willingness to Shelter committees
participate in the reconstruction of The organisation started con- Shelter committees were estab-
the shelter. struction once the beneficiaries had lished to empower and mobilise
produced the required number of people in the project. They regularly
The provisional lists were publicly mud bricks and dug the latrine pit. checked on the number of bricks
posted for two weeks to allow for The first step in the construction made, and created a ranking which
feedback from the community. was the trenching and laying of the determined the order in which they
foundation. would build houses.
Land
Formal land documentation The wall was built in three steps, Training
generally does not exist in Western with two days to dry at each step: Basic messages were shared
Côte d’Ivoire. Sites were visited 1) five rows of bricks, 2) five rows of about maintenance of the drainage,
with the traditional authorities to bricks, and 3) build the gable. The plastering the sill as well as door
certify that the head of family was work was done by a mason while making and installation of latrines.
the landowner. the household prepared the mortar
and supplied the necessary water. Each household received a brick
There were some conflicts mould. Trainings about mud bricks
between different communities, Constructing the roof took two production and self-help group
often between Autochthon com- days: one day for the carpentry and work were held at the start of the
munities and migrant communities. another day to fix the corrugated project. These trainings took about
roofing sheets. Simultaneously, the half a day per group and were led
mason built the latrine walls. by a site manager.
Only 40 households out of
2,200 had land tenure issues Once all houses in the village Each household received two
were completed a closure ceremony 1½ hour trainings on shelter and
that required some external
was held. hygiene promotion. In total there
intervention...
were six trainers (five technicians
and one social mobiliser).
The project aimed to support durable returns. 80 per cent of the shelter occupants maintained the drainage.
Photo: Yao Albert Konan
Coordination Staffing and structure Shelter committees distributed

The main humanitarian actors The organisational structure materials supervised by the or-
acting in the shelter response met was: ganisation. Materials were distrib-
twice a month until July 2012 uted on completion of each phase
one social mobiliser responsible
thereafter meeting once a month. of construction. Special attention
for group mobilisation, hygiene
Meetings were held in both Abidjan was paid to the corrugated iron, as
promotion and assessments
and in the West. households were tempted to sell it.
five site managers (one for two
Coordination helped to define to three locations) responsible 60 to 80 different masons and
the areas of intervention between for following works, masons, 10 to 20 different carpenters were
the different organisations as well carpenters, trainings and directly contracted, mainly from the
as to communicate figures from the materials supply. Site managers villages where the shelters were to
start of the returnee movement. and mobilisers spent 80 per be built.
In addition, coordination was cent of their time on the ground
essential in order to share informa- six community mobilisers (one Maintenance
tion on design, costs and to adopt for two locations). Locally hired Around half of the shelters were
a common response on the ground. community mobilisers received upgraded by their occupants with
a monthly allowance and concrete screed and plastering.
Technical solutions monitored construction However people mainly plastered
Mud bricks were selected as 11 committees in which inside the room in preference to
the easiest way to ensure a good positions were chosen to plastering the façade, failing to
quality of implementation, as it is a represent the three communities maximise shelter durability.
very common construction material in the region At the end of 2012, about 80
in western Côte d’Ivoire. Cement one project coordinator to per cent of the drainage around
was not used in the mortar as it supervise the operations. the shelters was still maintained.
would be above local standards and More than three-quarters of the
would increase the cost per shelter Logistics
Tenders were issued for rein- latrines were in use, although some
thus decreasing the number of ben- were used as showers. Hygiene
eficiaries. forced concrete slabs for the latrines,
corrugated iron sheets, timbers and promotion activities continued into
The organisation referred to other materials. Suppliers delivered 2013.
the shelters as “improved design” directly to each community, except Some masons contracted by the
relative to other houses on account for roofing sheets, which were organisation built the house design
of the corrugated iron roofing centrally warehoused. for other private contractors, but
sheets, latrines and quality of the they did not use metal roofing
platform. It was based on a common The mud bricks were produced
locally in the communities. Each sheets due to the cost.
design of shelter in Western Côte
d’Ivoire but was larger than many household stored them close to the Materials list
houses in the area. future construction site. Materials Quantity
Field warehouses were set CGI sheets 45 pieces
up to store timbers, frames and Timbers 33 pieces
equipment. Mud bricks 2,500 pieces
20
A.7 Democratic Republic of Congo – 2002 – Volcano

Keywords: Returns, Urban neighbourhoods, Household NFIs, Construction materials, Transi-
Update: tional shelter / T-shelter, Community engagement, Mass communication.

Democratic Republic of Congo
Project location:
Goma
10 years – – Long term assess-
Disaster: ment of impacts.
Goma volcano eruption in 2002
No. of houses damaged:
15,000 houses destroyed
(20 per cent of Goma’s housing
stock) – Periodic monitoring
300,000 people displaced
Project outputs: 9 months – – 5000 families as-
5,000 families supported with sisted with transi-
shelter and latrine packages tional shelter
Shelter cost:
US$ 250 average cost: Shelter
Goma 3 months – – First transitional
and latrine (materials and
labour) shelters erected
Democratic Republic
of Congo
January 2002 – – Volcano erupts
Project description
This case study summarises an assessment by a major donor of the transitional shelter and recovery
programming that it funded in Goma following the volcanic eruption in 2002. The assessment was conducted ten
years after the initial response. The assessment found that transitional shelter did help to facilitate the transition
to permanent housing, and became a base for many livelihood activities. It also found lasting impacts from both
the settlements approach taken and from the supporting activities to help people in Goma to “live with risk”.
Strengths and weaknesses transitional shelter in the city, based on the “city-
9 Transitional shelter really can facilitate the transition focused” approach, maximised and concentrated the
to permanent housing. As intended, nearly all of the economic benefits associated with investments made
original 5,000 "t-shelters" have been improved in some by the humanitarian community. In turn, residents have
way as part of making it a permanent home. A site visit had better access to jobs and public services in an urban
2012 noted that most beneficiary families continue to context than in a remote camp, contributing further
live in their transformed transitional shelters. After ten to the recovery of their city. Disaster Risk Reduction
years, some families are still making improvements measures were incorporated into the reconstruction of
leading to permanence, suggesting that the process of road and service networks, to enhance both evacuation
incremental housing development is both evident and options as well as access to land and housing markets.
likely to continue in the foreseeable future. The city-focused approach orientated humanitarian
9 Transitional shelters have become “shophouses”. assistance towards settlement planning and also
As intended, many project beneficiaries have expanded reflected beneficiaries' wishes to return to their own
their shelters to create space for livelihood activities of neighbourhoods.
all kinds, thereby either restoring livelihoods lost in the 8 In Shelter Projects 2008, the implementing
disaster, or creating new economic activity using the organisation noted:
shelter as a much-needed platform for production. 8 For families with eight or more people, shelters
This has contributed to both community and regional were initially not big enough.
economic recovery since the volcanic eruption. 8 Some people felt that plastic walls compromised
9 A deliberate focus on “Shelter and Settlements” is a their privacy and security.
critically needed approach to humanitarian assistance - The project was one of the first-ever attempts by the
in urban areas. Longer-term recovery was dependent donor to promote an explicit shelter and settlements
upon regenerating its urban economy. Providing approach to shelter activities.
Left: Recovery work began as lava cooled.

Right: Road work was linked to evacuation planning.
Photos: USAID/OFDA
The volcano Although eruptive activity ended People displaced by the volcano
(See case study A.1 in Shelter within 24 hours, seismic activity needed a place to call “home”.
Projects 2008) related to the volcano continued
Options for meeting this need
until early February 2002. On
Nyiragongo, a volcano located included moving the entire city to
February 9 seismologists declared
approximately 16 kilometers a new site, dispersing people to
that the eruption was over.
(ten miles) north of Goma, the different regions of the country,
major town in Eastern Democratic Since early 2002, Goma has moving people into camps, and a
Republic of Congo (DRC), began subsided by nearly 50cm. Minor “city-focused” option aimed at re-
erupting on 17th January 2002. subsidences have periodically habilitating Goma itself, allowing as
Lava flowed from the southern occurred as a result of on-going many people as possible to remain.
flank of the volcano, heading tectonic activity. These options were discussed at
towards Goma. length among representatives of all
Response key stakeholders.
This eruptive activity triggered With thousands of jobs lost, and
an exodus of Goma, a city of ap- the urban and regional economy The perceived and real security
proximately 450,000 people. Of devastated, national and interna- and political conditions in the
which an estimated 300,000 people tional organisations mounted a immediate region affected decisions
fled briefly to Rwanda, while others rapid response, with the interna- in shelter assistance by constrain-
fled to settlements to the west of tional community contributing a ing relocation options to the east,
Goma as well as elsewhere within total of US$ 40 million in assistance. north, and west of Goma. The city
the DRC. Most people returned to is also located on the northern
the city within three months. In this case-study, the donor’s shore of Lake Kivu, making large-
share of the contribution was scale southern movement of the
The lava flows and subse- nearly US$ 5 million. This included displaced impractical.
quent fires caused severe damage US$ 2.6 million in emergency relief:
in Goma. An estimated 13 per water, food, health, and non-food There was also the local security
cent of the city’s 35km2 land area assistance (including blankets, consideration that many people
was covered by lava. It heavily household goods, and plastic wished to remain close to their
inundated the central part of the sheeting); and a US$ 2.3 million former houses to prevent appro-
city, destroying up to 15,000 programme featuring a transition- priation or looting.
dwellings (20 per cent of the city’s al shelter project and disaster risk Following consultations with
estimated housing stock). In in- reduction activities. affected communities and au-
undating the most developed thorities, the donor devised a two-
portion of the city, arguably the The response featured the
design and implementation of one pronged strategy that would bring
most developed portion of eastern new life to Goma and reduce the
DRC, the lava flows destroyed of the donor’s first transitional
shelter projects. impacts of future disasters.
numerous economic enterprises
and community structures, and Recovery Transitional Shelter
thus thousands of livelihoods. After critical needs had been Due to the security, safety and
An estimated 90,000-105,000 addressed, the humanitarian economic concerns of the affected
people, many of whom were community began to develop strat- population, the first element of
already vulnerable because of con- egies for helping residents of Goma the programme was to support
flict-induced insecurity and limited rebuild their lives and livelihoods. a city-focused transitional shelter
economic opportunities, lost their Shelter quickly emerged as the most program, devoting 80-85 per cent
homes and other assets, and were pressing need for affected families. of program funds to the Goma
in need of shelter. urban area. The donor and its
22
“The central business district,

buried under rock, is re-emerging;
there is even a new Volcano
Internet Café on the edge of the
destruction. The camps set up
for displaced residents are now
mostly shuttered, and Goma is
experiencing something of a housing
boom.”
The New York Times (emphasis Transitional shelter (left) has evolved into permanent housing (right) for
thousands of families.
added), November 10, 2002
Photo: USAID/OFDA
partners determined that there itoring (provision of equipment, The Goma Volcano Observatory
was sufficient space in Goma to staff support, and technical assis- continues to operate many
shelter residents there, and that tance to the Goma Volcano Obser- community-based education
the existing social and economic vatory). activities, although updating
infrastructure, even post-eruption, is required. Activities include
The donor also sponsored
made it easier to assist people in the providing volcano activity
a two-year, community-based
city rather than elsewhere. reports to radio stations,
Disaster Risk Reduction program
sharing information at a local
The donor funded a single or- linked to the Goma Volcano Ob-
volcano information center, and
ganisation to provide shelter in servatory to enhance early warning
updating alert levels in public
Goma to assist 5,000 households. systems, upgrade evacuation
areas.
All of the households were assisted routes, and improve community
Over time, nearly all beneficiary
within nine months of the eruption. awareness of what to do and where
families transformed their
Other donors saw the efficacy to go when eruptions and earth-
transitional shelter into
of this strategy and provided a quakes occur.
permanent housing, resulting
combined total of 8,000 addi-
The road network was expanded in the re-establishment of local
tional households with transitional
following discussions with local markets and communities,
shelter. A further 2,000 households
officials and representatives. This contributing to overall recovery.
received other assistance from a
was intended to increase the The rapid response to the 2002
variety of other organisations.
number of evacuation routes. volcanic eruption, the incorporation
Assistance was used to expand of Disaster Risk Reduction into the
or supplement host family homes, Outcomes
Despite the considerable response, and the explicit shelter
or build on under-used or vacant and settlements approach adopted
private residential parcels of land. changes in Goma during the
2002-2012 period, including recent were aimed at strengthening the
Shelter supplies were sufficient resilience of Goma’s citizens by
to create 21m2 of covered living conflict in and near the city, several
outcomes of the donor-support- promoting recovery and lessening
space for an average beneficiary the impact of future disasters.
household of up to six people. The ed post-eruption activities have
supplies included plastic sheeting, become visible over time: The 2012 assessment by this
zinc roof sheeting, wood framing, In addition to providing much- donor found that the activities
and concrete screed flooring. A needed shelter, the city-focused that it supported have contribut-
modest latrine was also provided. programme had a significant ed to a transition to recovery and
impact on Goma's economy. reconstruction. This outcome is
Three-quarters of households notable, for it demonstrated the
were assisted on land occupied by Beneficiary families supported
nearly 45,000 person-days utility of using shelter as a means of
host families (relatives or friends); promoting economic recovery and
many of these beneficiaries have of labour to transform their
transitional shelters into linking humanitarian community
remained on hosted land. shelter activities to the process of
permanent homes. This
Living with risk generated nearly 3,600 new longer-term permanent housing
The second element of the jobs, and helped to jump-start development. Furthermore, shelter
strategy was rooted in the basic economic recovery in Goma. activity was deliberately concen-
message of learning to live with risk: Volcano monitoring is ongoing, trated in neighbourhoods, where
a Disaster Risk Reduction program with most of the equipment people wanted to resume their lives
me was designed to promote im- provided still functional, though and livelihoods. It enabled people
provements in volcano hazard mon- upgrades are needed. to learn to live with risk, supporting
them with risk reduction activities.
A.8 Conflict
A.8 Ethiopia – 2011 – Sudanese Conflict

Keywords: Planned and managed camps, T-shelter, Site planning, Training, Tools, Construction
Case Study: materials, Infrastructure.

Ethiopia
Project location:
Bambasi camp, Assosa
Conflict:
12 months - – Second partner
Sudan and South Sudan conflict begins shelter work
Conflict date: 11 months - – First partner begins
September 2011 shelter work
10 months - – Work on site begins
40,000 refugees by end of 2012
Project outputs:
Camp for 12,000 people (3600
6 months - – Site master planning
households)
2,175 shelters built (two
organisations, 70 percent built
by one organisation) – Need for new camp
identified
Shelter size: One month - – 54,000 Sudanese
Sudan
<2 people: 10m2 refugees in Ethiopia
3-4 people: 14m2
4-6 people: 21m2
September
Cost per shelter: 2011 – – Conflict restarts in
US$ 640 - 10.5m2 Ethiopia Sudan
US$ 800 - 14m2 Assosa
US$ 920 - 21m2 2006 - – Returns begin
South
Sudan 1990s - – Previous
1987 - history
1983 - of conflict and
1969 - refugee influxes
Project description
The organisation planned and built a camp for Sudanese refugees. Semi-permanent shelters were constructed
by refugees, with two partner organisations providing materials, carpenters and training. Refugees were able to
chose their own plot configuration and the shelters were constructed with locally procured materials.
Strengths and weaknesses 8 Difficulties in sourcing and transporting mud for the
9 The shelters followed local housing designs to make walls were not foreseen.
them cool in the day and warm at night. 8 Initial estimates of construction time were too low,
9 The shelters were cost-effective, and were durable and additional carpenters and masons were required.
alternatives to tents. Fewer shelters were built than initially anticipated.
9 Materials were procured locally, reducing transport 8 Many refugees did not recieve a shelter. Of those
costs and injecting cash into the local economy. This who did, many received a tent whilst waiting.
provided some economic compensation to the host - Bamboo is grown extensively in the area, the
community. eucalyptus was sourced from a neighbouring state
9 Shelter dimensions were tailored according to owned forest.
family size. - There was a very strong input from the government
9 Each plot was provided with a fence, a latrine and of Ethiopia in all issues relating to the camps.
a shelter. - Many families were separated when the first families
9 Refugees contributed labour to build the shelters. arrived. The rehousing of refugees was undertaken
This helped to foster a sense of ownership. in parallel with replanning the camps and a family
8 Inital plans to for the organisation to build the reunification exercise.
shelters itself were dropped as other organisations had - Sudanese refugees brought large numbers of
management systems better suited to implementation. livestock with them. Space for animals in the camp had
8 Technical staffing capacity was a constant challenge. to be allocated (See B.3).
24
Number of Zones: 3
Number of Blocks: 40
Number of Communities: 265
Number of Family Plots: 5,240
Average Plot Size per Household:
15mx10m
Site development
Despite delays, by the summer
of 2012 plot demarcation had
begun and the road was upgraded
in order for it to be functional
during the rainy season.
Once Bambasi camp was estab-
lished, water was provided from
eight shallow wells (up to 60m
Refugees were supported with materials, carpenters and deep). Later three boreholes were
masons to build tukuls based on the designs used by the host developed and a system of 34 tap
community. The refugees contributed labour. stands was established.
Photo: UNHCR
Shelter construction
Before the influx Most of the new arrivals from The organisation initially planned
Sudanese refugees have sought the Blue Nile State stayed with host to build all of the shelters itself,
safety in Ethiopia since 1969, first communities in border areas, and and built some sample shelters.
settling in the Gambella region. a transit centre was established at However, it became clear, that the
Additional refugee influxes 1983, Ad-Damazin. With the camps at full organisation lacked the manage-
1987 and the early 1990s led to capacity, this transit site became ment systems required to build the
the creation of five refugee camps more permanent. Given the scale of numbers of shelters required. As
in western Ethiopia. Three were es- the influx of refugees, new camps a result an alternative implemen-
tablished in the Gambella Regional were needed. tation process was chosen, using
State and two in the regional state partner organisations.
of Benishangul-Gumuz in the area Site selection
surrounding the town of Asossa. Negotiations began with the Two organisations were iden-
national government’s refugee tified to implement the shelter
Following a peace agreement agency and the local government programme.
between north and south Sudan, to identify sites.
refugees began returning home Implementation by the partner
from March 2006 onwards and A 450 hectare site owned by organisations began in August/
three of the camps could be closed. the adjacent village was identified September 2012.
23,000 refugees remained in one at Bambasi, 50km from the border.
It had with suitable drainage and A fixed design of shelter (a
camp in Gambella and a further tukul) was built. It was based on
4,000 remained in one of the Assosa access and was around 600km or
a two day drive from Addis Ababa, the shelters built and lived in by the
camps. The refugee population host community, differing from the
included several hundred refugees the capital of Ethiopia.
shelters that the refugees were ac-
from the Democratic Republic of The host community and the customed to building. As a result
Congo. Those remaining in the refugee population had a similar construction training was required.
camps included several hundred tribal heritage which, once some
Congolese people. initial differences were resolved, led Carpenters and materials were
to a good relationship between the provided and managed by the im-
Displacement in 2011 two communities. plementing organisations, while
Fighting recommenced in families had to provide the labour.
September 2011 in the Blue Nile Site planning Most families were able to provide
State of Sudan, displacing more From March 2012 the prepara- the labour, but in the case of the
refugees into Ethiopia. tion of the master plan began. The most vulnerable households, some
plan took four months to develop support was required.
By mid October 2011 there were
an estimated 54,000 Sudanese and agree. The process was signifi- The implementing organisations
refugees and asylum seekers in cantly delayed by complications in both provided a site engineer to
Ethiopia. About 34,000 were reg- awarding the contract to build the lead the project and a site foreman
istered and accommodated in access road. to manage the teams of carpen-
three refugee camps: Sherkole and The camp was designed to be ters and masons in the camp. Both
Tongo near Assosa and Pugnido no closer than 500m to the village. organisations required significant
near Gambella. The camp was divided as follows: logistics support.
A.8 Conflict
Timber and bamboo frames shelters were built with thatched roofs as a more durable shelter solution than tents. It proved
challenging to source mud to plaster the walls as originally intended.
Photos: Left: UNHCR, Right: Demissew Bizuwerk / IOM Ethiopia.
In the project plans, a carpenter Logistics in June and July 2012. A few
and a mason, working with families All of the shelters were built dozen refugees moved spontane-
would be able to build ten shelters using locally available materials: ously to Bambasi in September.
in fifteen days. In practice, only half bamboo, grass (for a 15cm thick In October 2012, 2,000 refugees
the number of shelters could be roof), rope and mud. This approach were relocated to the camp by local
built. This was due to an underes- was much cheaper than sourcing officials and were accompanied by
timation of the training required by materials in the capital, also cutting around 8,000 livestock.
those constructing the shelters, and transportation costs.
an underestimation of the number Materials list
Each shelter required signifi- Below is a materials list for
of households who would require
cant volumes of grass for thatching different shelter sizes.
additional assistance.
the roof and for strengthening the
Selection of beneficiaries mud walls. The grass could only be Material Quantity /
Refugees were brought to harvested seasonally with the main shelter size
Bambasi camp from the transit harvest being in March. This did Shelter (small) 10m2 14m2 21m2
site near the border at Ad-Dama- not coincide with the construction, Eucalyptus poles 10 11 14
which needed to continue all year 5m x 10cm
zin. The refugee population had
round to meet the needs. Eucalyptus/ bamboo 27 33 40
continued to rise while the camp 5m x <8cm
was being built, and many refugees The sourcing of sufficient quan- Eucalyptus pole 1 1 1
had settled near the border. tities of mud also proved more chal- 5.8 m x 12cm
lenging than anticipated. Initially Bamboo 5m length 37 39 90
Shelters were allocated
according to family size. Each family mud came from digging the latrine Mud with grass (m3) 2.45 4.37 4.89
was allocated a 10m x 15m plot. For pits but this was insufficient for Bamboo (roof) 20cm 57 66 80
families with seven or more people, the initial shelter needs, and for re- Bamboo reeds 50cm 25 45 50
two plots were allocated. mudding after the rains. By the end Grass (bunch) 15 22 30
of 2012, the organisation was still Rope and strings 100m 150m 200m
trying to identify sources for mud Door with frame, 1 1 1
and to organise sufficient trucking hinges and lock
for the large volumes required. Window 0.6m x 0.6m 1 1 1
Nails 2. 5’’ (kg) 0.5 0.5 1
The camp water supply was Nails 4’’ (kg) 1 1 2
sufficient to cope with the volume Used motor oil 3litres 3litres 5litres
needed to mix with the soil.
CIS Nails 0.25 2.5 0.25
Situation at the end of Fence
2012 Eucalyptus 5
5m x 10cm
By the end of 2012, there were
Bamboo (1m spacing) 80
over 86,000 Sudanese refugees
Tools
living in Ethiopia.
Claw Hammer 1
Approximately 3,700 refugees Bow saw 1
formerly registered in Ad-Damazin Shovel 1
A carpenter and a mason worked still remain in the local community Meter rule 1
with each family to build shelter. after opting out of the formal re-
Photo: Demissew Bizuwerk / Pick axe 1
location process to Bambasi camp Axe 1
IOM Ethiopia.
26
A.9 Ethiopia – 2012 – Conflict and Drought

Keywords: Planned and managed camps / relocation sites, Transitional shelter / T-shelter, Site
Case Study: planning

Ethiopia
Project location:
Dollo Ado 18 months – – Construction
Conflict / Disaster: ongoing
Conflict and drought in Somalia 18 months – – Transitional shelter
strategy reviewed
Conflict date:
Conflict since 1992
Over 1 million registered Somali
refugees
By the end of 2012, 177,000 14 months – – First transitional
refugees were registered in the shelter strategy
five Dollo Ado refugee camps
9,000 families (2011-2012) 13 months – – Shelter prototype
Project outputs: evaluation
7 months – – Criteria for Tran-
7,127 shelters by end of 2012 sitional Shelter
Occupancy rate on handover: adopted
High
Shelter size:
6m x 3,5m (21m2) Ethiopia
Materials cost per shelter: February
US$ 525 including transport 2010 – – First camps opened
at Dollo Ado
Project cost per shelter: Dollo-Ado
US$ 800 excluding overheads
1992 – – Conflict starts
Project description
Four organisations built semi-permanent shelters for Somali refugees living in the camps at Dollo Ado. Each
organisation set up production lines in the camps to prefabricate the components. The projects worked within
the constraints of challenging logistics and very different social environments between camps. The shelter design
was selected following a consultative process during which different options were shared with camp residents.
Strengths and weaknesses 8 The original design used mud render, but
9 The design process was coordinated between this required a significant amount of water and
organisations to avoid conflict between refugees over transportation, and was not possible to implement.
different shelter standards. Negotiations with the host communities over the use
9 The process to select the shelter design was of mud slowed progress.
designed to be transparent and include all stakeholders, - Different organisations have had very different
including camp residents. completion rates as a result of different budgets,
9 Shelter construction provided refugees and the management structures, logistics, supply and relations
host community with paid work. It is estimated that with camp residents and host communities.
the shelter projects contributed US$ 16,000 per month - Most materials were not available in Dollo Ado
to the economy of each camp. markets and were imported from other regions. Price
8 Joint procurement of supply of materials was fluctuations led to a 16 per cent increase in the total
attempted but did not prove successful. shelter cost.
8 Significantly fewer shelters have been built than - The shelter strategy was developed based on the
initially anticipated. The strategy supported less than assumption that it would acheive 100 per cent shelter
20 per cent of the population of the camps by the end coverage. Production and delivery remains short of
of 2012. these targets.
Complted shelters in Buramino camp at Dollo Ado. It was initially intended that the shelters would be plastered in mud but
this proved difficult to implement.
Camps at Dollo Ado Beneficiary selection A well-organised workshop

Following a resurgence of the Shelters were built by four or- with effective quality-control
conflict and drought in Somalia, ganisations and each was allocated mechanisms was necessary to
a series of five camps were estab- one or two camps. Camps were maximise production efficiency. The
lished in 2010 and 2011 within 100 established within host communi- minimum workshop and storage
km of the Ethiopian Border town of ties or in isolated locations. Some area for efficient production was
Dollo Ado. had been established for months 1 hectare (10,000m2). Workshops
whilst others had existed for years. were staffed by a mixture of skilled
By the end of 2011, the five
As a result each site presented very carpenters and daily labourers.
camps of Bokolmayo, Melkadida,
different challenges. One organisation found par-
Kobe, Hilaweyn and Buramino
hosted 34,000 Somali families, There was some variation in ticular challenges with the splitting
the largest refugee presence in beneficiary selection: One or- of bamboo, facing a 50 per cent
Ethiopia. The refugee population ganisation targeted blocks in each shortfall at the time that the
increased during 2012, and by the camp according to agreed criteria. bamboo had to be fixed. It turned
end of the year, 180,389 individu- Within each block the most vulner- out that this was due to many
als refugees were registered in the able households were prioritized, bamboo poles being split into two
camps. As these camps became and all shelters in each block were pieces by the daily labourers as
more established and the numbers completed before moving on to the opposed to four or six.
of registered refugees continued to next. Another organisation prior-
increase, it seemed likely that the itized individual households across Training and supervision
camps would remain open for some the camp rather than prioritising Training was provided for skilled
years. individual blocks. labourers who were responsible for
the on-site construction. On-site
Being close to the equator and In addition to building shelters works included digging holes for
at low altitude, Dollo Ado is subject for camp residents, 120 shelters foundations, erecting the frames,
to harsh weather conditions with were built for host community fitting the roofing, covering the
high temperatures, strong winds households in 2012. Additional walls with bamboo slats and fixing
and seasonal heavy rains. targets were set for 2013. windows and doors. Training in
The people living in the camps mud rendering for walls was given
Implementation
mainly come from rural parts of where mud was available.
Each implementing organi-
Somalia. A significant proportion sation started at different times All organisations directly
are nomadic pastoralists, accus- with different total budgets and in hired both skilled staff and daily
tomed to lightweight and movable different sites. The most effective labourers. To select carpenters for
shelters. projects established strong supply on-site works, candidates were
In 2011, shelter was identified routes, prefabrication facilities and asked questions on minimum foun-
as an urgent need in the refugee clear procedures for managing dation depths and how to best nail
camps. The number of new arrivals supply and construction. a joint. They were then assigned
peaked with an average of 168 one shelter to prove their skills.
Workshops Staff monitored the construction.
persons per day in June 2011. They
Each organisation established
were provided with tents. However, The ground at the different sites
a workshop and materials storage
the life span of the tents proved to varied. In some sites it was relatively
area close to construction sites. In
be around 6 to 8 months, menaing straightforward to dig 60cm deep
the workshops, timber was precut,
that alternative solutions for the holes by hand, in other sites the
bamboo was split, and doors,
shelter in the camps were required. ground was hard and concrete was
windows and roof trusses were pre-
fabricated. required in the foundations.
28
Three different prototype shelters were built and a group of refugee representatives, the government and the key organisa-
tions agreed on a common design
Image: NRC
Shelter selection communities. Each shelter required Logistics and supply

In September 2011, the organi- slightly more than 2m3 of soil to Although highland Ethiopia has
sations agreed to develop common render it with mud, as well as a significant plantations and pro-
shelter standards and build shelter significant volume of water. This duction of both eucalyptus timber
prototypes for review. worked out at over 2,000 truck and bamboo, the nearest eucalyp-
loads for the 10,000 shelters that tus and bamboo plantations are at
At this time, the three or-
were planned in the first year. Up to least a day’s drive from Dollo Ado.
ganisations involved in the shelter
1m3 of mud would additionally be The suppliers who can produce the
programme were invited to produce
required each year for repairs after paper work required for large pro-
prototypes based on the shelters
the rains. curements are further away, mostly
that they had been building. Each
By the end of 2012 there were based in Addis Abbaba.
of the three shelters was built to the
same design brief. sufficient resources available to The transport requirements
implement mud walling for 60 per proved demanding. One truck only
Each shelter was evaluated by a
cent of the shelters in Dollo Ado, carried enough materials for 15
gender balanced group of refugee
and the decision was made to dis- shelters. Building 10,000 shelters
representatives, the government
continue the use of mud in the would require over 600 trucks.
and the key organisations.
future. The design was adapted
Over the year, the biggest cost
The model selected had a cor- using closer spacings between the
increases were with bamboo and
rugated iron sheet roof, a eucalyp- bamboo strips for walls instead of
transportation. This led to a 16 per
tus post-and-beam structure and rendering it with mud.
cent increase in the cost of a shelter.
split bamboo wall cladding. The
For bamboo-only walls, plastic
intention was to plaster the walls Materials list
sheets or fabric were often placed
with mud.
inside to provide protection from Materials Quantity
The shelter had an internal the rain and the wind. The resulting Corrugated galvanised 24 pieces
partition, two lockable windows, structure was relatively well ventilat- iron sheet sheets roofing
(2m x 0,90m)
and a door that could be locked ed in the hot climate and provided Eucalyptus poles (8cm diameter) 32 pieces
both from the inside and the adequate protection from the rain. Bamboo (6cm diamter, min.
outside for improved security. Cor- Households were relatively satisfied 6.5m, dry, straight) 62 pieces
Nails (#9, #8, #6, #4) 10.5kg
rugated iron sheet was chosen for with these shelters. Roofing nails 3kg
roofing on account of its durability Metal straps (2cm wide; 1.5- 10m
Tighter construction quality 2mm thick)
and fire safety.
controls were required for bamboo- Wire mesh (1.8m x 30m; 2cm 1 piece
only walls to ensure that no large opening)
Mud plastering Hinges (T hinge 4 cm long sides) 6pieces
The shelter was originally gaps were left between the lathes. Lock system 4 pieces
designed to have wattle and daub Black wire (10 kg rolls) 0.1roll.
The use of local wood for shelter
walls using local mud. Bamboo laths construction was a major issue Workshop tools
would be covered with chicken wire for the host population, and as a Materials Quantity
and the shelter would be rendered result, timber was brought in to the Electric Radial arm saw 2 pieces
with mud. This was initially consid- area. However, each camp resident Hammer 5 pieces
ered to be a low cost and sustain- Tape measurer 4 pieces
burned a significantly greater Cutting table 2 pieces
able walling solution. volume of wood when cooking Assembling table 3 pieces
than each shelter used in its con- Oil barrel for treating timber 1 piece
Unfortunately, mud of suitable
quality was only available from struction.
certain locations in river beds and
these were owned by the host
30 Top: View of a camp before construction.

Each organisation established a large workshop in each camp to store materials and prefabricate components.
Images: Joseph Ashmore
A.10 Haiti – 2010 – Earthquake

Keywords: Returns, Unplanned camps, Planned and managed camps, Urban neighbourhoods, T-
Overview: shelter, Rental support, Housing repair and retrofitting, Cash / vouchers, Mass communications.
Summary
In October 2010, ten months after the Haitian earthquake, a
humanitarian organisation began a project to close a small camp of
around 200 families. Families were given rental support cash grants
to cover the costs of renting accommodation for one year and to
support the transition from camps to their new accommodation. The
project succeeded in its aims and became a test case for a much wider
programme of rental support.
Promoted by a small number of organisations, the rental support
approach relied on donors’ willingness to take a risk on a project-type
with few precedents. By mid 2011, rental support cash grants had
become a key part of the return strategy and by November 2012 over
23,000 households had received grants.
Early indications are that rental support cash grants have been
successful. A survey of households that have completed their year
of rental subsidy found that all of the respondents (90% of the
total caseload) had been able to organise their own housing for
the foreseeable future. None had returned to camps or moved to
informal settlements.
Le Marron Inconnu (the unknown slave)statue surrounded The statue in April 2012 after the Champ de Mars camp
by shelters in the Champ de Mars camp in early 2012. closure.
Photo: IOM Photo: IOM
Background Permanent housing plots in the managed camps were

The Haiti earthquake of January reconstruction: Rebuilding taken by others moving in from
2010 caused massive loss of life and irreperably damaged houses. spontaneous settlements.
damaged or destroyed 180,000 The lack of buildable space in Camps were not only bad for
houses. (See Section A.4 Shelter densely-populated urban areas the displaced people but they also
Projects 2010 for more background and complex issues over land rights prevented occupied public spaces
on the Haiti response) meant that the three main responses from being rehabilitated.
Responses generally took one of would only benefit those with land
rights or those who owned houses. In this context some Haitian
three forms following the distribu- officials began suggesting that
tion of non-food items in the initial Those displaced in camps over- displaced people should be paid
emergency phase: whelmingly did not own either land to leave camps. These proposals
T-shelters: This was the main or housing before the earthquake. were dropped due to protection
response by many organisations. Consequently, only a quarter of T- concerns as it would be impossible
Transitional shelters (T-shelters) shelters built went to Haitians who to verify if the families had found a
were built using basic frames were living in camps. Not only did durable solution. However, interest
which could later be adapted this mean that camp populations in properly planned rental support
into more permanent structures. were being reduced at a slow rate cash grants grew and presentations
Yellow House repair: Buildings but it proved almost impossible to were made to donors to encourage
were assessed by engineers close camps completely. If only a adopting the approach.
and classified as Green (safe), small proportion of a camp had a
Yellow (to be repaired) or Red durable solution available for them
(to be demolished). it wasn’t long before the empty
An edition of Chimen Lakay, a graphic newspaper, featuring the 16/6 program and a newly cleared market place.
Designed by: Chevelin Pierre, Script: Mike Charles.
Neighbourhood approach ing neighbourhood businesses to Protection and assistance:

Rental support was closely start-up or expand in order to offer Identification of vulnerable
combined with the neighbourhood those returning real income genera- families who qualify for
approach to reconstruction. tion opportunities. The grants were additional help
available to anyone with a business Beneficiary communication:
The concept of the neighbour-
idea and not restricted to those Facilitation of informed choices
hood approach is that projects
returning from camps. by beneficiaries using wide
such as rubble clearance, rebuild-
range of multi-media and face-
ing, water, sanitation and liveli- The 16/6 programme relied
to-face communications
hoods programming should be heavily on the use of rental support
Choosing a housing option:
joined together across sectors and cash grants to offer all families living
Either T-Shelter, Yellow-house
that agencies create a coordinated in camps a realistic housing option.
repair or rental support cash
and efficient response support-
Rental support grant
ing families to move from camp to
Rental support projects differed Choosing a rental property:
community. As of December 2012,
between agencies but largely Family chooses a property
this goal had not been fully realized,
followed the same pattern: (independently assessed for
but efforts were being made to take
safety) and negotiates the rent
a more holistic approach. Registration: Emphasis on Cash grant transferred: The
This approach minimises the obtaining accurate beneficiary year’s rental cost of US$ 500
possibility of families “rebound- lists through other health is transferred directly to the
ing” back into camps. For example, or distribution activities, in landlord and the family receives
“rebounding” could be caused by collaboration with Haitian the money left over
a lack of employment opportuni- authorities
ties or extremely poor sanitation
standards in the neighbourhoods to
which people return.
The 16/6 program

The 16/6 program, led by the
Haitian government, targeted
income regeneration in sixteen
neighbourhoods coupled with the
closure of six camps.
The programme focus on
neighbourhoods meant that liveli-
hoods grants were not allocated to
families leaving the camps. Instead,
a targeted livelihoods program was Graph to show completed and planned housing solutions, November 2012
implemented, aimed at support- Source : E-Shelter and CCCM Cluster
32
Camp closure: Families are

given a US$ 25 cash grant
to help in transporting their
possessions to their new home
Surprise visit: Agency awards
a US$ 125 bonus to families
continuing to live in their
chosen rental accommodation
following a surprise visit made a
few months later.
In addition to the US$ 650
grant costs, the relocation of one
household incurred an additional
US$ 350 in programming costs,
making a total cost of the return of
one household rise to around US$
1,000. Programming costs include
beneficiary registration, commu-
The ruined National Palace surrounded by shelters
nication of activities and protec- before the Champ de Mars camp closure.
tion activities such as providing Photo: IOM
two-years rental for vulnerable
families.
In comparison, a T-shelter costs
around US$ 2,000 and a permanent
house around US$ 6,000.
Concerns and safeguards

There have been vigorous dis-
cussions around the appropriate-
ness of a rental support approach
as a durable solution.
Some of the key concerns and
corresponding safeguards were:
Cash distributions can act The ruined National Palace after the Champ de Mars camp closure.
as a pull-factor to camps: Photo: IOM
Announcements about rental
A survey of households who Lessons
support programs were made
rented for one year achieved a Rental support could have been
publicly only after accurate
90 per cent response rate. Out of implemented earlier if it had
beneficiary lists were made.
those households responding, been considered or picked up by
Negligible pull-factors were
all had found their own housing other donors.
noted.
solutions and none had been Better links to livelihoods
Rental properties may not
forced to return to camps or programmes could be made
meet minimum standards: All
informal settlements. to further support families to
rental properties were assessed
Nearly 100 per cent of continue to cover rental costs
for safety and sanitation issues.
respondents reported that themselves in the future.
The emphasis was therefore on
their situation is better or much The neighbourhood approach
moving people out of the much
better than it was in camps. offers more chances for better
worse conditions in camps.
77 per cent of landlords used coordination between sectors
Cash grants would inflate
two-thirds of the rent money and organisaitons as well
rents: Rents were monitored by
to improve the standards of as between emergency and
organisations using the prices
the properties that they were development actors.
agreed between families and
renting out. The approach has been popular
landlords. Rents had not risen
by the end of 2012. with the general public,
particularly as it emphasises
Indicators beneficiaries’ rights to actively
The rental support approach choose where to live. Haitian
shows the following early indicators politicians have been keen to
of success: promote and be involved in
rental support programs.

Keywords: Returns, Unplanned camps, Urban neighbourhoods, T-shelter, Rental support, Housing
Case Study: repair, Cash, Training.

Haiti
Project location:
Port au Prince
Disaster/ conflict: 33 months – – Project completion
Earthquake
Disaster/ conflict date:
January 2010
Total number of houses
damaged or destroyed:
180,000
1,208 families relocated from 5
IDP camps
10,518 T-shelters built with
services and support 22 months – – Project start
Occupancy rate on handover: 21 months – – Project agreed
95 per cent
Project cost per household:
Approximately US$ 990 / family
T- shelter programme costs Haiti
were higher
Port au
Prince
January 2010 – – Earthquake
Project description
The project offered several service packages, including rental assistance, transitional shelter construction and
repairs to damaged homes, to incentivise families to leave camps and find suitable housing solutions. Central to
this project were life skills training, household livelihood planning, temporary health insurance and psychosocial
services. Over one year, the project closed all five camps that were targeted and helped more than 1,200 families
resettle.
Strengths and weaknesses service to families.

9 The organisation fully achieved its target of closing 8 Early beneficiaries could have been better prepared
five camps. for the risks of receiving cash.
9 Life-skills training was delivered in time to prepare 8 Staff safety was a serious concern due to the
families for their resettlement. challenging sites chosen (site selection based on level
9 Effective mechanisms to prevent fraud and to of need rather than ease of implementation).
minimise inflation of rent prices. 8 With a large part of camp residents making a living
9 Participants were given a choice in their resettlement from small commerce, there was an opportunity to
option. work with IDPs before they relocated to improve their
9 Good mechanisms for preventing programme abuse small commerce activities. This component was not
and to reduce the chance of housing price inflation. added until the second phase of the project.
9 Use of mechanisms such as a hotline improved the - The rental subsidy lasted one year which gave the
organisations accountability. family time to save and plan for their future and recover
9 Strong support from local government. from living in a camp for two years.
8 Some landlords canceled contracts due to the - Despite early sceptisicm from many humanitarian
organisation making late payments. staff, rental subsidy programmes did not noticably lead
8 Personnel costs were comparatively high due to the to rental price inflation.
large number of staff required to provide a personal
34
The project offered various options, including cash for rent, and provided additional support to help close camps.
Photo: Jack Reybold (CRS)
Background given photo ID cards to prevent Most families moved to houses

(See the overview section A.10, further disputes. in neighbourhoods near the camps,
Haiti – 2010.) while others moved to areas with
Coordination more open spaces such as Carrefour
The organisation created the In line with government and Croix des Bouquets.
Ann Ale Lakay project (meaning strategy, the project offered a
“Let’s go home” in Haitian Creole) standardized package of resettle- The rental subsidy was worth
in September 2011. The project ment options. The goal was for all US$ 500 and if the family could
aimed to support families remaining agencies engaging in camp closure negotiate a lower price with a
in camps to leave those camps. The projects to operate using a standard landlord they were able to “keep
project was a response to the fact approach, as this prevented families the change”. For example, if they
that pull factors alone (the setting from refusing the services of one or- find a place for US$ 400, they could
up of services in return neighbour- ganisation in the hope of receiving keep US$ 100. This helped prevent
hoods) were not sufficient to get a better deal from another. rent price inflation as project partic-
families to relocate. ipants had an incentive to negotiate
the best deal possible
The project began as a 6-month, “With the conflict resolution
US$ 600,000 pilot project to close [training], I can find Although the housing market
three camps (460 families). It was solutions to all my other in Haiti had not even begun to
extended for six more months to recover, it was flexible enough to
problems”
close an additional two camps. absorb more people capable of
paying rent.
The project was aligned with Project beneficiary
the Haitian government’s “16/6” Around 60 per cent of the
programme. This programme aimed people receiving rental subsidies
at closing six camps and rehabilitat- Implementation found accommodation for less that
ing sixteen return neighbourhoods. The project offered households US$ 500. The median rental price
three choices: one year’s rental was US$ 375. In informal inter-
Selection subsidy, construction of a transi- views, the majority of beneficiaries
The five camps were chosen in tional shelter or support to repair a reported using the remaining funds
coordination with other actors and damaged house. from the rental subsidy to support
had been identified as priority sites their activities in small commerce,
for closure. Some camp dwellers Of the 1,205 families supported,
98 per cent chose to take the such as purchasing a small quantity
were “renting” tents from those of goods for resale.
who have moved out of the camp. one-year rental subsidy and 2 per
The organisation ensured that cent chose to receive a t-shelter. There was a risk that people
these families, rather than the tent As few of the families living would be harassed and pressured
“owners”, received project services in the camps had owned a house to give the money to groups such as
by conducting a “surprise census” before the earthquake, there the police and the camp committee.
to ensure that the genuine residents were no housing repair services As a result the money was trans-
were registered. Beneficiaries were requested. ferred directly to the landlord
Life skills trainings were a an essential component of the Camps were slowly turning into permanent shanty towns and
project, preparing residents for life outside the camps. many had poor sanitary conditions.
Photos: Nathan Jayne Photos: Nathan Jayne
via a money transfer service. The organisation’s Code of Conduct contracts greater legal weight
remainder was transferred to the field teams based within the in favour of the family. The aim
head of the household via a mobile camps was that this would help to
phone base transfer system. an official, organisation-wide prevent evictions and reinforce
accountability framework. the government's leadership in
Some tent “owners” pressured
this project.
the “renters” to share the leftover Protection
money from the rental. In later Protection issues were dealt Trainings
phases of the project, it was with in different ways: Cash transfers were accompa-
suggested that everyone should nied by a life-skills training module.
Training included a family
keep information to themselves According to participant interviews
communication and conflict-
on whether or not they received and focus groups, this training was
management module, which
leftover money. It was suggested crucial to the success of the project.
focused on positive ways to solve
that renters immediately moved out
problems (including disciplining These six-module trainings
of the camp, and that they should
children) without resorting to prepared camp residents with life
not share their address with the
physical punishment. skills they would need for a suc-
tent “owner”.
Social workers checked the cessful transition. To help people
The project team had 19 people: safety and adequacy of all develop a sense of responsibility for
an international programme houses before families moved their future, families developed a
manager, a project manager, two in. The families who chose to “family plan”, a personal road map
project officers, a monitoring and receive rental support were for resettlement and recovery. The
evaluation officer, two psycholo- not allowed to choose houses plan not only helped families think
gists and twelve social workers. marked as damaged, nor homes about their goals for the future but
located in or along a ravine. also helped them plan for potential
Accountability The municipality signed all setbacks.
The organisation took some rental agreements to give the
specific steps to provide accounta-
bility to service recipients, including
setting up:
notice boards in all communities Summary of Training modules
with information about the Module Details
organisation and the project Family communication Interpersonal skill development and conflict
a free telephone hotline to deal resolution
with any questions Personal responsibility and problem solving Role within the country of Haiti, larger
community, neighborhood and family
regular community meetings –
Prioritising needs, planning for the future Helps families to identify needs and create a
for information dissemination family plan.
and feedback to the organisation Financial planning Banks, savings, lending options, health and
contracts with Beneficiaries, other insurance, negotiation
outlining mutual responsibilities Small business management Key concepts to improve the profitability of a
posters and trainings for small business
all project staff on the
36

Keywords: Dispersed, Construction materials, Housing repair and retrofitting, Training, Guidelines and
Case Study: training materials .

Haiti 31 months – – Phase 2 starts
Project location:
Rural south-eastern Haiti
Disaster: 29 months – – Completion of 300
Earthquake houses
Disaster date:
12th January 2010
22 months – – Phase 1: 28 houses
Number of houses damaged / repaired
destroyed:
180,000
17 months – – Repair project starts
Project outputs:
500 completed houses
More than 90 per cent
Shelter size:
22 m2 reconstructed houses – Starting of the re-
construction project
22 - 42 m2 repaired houses
7 months – – First prototype for
Materials cost per household: repair
US$ 3,190 (Including US$ 740
local contribution) Haiti 5 months – – Feasibility study,
local assessment
US$ 1,000 (including US$ 300
4 months – – First prototype for
local contribution) for repairs reconstruction
Petit
Project cost per household: Goave
US$ 4,000 reconstruction 1 month – – Partners request
support
US4 2,000 for repairs
– Pilot phase started
12 January
2010 – – Disaster date
Project description
This project worked in rural areas of Haiti beginning with an in-depth assessment of local building practices.
Builders were then trained in improvements to existing construction. This was followed by building assessment
and repair construction programme resulting in the construction of 500 houses to date. The overall project
goal was to improve local communities’ resilience to hazards and to improve living conditions through housing
improvements and construction-based economic stimulus.
Strengths and weaknesses housholds which were not provided with construction
9 The project was designed to be replicable by Haitians support.
without external support. 8 There is a low visibility of improvements as they are
9 A detailed assessment of cultural practices meant difficult to identify by a non-professional.
that social structures were enhanced instead of ignored 8 It was difficult to persuade local partner
by the project. organisations to repair more houses as they considered
9 Good ownership by local stakeholders. repaired houses to be less safe than new houses.
9 The project strengthened the capacities of existing 8 Technical, management and administrative
local organisations and created jobs linked to local capacities of partner organisations were not properly
market. assessed.
9 Construction skills training enhanced livelihoods - This project is ongoing and has received some
opportunities and has improved the general safety of interest from other organisations following positive
construction. impacts on other projects and national strategies.
8 Detailed assessment of local capacities meant that - The Ministry of Public Work, Transport and
the construction phase started relatively late. Communications gave its agreement for the use of the
8 Slow to demonstrate impacts. There was no designs and technical recommendations for housing
significant impact in the first years of the project on reconstruction in Haiti.
The project approach was adopted by other organisations. Left: shelter built by the project, Right: An adaptation by another
organisation in an urban area of Port au Prince.
Photo: Left: PADED / MISEREOR Right: CRAterre.
Before the earthquake Lists of affected people were Approach

(See the overview section A.10, drawn up by the local organisations Affected people were involved
Haiti – 2010) immediately after the earthquake. as much as possible, and five
A community meeting at the start partner organisations implemented
In many of the rural communi-
of the project was attended by 200 the project.
ties in south-eastern Haiti incomes
people from all the project areas,
are low and there is no access to The following steps were
and the following selection criteria
power or running water. The public followed:
were decided upon:
infrastructure that existed was in a
Local organisations defined and
poor state of repair. Compulsory:
managed the reconstruction
Most people in the region The house of the beneficiary was projects.
owned their own houses, grouped damaged by the earthquake. Specific designs and technical
or dispersed over a large territory. The beneficiary is the owner of solutions were developed
Many houses were in a poor the house plot. depending upon the context.
condition, and homeowners The beneficiary agrees to the An external expert was
often lacked the knowledge and rules of the project. embedded in each local
resources to maintain them. Regular Preferred: organisation for one month to
damage was caused by cyclones. build up their training capacity.
The household hosts displaced Building models were monitored
After the earthquake families. and evaluated. If necessary,
In south-eastern Haiti, more The household head is female. changes and adaptations were
than 50 per cent of rural houses The household head is a widow. made.
were partially damaged by the The household includes many
earthquake. However, very few children and the adults have Implementation
people were injured or killed by limited income-generating Households were put into
building collapse As the affects of opportunities. groups of 5 or 6 households. These
the earthquake were relatively less Households are committed groups had to work together to
severe in rural areas compared to members of the local repair their houses.
urban areas, there was a migration organisation (this was a Existing administrative staff
to rural areas immediately after the condition of the partner from partner organisations worked
earthquake. organisations). on the project. A social mobiliser
The beneficiary is regarded was hired to assess up to 50 house-
Selection of beneficiaries as having a good behavioural holds. Two engineers were hired
Project areas were selected record. per partner organisation.
according to level of damage and
Land issues were resolved by the
whether partner organisations had
local partner organisation.
a presence before the earthquake.
House owners bore part of the responsibility for monitoring on-site construction.
Photo: PADED / MISEREOR
38
There were two monitoring and

evaluation missions each year, and
the project was managed by a full
time foreign expert based in Haiti.
During the house repairs, the
inhabitants were given a guided
tour of a damaged house to point
out defects and reasons for failure.
With this new knowledge, people
were able to take on part of the
responsibility for the quality of con-
struction and repairs to their own
houses. The project included repairs and had a strong social mobilisation component.
A registration card was Photo: PAPDA/ VEDEK / Secours Catholique.
completed for each household. To resist high winds, houses Logistics
This included: identification were built with a low profile, and Each partner organisation
details; reason for their selection; households were encouraged to procured construction materials
ownership of the land; access to grow high vegetation surrounding from local suppliers, though these
water; proposed repairs; beneficiary the house to reduce potential suppliers imported part of their
contributions to the shelter and impacts of cyclones. materials.
construction completion dates. An
agreement was then signed with Houses were built with four In some cases the partner or-
the householder. roof slopes to prevent there being ganisations formed a procurement
a weaker gable end. In some areas, collective in order to negotiate
Households selected a builder, people preferred a traditional roofs better prices.
paid for by the organisation, from a design with two slopes as they
list of craftsman who had completed could use the space under the roofs Broader impacts
the training programme. Local site for storage. Most of the newly built houses
supervisors made technical checks in the project area that were not
on each building. Training funded by this project had small
New houses were constructed in The project involved three stages improvements to bracing, stone
groups, while repairs were made on of training: a training of trainers, masonry, and stone foundations.
a house by house basis. a training of artisans and a more Although it is too early to really un-
basic training for house owners. derstand the broader impact of this
Technical solutions project, it is hoped that it has led
As many of the families were Participants were trained on the to a change in construction culture.
poor, technical shelter solutions different ways hazards can affect
buildings. Other organisations have
had to ensure low maintenance
adopted this project approach and
costs. As part of the repairs are conducting their own trainings
The core technical criteria was programme, each household group in other areas.
that shelter failure would not lead was given training on water and
to further injury and death. sanitation issues and provided with Materials list
a community water tank. Materials Quantity
Traditional local houses were
Trainings materials included Repairs (for 100 houses)
built on wooden posts dug directly
printed illustrations of best practice Corrugated iron sheet (34 2,000
into the ground which were quickly guage)
weakened by rot. The new design in Créole.
Cement Bag 1,500
added a proper foundation. Artisans were trained in groups
Local wooden pole 1,500
Masonry skills were very basic in of 20 for 2 to 4 weeks, during
which they constructed a prototype Roofing nails 100 lbs
the area and filling this knowledge
house. Payment for participants to Reconstruction (for 100 houses)
gap was an important part of the
attend trainings depended upon Corrugated iron sheet (34 3,000
construction training. guage)
the partner organisation. In some
Cross-bracing was used in the case, only food was provided, in Cement Bag 1,100
walls. This reduced the risk of the other case, full salaries were paid. Wooden rafter imported 4,600
wall collapsing in cyclones and
Wooden plank imported 1,500
earthquakes.
Roofing nails 700lbs

Keywords: Returns, Unplanned camps, Urban neighbourhoods, Infrastructure, Community
Case study: engagement

Haiti
Project location:
Port au Prince 24 months – – Construction of
300m canal
Disaster/ conflict: – Construction of
Earthquake market place
Disaster/ conflict date: 23 months – – Presentation of
outcomes
12 January 2010 22 months – – Community action
Number of houses damaged / plans
destroyed:
21 months – – PASSA Process
180,000
Project target population: 18 months – – Community sensiti-
700 families sation
17 months – – Training project
Project outputs: team on PASSA
Increased awareness on safer
construction, 15 months – – Relocation of most
Cadastral map, at-risk shelters
Community Action Plan, 14 months – – Neighbourhood
assessment
300m of canals 10 months – – Recovery phase
Community Market Place starts
Haiti
8 months – – Livelihoods grants
7 months – – Return starts
Port au
Prince 5 months – – Assessment
– Project start
12 January
2010 – – Earthquake
Project description
The organisation used the Participatory Approach for Safe Shelter Awareness (PASSA) process to support the
community make the transition to neighbourhood recovery. A range of participatory activities were carried out
to decide both a comprehensive community plan for reconstruction, and a detailed list of related programme
activities by the organisation. The identification of problems and solutions enabled the community to make plans
for their own long-term recovery activities.
Strengths and weaknesses technical skills are required for safer construction.
9 A participatory planning approach promoted a 8 More time was needed to explain that participatory
high level of engagement by the community which tools only informed planning, and expectations for
led to a programme that responded to people's self- concrete results needed to be managed.
determined needs. - PASSA was developed in rural contexts, the focus
9 The process empowered and gave a voice to on ‘shelter’ needed to be adapted to ‘habitat’ to
members of the community who are not often heard. encompass the infrastructural and social aspects of
9 The plans that were developed cut across a number living in an urban context.
of different sectors which resulted in an integrated - "PASSA" can be carried out simultanously with
approach to settlement planning. other assessment techniques.
9 The project built on relations with camp residents - GIS mapping weas essential to monitor progress.
early in the response to support recovery. - Considerable time is required to plan the participatory
9 Enabled the community to directly act in their process and analyse the information from workshops.
neighbourhood to improve their quality of life. - Local terms needed to be used to ensure a full
8 PASSA was not used in the first year of the response understanding of issues.
leading to delays in the recovery planning. - Participatory tools developed for rural contexts can
8 Participatory tools are only the first step for be adapted for urban contexts.
reconstruction. Additional training, planning and
40
People returned to crowded areas The organisation used a PASSA process: residents identified their own
with limited services. problems and the actions needed to address them.
Photo: Amelia Rule Photo: Mandy George
Before the earthquake molished houses, makeshift shelters not include the existing committee
After land was reclaimed from and a lack of services. members. However, all activities
the marshes in the 1980s, an were carried out in coordination
The groups with the highest
informal settlement developed in with the committee members.
shelter vulnerability were renters
Delmas 19, Port au Prince. The
and those who lived next to PASSA comprised eight partici-
houses were self-built structures
the canal on land that could be patory activities, which were carried
made with poor-quality materials
reclaimed by local authorities. out over two to three months:
such as concrete blocks, corrugat-
Those facing possible eviction had
ed iron and wood, and construct- 1.historical profile and everyday
a broad range of backgrounds in
ed with little knowledge of safe problems
terms of education levels, livelihood
building techniques. 2.community mapping and visit
strategies and home ownership.
3.frequency and impact of
Infrastructure was poor with
Direct support was given to hazards
limited water and sanitation
specific households based on vul- 4.safe and unsafe habitat
services, and the site was badly
nerability assessments developed 5.options for solutions
drained with limited access.
with the community, while the 6.planning for change
After the earthquake whole community benefited from 7.problem box (future planning)
The earthquake destroyed half improvements to site drainage and 8.monitoring plan (future
of the houses in the settlement public spaces such as the market. planning)
and damaged half of the remaining After each activity, the group
Implementation
structures. The main drainage canal shared their work with family and
The participatory process began
was also damaged and blocked by neighbours to encourage under-
with an explanation to participants
rubble and debris. standing of the process across the
of how a detailed planning process
community.
Many water reservoirs belonging would result in the best solutions
to individual households and com- for reconstruction. The coordina- At the end of the process, all
mercial suppliers were damaged tion of different sectoral projects, the work, findings and plans were
and pit latrines were inaccessible such as solving drainage issues shared firstly with the committee
or broken. There were more than before providing shelter solutions , members for feedback and input,
100 families, with only one public achieved a joint approach to settle- and secondly presented to the
latrine, living in makeshift shelters. ment rehabilitation. whole community at an open day
held in the community centre. The
Selection of beneficiaries The organisation used the “Par-
PASSA group members shared
Following the earthquake the ticipatory Approach for Safe
what they had done and received
organisation provided emergency Shelter Awareness” (or PASSA
their participation certificates.
assistance in the targeted camp, see - PASSA, Participatory Approach
and identified the clear need for for Safe Shelter Awareness, IFRC The main problems faced by the
joint livelihoods and shelter support. 2011). PASSA was a relatively new, community were:
and formally structured approach
In June 2010, the private weak infrastructure and
to participation in shelter projects.
landowner offered US$ 200 to flooding
It was based on a tool commonly
families to leave the site. Conse- public health, water, sanitation
used in WASH programming.
quently two-thirds of the camp and waste management issues
population relocated. The majority The PASSA process involves safe access routes and personal
were from the adjoining neighbour- working with a group of 40 rep- safety
hood, and the organisation followed resentative people. This group was unsafe shelter and settlement.
them as they returned home to de- selected by the community and did
Both the market (left) and the path (right) were identified needs and both were
built by residents with the support of the organisation.
Photo: Amelia Rule
The identified solutions were to: ment with relevant authorities. The Poor waste management and
projects also aimed to improve skills lack of upkeep of the canal lead to
construct the canal
for the housing construction and serious blockages and subsequent
install solar street lighting
repairs which would follow. flooding of low-lying houses with
construct shared latrines
waste and sewage.
improve waste management Challenges with PASSA
improve housing and planning The community had raised ex- To mitigate against these
improve technical expertise pectations about what PASSA could problems the PASSA process helped
through supervision and provide. They thought they would participants to identify simple
training. immediately receive the solutions actions that they could conduct.
they identified. The facilitators These included improved construc-
Community projects
spent a lot of time explaining that tion and environmental manage-
Planning for change started
the participatory approach would ment, and how to prepare, plan
with mapping the issues in the
help to identify priorities and the and respond to a natural disaster.
neighbourhood and understand-
ing their relationships. This enabled solutions that the community them-
Technical solutions
the community to take into account selves could achieve. It would also
When provided with the
issues, including gender, protec- analyse where support was needed
materials and technical support
tion and security. Once the issues from the organisation and the local
necessary to carry out the recon-
had been identified the groups authorities.
struction the PASSA process had
discussed each problem in turn. The PASSA tool was developed ensured that the community was
Working groups, called ‘cells’, in a rural context with a specific highly motivated.
took on each subject and carried focus on ‘Shelter’. As a result, some
At the end of 2012, Haiti had no
out further work, before creating limitations were found using the
official building codes and material
an overall Plan of Action. tool in an urban context and within
standards were not enforced.
an integrated approach. The team
A security cell positioned solar The general level of understand-
adapted the activities to take into
lighting while a community waste ing by architects and builders of
account the wider issues of infra-
management group cleared waste. seismic construction techniques
structure, water sanitation, urban
was limited. A great deal of time
Community contracts were issues such as spatial planning and
was spent with engineers, seismic
written for people from the neigh- security problems.
specialists and construction profes-
bourhood to build the canal. This sionals to ensure that the shelter
DRR components
employed over 300 people. solutions were safe and that the
The area was suffering from
Materials and technical supervi- poor drainage, poor waste man- community understood the reason
sion were provided by the organi- agement, poor housing construc- behind the application of new tech-
sation and fifteen shared latrines tion and poor infrastructure. All niques.
were constructed by the families these aspects made the population This knowledge was trans-
themselves. vulnerable to flooding, the effects ferred outside of the participatory
of hurricanes, outbreaks of disease planning sessions, delivered instead
Community construction teams
and earthquake risks. through on-site practical training
that had received training before
working on the canal also built the PASSA raised understanding of sessions.
market. how risks to health and safety were
caused not only by natural disasters
All of these activities started
but also by the everyday practices
with awareness raising and engage-
of the community.
42
Participatory Approach for Safe Shelter Awareness (PASSA) is a

participatory method of disaster risk reduction (DRR) related to shelter
safety. It is a variation of Participatory Hygiene and Sanitation Trans-
formation (PHAST), which has been used by many Red Cross Red
Crescent National Societies in water and sanitation programmes since
the late 1990s.
The aim of PASSA is to develop local capacity to reduce shelter
related risk by raising awareness and developing skills in joint analysis,
learning and decision-making at community level.
PASSA is a process, facilitated by volunteers, that guides community
groups (called PASSA groups in this manual) through eight participa-
tory activities which enable the participants to do the following pro-
gressively:
Develop their awareness of shelter safety issues in their community
Identify hazards and vulnerabilities that create risk related to
shelter
Recognize and analyse causes of shelter vulnerability ÌÌIdentify and
prioritize potential strategies to improve shelter safety
Make a plan to put those shelter safety strategies into place,
based on local capacities
Monitor and evaluate progress.
Source PASSA, Participatory Approach for Safe Shelter Awareness,

IFRC 2011
“PASSA helped us to see Defining the community:

that many problems in our
In this complex urban
area are not complicated
context, the community
to fix, they are small things
was defined by: housing
that can have a large
typologies, level of poverty,
negative impact – such
physical boundaries of
as the rubbish blocking
roads (making the area a
the canal and causing
pedestrian community), a
flooding.”
representative committee
and the familial and
PASSA participant neighbourly networks that
Delmas 19 were already in place.
Drainage was identified as a key safety issue. 300m of drains were cleared and covered to make a path.
Photos: Amelia Rule
A.14 Japan – 2011 – Earthquake and Tsunami

Keywords: Urban neighbourhoods, Housing repair and retrofitting, Cash, Structural assessment.
Update:

Japan
Project location:
Ofunato, Iwate 21 months – – Project completed
– Housing payments
Disaster: completed
Earthquake and Tsunami – Housing repairs
Disaster date: completed
11th March, 2011
19 months – – Investigations of
Number of people displaced: housing completed
390,000 houses total
5,500 houses in Ofunato
Project outputs:
150 houses repaired
Advice provided to 1,155 11 months– – Applications closed
households
8 months – – Assessments com-
Occupancy rate on handover: plete
Unknown
Shelter size:
Variable 6 months – – Pilot project com-
plete
Up to US$ 5,600 Ofunato
Japan
3 months – – Project start date
11 March – Tsunami
2011 – – Earthquake
Project description
This project provided cash assistance to repair 150 houses after the tsunami in Japan. It was mainly targeted
at families unable to apply for the government’s Emergency Repair Aid and for those who required further
assistance on top of the government’s aid package. The project provided an information and support centre with
outreach to support 1155 households. This service provided information to those who had difficulty in accessing
other sources of information, primarily the elderly or people living alone.
Strengths and weaknesses the same time the affected people were thouroughly
9 The project hired architectural specialists to provide consulted to gather information and understand needs.
advice to rebuild durable housing. 8 By supporting rebuilding in potentially hazardous
9 Affected people were supported to return to areas, there was the possibility that people could be
their original homes, and to reintegrate with their re-exposed to future disasters.
community.This helped to improve security in the area. 8 A limited number of households were supported as
9 Local construction firms used local labour and more a result of the high cost per family (a result of Japanese
environmentally sustainable resources. materials prices and culture).
9 Houses of both evacuated and non-evacuated 8 The financial aid was only for repair and was not for
people were rebuilt. the construction of new houses.
9 The project's focus on preserving parts of the old 8 Only families who had reached a certain stage in
town helped to give a feeling of continuity to the their recovery could benefit from the repair project.
community. - Vulnerable households were prioritized, including
9 By coordinating with other organisations, the project those made up of only elderly people or those with very
was part of a sustainable, sector-wide response. At low incomes.
44
Community consultation meeting . the organisation established an information centre and provided advice through indi-
vidual consultations to 1,155 households.
Photo: Habitat for Humanity Japan
Before the earthquake rented private houses and apart- many families in Iwate prefecture
Japan is a high income country ments for affected people. and Ofunato city were still unable
with nearly 70 per cent of its popu- to repair their homes.
lation living in urban centres. The The government provided grants
of US$ 6,000 to families who were The organisation selected house-
affected area was known for its holds based on financial need, the
deeply indented coastline, fishing able to return home but lacked
the financial resources to carry out relative impacts that repairs would
and marine farming. have on the family’s quality of
repairs.
The population of the area was life and how much the total cost
ageing and decreasing, while the In Ofunato city the fishery of repairs would be covered by
percentages of detached houses business was devastated, and the available financial aid.
and self-owned houses were high up to 5,500 houses were damaged.
compared to urban areas. The hilly landscape in Iwate prefec- Implementation
ture, this resulted in many houses The organisation provided cash
After the earthquake being damaged but not washed assistance for repairs to houses in
The Great East Japan Earth- away, despite being flooded by the Ofunato city. As households had
quake was the biggest in Japan’s tsunami. to make applications to receive the
recorded history. The earthquake financial aid, information about the
Despite the high levels of
caused a huge tsunami with a wave project was advertised in different
damage, not all home owners
height of over 10m. It also caused media and communicated through
received grants through initial gov-
a temporary rise in sea level of up individual house visits to ensure
ernment schemes. This is because
to 40m. that all residents were aware of the
Iwate prefecture closed its appli-
process.
The tsunami devastated the cation process earlier than other
Pacific Coast of the Tohoku and affected prefectures. The organisation approved 150
Kanto area, cutting off communi- of 270 applications.
In Ofunato city, the government
cation routes. More than 390,000 immediately constructed temporary The works were done by local
houses were damaged or destroyed, housing for 1,800 households as an carpenters selected by beneficiar-
and more than 400,000 people emergency measure. Later on, 150 ies themselves. This approach
were displaced in the immediate unoccupied public housings and respected local Japanese culture
aftermath. 500 unoccupied private apartments and existing relationships in com-
The Fukushima Daiichi nuclear were rented by the government to munities.
plant was also hit by the tsunami, serve as temporary housing.
Each household signed a
causing a nuclear crisis that led to Those that had not been contract with the organisation,
long-term evacuation for those evacuated often had difficulty re- stating that the household and
living nearby. habilitating their houses due to carpenter (rather than the organisa-
The government built 54,000 lacking resources. tion) had joint responsibility for the
temporary housing units. These construction process. The organisa-
were to last until safe permanent Selection of beneficiaries tion paid the repair fees directly to
housing could be built. The govern- Once the allocation of carpenters and monitored the con-
ment also released empty rooms Emergency Repair Aid grants was struction.
from existing public housing and completed it became clear that
The organisation made technical

assessments before and after the
construction, using local, qualified
architects. These architects were
specialised and qualified with
relevant licenses for the works.
Their titles would translate as
“qualified architect of the first
class”, “qualified architect of repair
technique”.
The technical assessments were
used to decide whether a house was
safe to repair and to provide recom-
mendations on how to improve the
disaster resistance of the buildings.
Advice
The organisation provided
recovery advice to 1,155 house-
holds, including the 150 that
received financial assistance. Advice
was provided on housing, financial
issues, house repairs and how to
access government aid.
The project staff conducted the
advisory sessions in the temporary
housing complexes and local
community centers. Staff explained
the aid available for each individual
case of damage.
In some cases external experts
such as lawyers, architects and
financial planners, accompanied
project staff.
The staff provided the residents
with information that they regularly
gathered from the city offices and
Before and after photographs of housing repairs.
updates on various government as- Photo: Habitat for Humanity Japan
sistance programmes.
The project tried to ensure that and that encouraging people to Logistics
people who were less mobile, par- repair their original homes was not Financial aid was directly paid
ticularly elderly people living on reducing their exposure to future to the builders and not to the
their own, had equal access to in- disasters. homeowner. The payments were
formation by making home visits. The organisation tried to obtain made by the organisation’s head-
specific hazard information on each quarters in Tokyo to guarantee
The organisation hired two full secure bank transactions. All of
-time staff to manage the project. location before they visited each
house. the construction firms that the or-
Coordination ganisation worked with had bank
Before repairing a house the accounts.
By coordinating with the city
team checked whether it was in an
administration and NGOs, the The organisation checked in
area defined as a hazardous area
correct up-to-date information was advance a list of appointed carpen-
in the city revival planin which case
delivered to as many households as ters / builders from city offices. This
the building did not take place.
possible. was required to ensure that the
Before rehabilitation, architects organisation did not contribute to
DRR components reviewed each house to ensure anti-social entities.
There was some concern that their suitability for living.
the tsunami-affected area would
remain at high risk of future hazards
46
Before and after photographs of housing repairs.

Photo: Habitat for Humanity Japan
A.15 Kenya (Dadaab) – 2011 – Famine / Conflict

Keywords: Planned and managed camps, Emergency shelter, T-shelter, Core housing, Site plan-
Update: ning, infrastructure,

Kenya
Project location: January 2013 – – Kambioos camp of-
Dadaab ficially recognized
Conflict / disaster: June 2012 – – End of use of ISSB
July 2011 famine and continuing blocks.
conflict
Camp population:
At maximum over 450,000
registered refugees
Project outputs:
Camp planning and site
construction for 76,000 people on
two sites
Plot size:
Up to five people: 10m x 12m August 2011 – – Kambioos camp
More than five people: 12m x 15m opened
– Ifo 2 camp opened
Kenya Somalia July 2011 – – Famine declared
Dadaab
Early 2011 – – New arrivals start to
increase
March 2010 – – Ifo 2 site identified
Project description
Following a massive influx of Somali refugees to the camps at Dadaab in Kenya, two new camps were
planned and built. Camp services were set-up and a refugee-led committee was established to manage the
camps. Planning was for 200,000 people, but poor security and lack of government recognition meant that far
fewer people settled at the sites. The majority of families were sheltered in tents. Later shelters were built with
plastic sheet on timber frames. As families became established, many built their own structures. After some initial
construction, use of Interlocking Stabilised Soil Blocks (ISSB) was prohibited by the government.
Strengths and weaknesses 8 There were difficult labour relations between

9 Qualified and experienced technical experts oversaw the host community and the refugee population,
camp planning and construction. initially exacerbated by differnt policies by different
9 The Government of Kenya supported site implementing organisations.
identificaiton, physical planning, shelter construction 8 Security was poor and the lack of official recognition
and registration of refugees. of Kambioos camp meant police presence was
9 Significant refugee and host community insufficient.
participation in the project. 8 Poor security, challenging host community relations
9 Complete settlement services were established. and difficult access meant that the camps ended
(including water supply and sanitation, health, up being much smaller than planned. The growing
education). population of the camps at Dadaab, ended up moving
9 Settlements were built rapidly once there was into the existing densly populated sites.
agreement to start. 8 There were insufficient materials available to the
9 Site planners learnt lessons from the current camps refugee population for shelter construction and fuel.
and paid particular attention to improve firebreaks. - After 300 shelters were built with Interlocking
8 The camp populations removed vegetation and Stabilised Soil Blocks (ISSB), the government prohibited
damaged the surrounding environment. further construction to avoid the sites becoming
permanent camps.
48
New sites were identified, planned and constructed within 20km of the existing town of Dadaab to cope with major new
influxes and a backlog of non-registered new refugee arrivals. Initial planning was for sites for 200,000 people.
Background government that additional camps Family latrines and showers were
(See Shelter Projects 2009, A.10) would signify increased insecurity, built at the corner of each individual
not a positive message for Kenya plot, 8–10m from the shelter. It was
The conflict in Somalia led to
to be promoting internationally. expected that refugees would take
forced migration of thousands of
Additionally, there was the proper care of them and not allow
Somali nationals into the neigh-
concern that refugees would clear foul smells to develop. When a
bouring countries, including Kenya.
vegetation, potentially causing latrine was full it would be decom-
Since 1991, the Garissa County of
conflict with the host communities. missioned, backfilled and replaced
Kenya became a home to refugees
with another one close by.
fleeing war torn Somalia. Dadaab, Site planning
a small town within the County is Both camps were planned based Implementation
located 100km from Garissa town on a community concept with 10 or Parts of “Ifo 2” camp had flood
and 90km from the Somali border. 12 shelters. zones. As a result a Topographi-
In 2009, Dadaab had a popu- cal Survey was conducted in 2011
Each camp was planned with an
lation of 250,000, mainly Somali and recommendations were made
8m wide sanitation line between
refugees. They were settled in to for flood mitigation and control
communities. This break was
three major camps known as Ifo, measures.
for sanitation facilities, including
Dagahaley and Hagadera. communal showers, latrines and The site of Kambioos had fewer
Continuous drought inside garbage pits for the initial settle- flooding issues, but there were
Somalia coupled with persistent ment. Roads were 15m wide. initially concerns about the water
fighting led to further displace- scarcity, and the additional chal-
Strong camp management was
ments from Somalia into the existing lenges caused by the sandy soil and
required to enforce these breaks,
camps in Kenya. By mid-2011, up a bedrock. This created issues in
as there was a tendency for house-
to 1,400 Somalis were arriving per pitching tents and digging latrines,
holds to build fences out of thorns
day, leading the camp population while access roads, both to and
and brushwood that encroached on
to increase to over 450,000 people. within the site were a challenge.
them. This had also been an issue in
existing camps in Dadaab. Both sites had significant
Site selection
security issues, hampering access,
The massive influx of refugees The camps were built in phases.
with major incidents, including kid-
led to the need for new camps. Each of the phases of “Ifo 2” camp
nappings occurring at both sites.
was planned to measure 2.5km
The process to identify new
x 1.5km. Kambioos site was built The following roles were taken
camps began in 2009 with three
in four planning phases each with on by different organisations in the
possible sites being identified for
10 sections and seven residential two camps:
Kambioos in Fafi district and one
blocks.
site for Ifo extension in Lagdera Camp management agency:
District. Each plot initially measured 10m responsible for site planning
x 12m. However, as households and shelter in the two camps. A
After a series of negotiations
encroached on sanitation lines and team of 6 surveyors and planners
with the respective host commu-
roads, the plot sizes were adjusted per camp and one overall site
nities, it was agreed that the two
to 12m x 15m when relocation planner was responsible for
camps were vital for decongesting
of refugees living in the outskirts supervision of all works.
the existing camps.
began in mid-2012. Depending Construction: responsible for
During these negotiations, there upon their size, larger families were roads, schools, health facilities
was a significant concern from the allocated two or three plots. and general infrastructure
provision in the two camps
The site of Kambioos was covered in dense and thorny vegetation and had very sandy soil, requiring additional care to
be taken with construction of shelters, latrines and other infrastructure.
(six staff were involved). An sheets which they used to construct with Section Leaders, Community
organisation assigned two shelters in addition to the tents. Peace and Protection Teams, Site
engineers in Kambioos and Planning, Shelter, Food Advisory
To provide families with a more
another two in "Ifo 2" to WASH and Graveyard committees.
durable solution, Interlocking Stabi-
directly monitor the works that
lised Soil Block (ISSB) shelters were The future
were sub contracted to local
built from May 2012 onwards. Kambioos camp suffered from
building contractors.
The plan was to construct 16,000 several serious security incidents.
Shelter partner: responsible
shelters in a 2 – 3 years period. One of the reasons for poor
for emergency shelters in the
two camps (four staff were By the end of June 2012, 296 of security at the Kambioos camp
involved). these shelters had been completed. was that there was a lack of police
Operating partner: In July 2012 however, the Kenyan presence, despite a plot measuring
responsible for sanitation and government stopped the produc- 300m x 300m had been allocated
hygiene in Kambioos alone and tion of ISSB shelters stating that for a police station. By 2013 plans
over 40 labourers to support these were permanent structures were underway to construct police
in latrine construction (six staff rather than refugee shelters. station.
were involved). Police were not deployed since
In November 2012, the con-
Most of the challenges expe- struction of temporary shelters the site was not officially recognized
rienced in the site construction was approved by the Kenyan au- by the government until 2013. As
were labour related. Early on in thorities. These had timber frames, a result, financial resources were
the construction, different partners plastic sheeting walls and a corru- limited, and only 18,000 people
paid different wages to labourers. gated iron roof. This design was an moved into the camp initially
Wages for labourers were later interim structure, to facilitate rapid planned for 150,000 people. Similar
standardised to reduce delays. Pay delivery of a durable roof, while ne- issues were faced at Ifo 2 camp
delays also caused strikes for up to gotiations on other shelter options with a total population of 69,000
7 days. continued. by the end of 2012. The rest of the
new arrivals settled in the outskirts
Initially there were challenges Tents were also used for shelter of existing but congested camps.
in identifying who represented the at Kambioos camp, and plans were
host community, as many forceful put in place to replace these with No significant fires were
individuals claimed to be a repre- temporary shelters as well. reported in either camps. This was
sentative. attributed to the proper planning
Services and good management, reducing
Shelter When the sites were set up, encroachments into open spaces
At Ifo 2 camp, a total of 16,000 water was brought in by truck. and effective firebreaks. This was
tents were issued to refugee Boreholes were made and 16.5km in contrast with the congested old
families during the relocation in of water pipeline, 41 tap stands and camps.
July – October 2011. By the end of 246 taps were later installed.
2012, they were in a bad condition On 11th January 2013 Kambioos
and 6,000 were replaced with new By the end of 2012, one opera- camp was officially recognized
tents. tional borehole in Kambioos camp by the government, and became
delivered sufficient water for its "foreseen as one of the camps
Tents in the camps had a population to receive 20 litres per where refugees residing in urban
limited lifetime averaging just person per day. A health post and areas in Kenya will be relocated
over 6 months. Although many of primary schools were also serving to, in accordance with the govern-
the arriving families from Somalia the population, and plans to build ment Directive issued in December
were accustomed to nomadic and a secondary school were underway. 2012 calling for the relocation of
moveable structures, training in refugees and asylum-seekers from
maintaining tents was required. Camp management structures
urban areas to refugee camps".
were established in both camps
Over time, many of the refugees (one chairman and one chair lady)
covered their tents with plastic
sheets while others purchased iron
50
Top to bottom: Site marking; Tent erection on a windy day; Newly established blocks at IFO camp extension.
Camps were organised into a) plots, b) communites, c) blocks, and d) sections.
www.ShelterCaseStudies.org Photo: Joseph Ashmore 51
Dagahaley
Refugee Camp
m DADAAB OVERVIEW
jir
Wa
To
m
i
L ibo
To
Ifo 2 East & West
Refugee Camp
Kenya- Fafi District Sources; UNHCR, LWF/DWS-Dadaab
©2012
m
Ifo 1 To Liboi
Datum: WGS 1984 Refugee Camp
52
Kambioos Refugee Camp Overview Projection: UTM Zone 37
A.15
The boundaries and names shown and the designations

used on this map do not imply official endorsement or
GOVERNMENT OF KENYA
As of June 2012 acceptance by the United Nations
LWF/DWS
Dadaab
Hagadera
POLICE Refugee Camp
!
C
s sa
G ari
m
1C 1B 1A 2A 2B 2C 3A 3B 3C 4A 4B 4C 5A 5B 5C 6C 6B 6A 7A 7B 8A 8B 8C 9A 9B 9C 10A 10B 10C To
ISSB PS
Kambioos Alinjugur
Refugee Camp
HC WC
HC HC
PF CC
WFP D.C 0 1 2 3 4 5
CS PS 1D 2D PS CS 3D PS CS 4D PS CS 5D PS CS PS 6D 7D PS 8D PS 9D PS 10D PS Kms
WFP D.C Market
CS CS
RC
To Yumbis
SP
UNHCR
!
C SP CRS m
1G 1F 1E 2E 2F 2G 3E 3F 3G 4E 4F 4G 5E 5F 5G 6G 6F 6E 7E 7F 7G 8E 8F 8G 9E 9F 9G 10E 10F 10G

m
RC
!
C RC
VC VC
HC HC
Green belt SS CS Green belt Graveyard Green belt Green belt Graveyard Green belt SS CS
WFP D.C
Green belt Green belt

Graveyard
Graveyard
Green belt
CFS
CFS
DC Green belt DC
®
To H a g a d e r a
Legend
20C 20B 20A 19A 19B 19C 18A 18B 18C 17A 17B 17C 16A 16B 16C 15C 15B 15A 14A 14B 14C 13A 13B 13C 12A 12B 12C 11A 11B 11C
Food Major road
Blocks Utility road

CS PS 20D 19D PS CS 18D PS CS 17D PS CS 16D PS CS CS PS 15D 14D PS CS 13D PS CS 12D PS CS 11D PS CS
Health Trails
Boreholes
Market
TBA TBA TBA TBA

!
C Operating
Security
Graveyard Graveyard Graveyard TBA TBA TBA TBA Graveyard Graveyard Graveyard
!
C Not in Use
!
C Education
WFP D.C WFP D.C
TBA TBA TBA TBA
Sanitation
Green belt
TBA TBA TBA TBA
WFP D.C WFP D.C
Environment
Graveyard Graveyard Graveyard TBA TBA TBA TBA Graveyard Graveyard Graveyard
!
C Camp Facilities
TBA TBA TBA
!
CTBA
Key
CC - Community Centre
CS PS 21D 22D PS CS 23D PS CS 24D PS CS 25D PS CS CS PS 26D 27D PS CS 28D PS CS 29D PS CS 30D PS CS CS - Community Space
CFS - Child Friendly Space
CRS Catholic Relief Services
DC - Disability Centre
HC - Health Centre
IMC - International Medical Corps
DC DC IRC - International Rescue Committee
CFS
CFS

Green belt SS CS Green belt Graveyard Green belt Green belt Graveyard Green belt SS CS Green belt
ISSB PS - Interblocking Stabilised Soil
To Yu m b i s
HC Green belt HC Blocks Production Site
Graveyard
Graveyard
VC VC
RC RC
PS - Primary school
RC - Reception Centre
WC - Women's Centre
m
SP - Section Post
SS - Secondary School
TBA - To Be Allocated
Plan from June 2012 of Kambioos site.

CS PS 40D 39D PS CS 38D PS CS 37D PS CS 36D PS CS CS PS 35D 34D PS CS 33D PS CS 32D PS CS 31D PS CS
VC - Vocational Centre
Conflict / Complex
WFP DC - World Food Programme

Distribution Centre
40G 40F 40E 39E 39F 39G 38E 38F 38G 37E 37F 37G 36E 36F 36G !
C 35G 35F 35E 34E 34F 34G 33E 33F 33G 32E 32F 32G 31E 31F 31G
1:12,500
!
C 0 250 500 1,000
Meters
A.16 Lebanon – 2007 – Conflict

Keywords: Unplanned camps, Planned and managed camps, Urban neighbourhoods, Housing
Update: repair and retrofitting, Cash, Structural assessment.

Lebanon
– Project ongoing
Project location:
Palestinian “gatherings” in and
around Saida, southern Lebanon Dec 2012 – – 557 shelters reha-
bilitated since 2005.
Conflict: May 2012 – – Completion of 160
Palestinian refugees rehabilitation in
Displacement date: buildings from the
1948 to present 17th and 18th cen-
tury in the historical
Number of people displaced: centre of Saida
40,000 Palestinian refugees
in gatherings (2009), 450,000 – First rehabilitations
Palestinian refugees in Lebanon without using a
contractor
(2012)
Project outputs: Jan 2008 – – Introduction of a
Repairs of 557 shelters, including new “complex”
412 roof repairs roofing solution
Occupancy rate on handover: Lebanon Oct 2007 – – 250 shelters reha-

Close to 100 per cent bilitated
Shelter size:
40m2-60m2, Average 50m2 Saida
Materials cost per shelter: 2005 – – Project start date
US$ 600 – US$ 2,500: Roof only
US$ 5,500: Full rehabilitation with
services
US$ 2,300: roof only
US$ 7,800 full rehabilitation with
services. 1948 – – Conflict date
Project description
The organisation ran a series of projects since 2005 to improve the shelter standards of Palestinian refugees
living in “gatherings”. Structured repairs focusing on roofs were conducted with associated water and sanitation
improvements. Eight gatherings in the Saida area were targeted with around 25 per cent of the shelters repaired.
The organisation also carried out other rehabilitations in other parts of Lebanon during the same period. Many of
the initial lessons learnt were adopted by other organisations in subsequent responses.
Strengths and weaknesses This could be mitigated against through improved

9 The project built on its own experiences in different social analysis and involvement of local leaders.
implementation methods. As it progressed it reduced 8 Community participation and support for the project
reliance on contractors, resulting in significant could have been improved through greater community
efficiency and quality gains. mobilisation. Greater inputs from beneficiaries in
9 Effective new technical solutions for roofing were terms of labour would also have helped to bring down
used. relatively high unit costs.
9 The organisation worked hard with multiple 8 Construction contractors performed poorly, leading
stakeholders to negotiate access to gatherings where to programme delays and poor quality construction.
civil works were previously forbidden due to land To remedy this, the organisation was forced to directly
tenure, political or conservation reasons. implement the construction.
9 Introduction of beneficiary participation in the form - The relatively small scale of interventions and the
of unskilled labour was a success. significant costs per household reflect the complex
8 There were protection issues with some renters operating environment and the nature of the works
being evicted from properties following rehabilitation. required.
A.16 Conflict
The project made improvements to different types of structures, including multi-storey buildings.
The organisation moved to from a contractor-led approach to a direct-build approach to construction to improve quality.
Photo: Julien Mulliez
Background barracks built for the Lebanese Implementation

The Arab-Israeli war of 1948 families displaced by the 1956 As a number of shelters were
displaced thousands of Palestinians, earthquake which were then found to be structurally unsafe,
with thousands seeking shelter in later sold or rented out stabilisation works needed to be
camps in Lebanon. There is still no multi-storey buildings with zinc conducted with care. Inhabitants
political solution to the displace- roofs and very limited space were advised to evacuate until
ment, and many refugees experi- between buildings repairs had been completed.
ence very poor living conditions. single-storey concrete housing,
By repairing the shelter the or-
often low quality with zinc roofs
The largest Palestinian refugee ganisation was effectively guaran-
new apartment buildings
camp, Ein El Hilwe, is in Saida. The teeing its safety to the inhabitants
with concrete roofs in good
gatherings in the Saida area are and therefore taking on consider-
condition.
found in three types of location: able responsibility for the quality of
The most dangerous housing the work.
within Ein El Hilwe camp itself was often found in the areas where
between Mieh Mieh and Saida land-use was disputed. The organisation made a transi-
city tion from contractor-led rehabilita-
within the old city of Saida in Selection of beneficiaries tion to direct-build. This decision
urban Lebanese communities The organisation’s social team was taken following concerns over
A survey of all Palestinian gather- made home visits in the target the quality of contractor’s work.
ings in 2009 concluded that around areas, filling in questionnaires with Those contractors that were able
30 per cent of the housing in Pales- both technical and social data. This or willing to work in the gatherings
tinian gatherings had shelter reha- was followed by a technical team often used unskilled labour and
bilitation needs. Gatherings within mapping housing with “highly amateur equipment.
the urban Lebanese communities urgent shelter needs”. This benefi-
The organisation found that it
in Saida tended to have less urgent ciary list was submitted to the gath-
could ensure better quality work,
needs compared to those gather- ering’s local committee.
and improve structural safety by
ings located in Ein El Hilwe camp. After the committee made implementing directly. It was also
The majority of gatherings had high additions to the list, the organisa- able to carry out the work cheaper.
or moderate shelter needs, often tion made a final decision based on
with leaking zinc roofs, water- By implementing direct-build
overall social and technical criteria,
damaged concrete block walls, and projects the organisation was also
including household income, age
serious structural problems. able to select community par-
structure, and whether members of
ticipants to receive cash-for-work
Water and sanitation problems the household were disabled.
and to provide basic construction
were also identified, mostly due The social team also commu- training for beneficiaries during the
to poor chlorination practices and nicated with the local population repairs.
poorly-maintained water networks. throughout the project to minimise
Rehabilitation followed a
Land ownership in Saida gath- potential conflicts and encourage
five-step process:
erings ranges from public land,. participation.
which is illegally occupied but 1.Information of stakeholders and
The gathering’s local committee
tolerated by the municipality, to selection of beneficiaries,
was involved in the identification of
illegally occupied private land 2.Bill of Quantities (BoQ) and
people who would be involved in
where evictions are being sought plans of selected shelters,
the cash-for-work part of the con-
by landlords. 3.Purchase of materials and
struction. The organisation reserved
equipment, preparation of
Shelter types included: the right to make a final decision
workers contracts
over who would work in order to
multi-storey buildings with 4.Implementation of works
ensure fair selection.
concrete roofs, converted from 5.Handover.
54
“In winter, me and my

brother used to fight as to
who is to sleep in the corner
where the leakage is worse;
we no longer have to fight
about that”.
Young focus group participant
Left: Example of poor construction by a contractor in 2007, Wooden girders insufficient to support the new roof.
Right: Direct build, correct use of ring-beam to support the roof.
Photo: Arnaud Fratani
A specific bill of quantities had Technical solutions small panels made the roof easier
to be drawn up for each household Working on multi-storey to repair which is useful in conflict
and each household had to sign an buildings required special consid- areas where localised roof damage
agreement before work could start. erations. Repairs often involved is common. However, skilled
the use of large amounts of sand, workers were required to lay it, and
The organisation spent consid-
cement and tiles, creating poten- therefore greater management by
erable time and effort to negotiate
tially dangerous loads on weak, the organisation was required.
with authorities for permission to
repair shelters in illegal gatherings. elevated structures. Floor loads 3) Structural reinforcements
A good relationship with the influ- were reduced by up to 50 per cent Concrete roof/floor slabs in mul-
ential Members of Parliament from by: tistory buildings were often poorly
all political sides was developed and cutting the amount of sand used supported. Steel beams were
they became keen to take partial for flooring which increased the installed, supported at both ends
credit for the assistance projects. strength of the concrete mix by reinforced concrete lintels or by
The organisation also required reducing the thickness (with a steel column fixed on an isolated
specific authorisation from the some resulting loss in levelness reinforced concrete foundation.
Lebanese army for the transport of of the floor); During the rehabilitation the steel
building materials to the shelters. reducing the amount of mortar beams were supported by metal
for tiling; props.
Once the materials were
purchased, meetings were held to using lightweight tiles in place A number of walls were found
provide households with a complete of traditional tiles. to be unable to bear the loads
overview of what work would (and Following experience from placed on them and new reinforced
wouldn’t) be done. previous projects, three key concrete columns were built to
technical approaches were adopted make the shelters safer.
Shelters were divided into by the organisation from 2008:
groups and work was carried out Impacts
on 8 to 12 shelters at a time. An 1) Reinforced concrete ring An independent assessment at
expatriate project manager was beams the end of 2008 concluded that
supported by a local engineer and To support rehabilitated roofs, family relations, decreased tensions
foreman for daily site supervision. concrete ring beams were intro- within the households, reductions
duced. Theses would reinforce the in infectious diseases and improved
A maximum of seven weeks structure, add a slope for the roofs
to complete a shelter was set as a personal hygiene practices were a
and provide connections to support direct result of the project.
target. the roofing girders.
The assessment noted that poor
DRR components Steel reinforcement was used housing conditions tended to have
Where possible, the organisa- in the corners to connect walls a disproportionately large negative
tion reinforced the structure of together and make the structure impact on young women and girls.
shelters in order to improve their more earthquake resistant. The impact of small things such
earthquake resistance. This included
2) New, insulated roofs as rehabilitated bathrooms with
improved foundations, lintels, ring
A french roofing product, made lockable doors made important
beams, reinforced slabs, and in
of zinc sheet, insulation material positive impacts on girls’ and
some cases, additional steel girders
and a bitumen was introduced. The women’s privacy.
supported with steel columns.
A.17 Conflict
A.17 Lebanon – 2011 – Conflict

Keywords: Hosting, Support for host families, Rental support, Housing repair and retrofitting,
Case Study: Cash / vouchers, structural assessment

Lebanon
Project location:
Bekaa valley and Wadi Khalid
(northern Lebanon)
Disaster: 17 months – – 120,000 registered
refugees in lebanon
Syrian conflict
Conflict date: 13 months – – Sealing off begins
March 2011 (ongoing)
Project start: 6,900 registered 10 months – – 26,000 registered
refugees in Lebanon refugees in lebanon
End of 2012: 119,596 refugees,
though numbers were rapidly
rising
Project outputs:
Rehabilitation of 980 houses
“Sealing off” 1,555 houses
2 months – – 6,290 registered
Non-food items for 1,200 refugees in Lebanon
households Lebanon
Shelter size: Beirut
Variable September
2012 – – Rehabilitation
One room with sanitation project begins
facilities per hosted family
Cost:
US$ 1,700 / family
housing rehabilitation
US$ 40 / family sealing off March 2011 – – Conflict start
Project description
This project rehabilitated houses where people fleeing from Syria were hosted. It also made quick repairs to
winterise dwellings and distributed non-food items, including stoves and fuel. Particular care was taken with
targetting of affected populations through detailed social and structural assessments of hosting arrangements.
Assessments were followed by phased cash payments for host families to make repairs. As refugee numbers
continued to rise, the organisation conducted pilot cash for rent and transitional shelter construction projects.
Strengths and weaknesses remain. In reality, however, there were few evictions.
9 The project built upon existing hosting capacities 8 The total hosting capacity in case of new influxes of
and provided support for the host families. refugees was not assessed in detail.
9 The investment in improved shelter and domestic 8 It was not always clear if hosted families stayed free
infrastructure remained with the hosting families. This of charge or had to make some payment to their hosts.
encouraged good relationships with the refugees. - There are significant numbers of privately owned
9 Although the project was focused on shelter, the empty and incomplete buildings in Lebanon.
project was flexible and included works to improve - Whether or not refugee families would be welcomed
water and sanitation for hosts' houses. by host families strongly depended on the political
8 The project required a large number of staff to allegiances of the local authorities.
make multiple visits per house. This made the project - Significant intervention costs per family were due to
difficult to scale-up quickly in response to rapid refugee high commodity costs in Lebanon.
influxes. - Although the total number of direct beneficiary
8 As a practical solution to assure tenure, limited households may seem relatively low, assessments
one-year hosting agreements were signed, after actively identified many families as being able to cope
whoch there was no assurance that the families could without assistance.
56
An urbanised area of Bekaa valley where many families were hosted. There were many privately owned and partially
completed houses that could not be used as the owners could not be identified or too much work was required.
Before conflict in Syria the Akkar Region and the Bekaa Weatherproofing - Quick
Lebanon has had a long history valley. The numbers of registered rehabilitation with plastic sheets
of immigration from Syria, with displaced Syrians rapidly increased to provide weather.
hundreds of thousands of Syrians from around 6,000 at the project Cash for rent and t-shelters
estimated to be working and living inception in October 2011 to over Planned in case of larger influxes
in Lebanon. 100,000 by the end of 2012. This in 2013.
far exceeded initial planning figures This case study focuses on the
Despite its relatively small size,
for the expected scale of migration. housing rehabilitation component.
the climate in Lebanon varies
greatly both seasonally and geo- The situation of Syrian refugees
Initial beneficiary
graphically. In summer it can be in Lebanon is made more com-
selection
very hot at lower altitudes. During plicated by political and religious
The project was established in
winter, parts of the north and east divisions. Refugees mainly settled in
locations with the largest popula-
of the country see snow, while locations with sympathetic munici-
tions of registered Syrians displaced
coastal regions remain warm. palities, where they felt safe. This led
by the conflict.
to relatively localised populations
There are a large number of
of refugees in the Bekaa valley. It As there were relatively few
privately owned, unoccupied or
also led to challenges in identifying actors involved in host family
partially complete houses across
refugee families, as some preferred support early in the response, the
Lebanon. Most houses are rein-
to remain in anonymous. organisation was able to coordinate
forced concrete buildings with
with the other organisations, both
cement block walls. A few older Shelter strategy formally and informally, in the field
houses have mud block walls. Most The strategy adopted by the or- as well as with the local authorities.
buildings in urban areas as well as ganisation focussed on six core ac-
in rural areas are multi-storey. tivities: The organisation received a list
of Syrian families from the United
Lebanon is classified as an up- Housing rehabilitation - Nations who were registered as
per-middle-income country by the Mid-term solution that provides being displaced from Syria, and who
World Bank. There is significant weather-proofing (doors met additional vulnerability criteria.
wealth, especially in coastal areas. and windows), and improves These criteria included families with
However, there is also consider- sanitation, safety, electricity and extremely low incomes, families
able poverty, and migrant workers privacy headed by women or elderly
can be found across Lebanon living Non food item distribution - people, families with chronically ill
in makeshift shelters made from This includes heating stoves and members, families with no adults
timber and plastic sheeting. refill coupons and hygiene kits and households without a water
Management of collective supply.
During the conflict
shelters - Capacity building
Following intensified civil conflict
and follow-up to provide the
in Syria in March 2011, thousands
larger collective shelters with
of Syrians fled into Lebanon, mainly
proper management.
into the north and east of Lebanon,
House being refurbished to house several refugee Migrant worker housing. Such structures were in use
households. before the refugee influx.
Photo: Joseph Ashmore Photo: Joseph Ashmore
Household assessment “Repair minor cracks in walls Collective centres

The organisation carried out - cut out the cracks with a In addition to rehabilitation of
a further assessment visiting all host family houses, some collective
masonry grinder to form square
shelter with a team of engineers centers were also repaired. Typical
and social mobilisers. This team edged slots and remove all works carried out include:
completed two assessment forms: dust and debris. Apply Epoxy
replacement of doors and
one highlighted structural and in- Primer liquid by brush and
windows and broken walls
frastructure needs with questions, whilst tacky lay in the putty- roof repairs
including: like repair mortar, Mouldable rehabilitation of sanitation
Are the kitchen and wc Epoxy Mortar. Apply mesh tape facilities
separated? and plaster. Work to include all provision of cooking facilities
Is the bathroom connected to a plaster.” water and electricity supply
permanent water supply? installation of partitioning for
Are there sewerage connections/ Example of detail in the BOQ privacy.
networks? Further works, such as decorat-
Is there structural damage? Cash grants were allocated to ing and the provision of additional
Is there access to drinking households so that they could pay social spaces were also assessed
water? for repairs. Cash grants were paid but not prioritised during the
Is there access to electricity? in installments following the organ- emergency rehabilitation.
How large is the room? isations’ monitoring teams’ confir-
How many people are sleeping mation that certain stages of the What next?
in each room? construction had been completed: In late 2012, as the number of
The second form focussed on Syrians in Lebanon continued to rise
1st installment of 25 per cent
social issues and other vulnerabili- rapidly, and winter approached, ad-
was paid when the contract
ties. Approximately one quarter ditional solutions were required.
agreement was signed.
of households were identified as New programming responses
2nd installment of 30 per cent
needing assistance while the others included rental subsidies, ongoing
was paid when sixty per cent
appeared to be coping adequately. distribution of non-food item,
of the 1st installment were
including stoves and fuel vouchers,
Implementation completed.
and simple “sealing-off kits”. These
3rd installment of 45 per cent
Once families had been iden- kits consisted of timber, plastic
paid on completion.
tified for inclusion in the project, sheeting, tools and fixings that
A contract agreement was could be used by mobile teams to
the engineering team returned
signed by all parties to prevent the seal windows and doors in unfin-
and conducted a detailed assess-
host family from demanding addi- ished buildings.
ment of the works required using a
tional rent from the Syrian family or
standard but detailed bill of quanti- Contingency planning was also
evicting them following the reha-
ties template. undertaken to include tents and
bilitation.
Each line in the bill of quantities other emergency shelters that could
The organisation operated from be deployed at scale, either individ-
was given a unit cost, from which
two field offices, each less than ually on small plots of land or inside
materials costs were calculated.
three hours drive from Beruit. Staff unfinished buildings.
The documents were reviewed worked in teams of three people.
in the office, and a schedule of work Project managers at each location
was agreed with the homeowner. supervised two teams each.
58
A.18 Madagascar – 2012 – Tropical Storm

Keywords: Construction materials, Core housing construction, Training, Guidelines and training
Case Study: materials.

Madagascar
Project location:
east and south east Madagascar
Disaster:
Intense tropical storm Giovanna
10 months – – Project completed
and moderate tropical storm
Irina
Disaster date:
February 2012
Number of houses damaged/
destroyed:
45,500 6 ½ months – – Construction
Number of people displaced: 6 months – – Materials delivery
332,204 affected – Beneficiary identifi-
55,060 people displaced cation
Project outputs: – Training of builders
5½ months –
Construction of 598 shelters
Training of builders and
beneficiaries Brickaville 5 months – – Shelter design and
drawings
Shelter size:
12m2 Vatomandry
Madagascar
US$ 128
Project cost per shelter: Farafangana
US$ 250
February 2012 – Tropical storms
Project description
This project formed community committees to select beneficiaries and monitor the building of 599 houses
in rural locations. Close monitoring by beneficiaries allowed a degree of remote management of the project to
improve quality in a difficult to access area. The project aimed to build safer shelters using local materials.
Strengths and weaknesses assisted by committee members in each village.

9 To reduce overheads whilst maintaining quality, the 8 Different approaches between organisations meant
project used remote management with community that beneficiaries did not always accept solutions,
committees to monitor and ensure material and making implementation problematic..
construction quality. 8 Not enough consideration was given to other local
9 The project used a committee to select beneficiaries materials such as bamboo.
and improve transparency. 8 Shelters should have varied according to materials
9 Ther project was accompanied by an education used.
programme on safer building practices to increase 8 Due to budgetary constraints shelter dimensions
project reach and support people who were not were not adapted to household size.
directly supported by the project. 8 Increasing prices of materials led to a reduction in
9 Municipal authorities were involved in issuing land the number of households supported from 680 to 598.
certificates for landless households. - The project only received two-thirds of the funds
9 The design process involved beneficiaries and required for its original budget. Cost savings were
local craftsmen from the start to ensure that shelters made by reducing staff and the number of beneficiary
were culturally acceptable and adapted to local households.
environmental conditions. - Problems with local suppliers caused a month long
8 Illiterate community members had difficulties using delay in project implementation.
the quality-control checklist. However, they were
Shelter under construction (left) and completed (right). The project used remote management through community commit-
tees to build 599 houses in difficult to access villages and conduct trainings.
Photo: CRS
Before the cyclone month after the storm, only 15 per 2004 cyclone. It replicated the
The island of Madagascar is cent of households had managed design components of the shelters
prone to cyclones, floods, droughts, to rehabilitate their shelters. which survived the cyclone and es-
epidemics and pandemics, fire and tablished a checklist for construc-
Many households headed by
locust swarms. Previously, in 2007, tion.
women, the elderly or disabled
a major cyclone directly affected people were often not able to re- A funding shortfall of nearly a
about 525,000 people. habilitate their homes within 6 third meant planned staff numbers
Over the past 35 years, Mada- months of the cyclone. were cut and responsibility for
gascar has experienced 46 natural monitoring construction quality
Materials to repair shelters were
disasters affecting a cumula- was passed onto the community
hard to come by, and many families
tive total of more than 11 million committees.
were too poor to buy them.
people. Government studies from The project was implemented
2008 indicate that there will be a Selection of beneficiaries in 83 villages across three districts.
greater intensification of cyclones During national coordina- Each district was supported by
and increased rainfall over the next tion meetings, organisations were three field workers, a technician
50 years. allocated different communes to and a project coordinator.
work in. A commune is made up of
After the cyclone several villages and each organisa- A typical construction required
Tropical cyclone Giovanna hit tion selected beneficiary villages two carpenters and eight labourers,
eastern and central Madagascar based on damage reports. paid through food-for-work. At
in February 2012, causing signifi- least two of the labourers in each
cant damage. Winds peaked at The organisation established team were women. Once materials
230km/h. It was followed by the a community committee in each were available, a house could be
severe tropical storm ‘Irina’ and village (see below), and households built in five days.
there were subsequent floods and who had lost their homes and who
landslides in the south-east. were unable to rebuild, were the Committees
target beneficiaries. The focus was The project was implemented
The two disasters caused signifi- mainly on the disabled, the elderly, through the village committees.
cant damage to housing, agricul- pregnant women and large house- Committees were responsible for
ture, livelihoods, health and schools. holds. identifying beneficiaries and moni-
Less than 5 per cent of the popula- toring the quality of materials and
tion had access to rice stocks and Implementation construction. Representatives
less than 50 per cent had access to The shelters were built on land included:
staple foods. Approximately 80 per belonging to the households before
cent of mixed-crop farmland and to the cyclones. In only one case, the village chief
rice fields were destroyed by the where the household had rented the mayor of the commune
storms or resulting flooding. The their accommodation prior to the a church representative
storm season coincided with the storm, was it necessary for the au- a beneficiary representative.
seasonal ‘lean period’ for farming thorities to allocate a new plot of One or two members of the
families. land. committee monitored housing con-
Many affected households struction using the construction
The organisation began by
sought refuge in welfare centres or checklist. These individuals were
reviewing the government shelters
with relatives and neighbours. One usually teachers or other literate
that were built in response to the
people.
60
The community committee Roof frame those who did not receive a house
worked with a community mobiliser Corner bracings were added. from the project.
from a partner organisation and A cross-beam was added to One month after the project
with the local government office strengthen the roof and to had been completed, eight addi-
responsible for facilitating the create a storage area. tional families had built new houses
emergency response. Corners were connected with following the project design using
strong ropes or metal straps. their own resources.
One person in each village was
nominated as a communications Roof covering
focal point to provide two-way For a thatched roof, wire or Logistics
communication between ben- strong ropes were used to The organisation purchased
eficiaries and responding organisa- connect the roofing to the materials, on behalf of beneficiaries,
tions. ground with heavy rocks. from local suppliers. The suppliers
For corrugated iron roofs, 26 to delivered materials directly to the
Technical capacity 29 gauge sheets were used, and villages. Contingency material
The committees provided the roof structure was secured suppliers were also identified in
technical training to people living in with wood battens. case of a delivery failure.
the village. This allowed the house- Training The beneficiaries made and
holds to monitor the construction The organisation produced a provided rope for the roof. Benefi-
quality themselves and allowed poster that illustrated key points ciaries also contributed to the cost
technical staff to provide more on strengthening houses against of the shelters by sourcing wood for
targeted assistance. cyclones and storms. The aim was the roof supports. The wood was
Staff members and committee to improve the understanding of commonly available, and could be
members were provided with found or purchased at low cost.
detailed plans to ensure quality in
Construction Checklist
construction. Home owners and
Storage & quality Are all the materials stored safely from storm, rain
committee members were supplied
and flood and are secured to prevent theft?
with a simplified construction
Are the quality of materials good?
checklist that helped them to follow
the progress of construction at a
number of key stages.
DRR components Foundation Is the wood dry?

The shelter design was an adap- Has the wood been treated with oil?
tation of traditional houses in Mad- Have you buried the footing to 75 cm?
agascar with the following improve- Have you used broken rocks in the foundation?
ments to ensure better resistance to Does the floor have corner bracings?
future cyclones and flooding:
Foundations Structure Are diagonal bracings used at columns?

Pillars were buried to a depth of Are diagonal corner bracings used at corners to
at least 750mm. connect the diagonal bracings?
A mix of stones (5-10cm in size) Are all joints between the columns and beam made
was compacted beneath and using timber joints and not nails?
around the pillars. Are connections between beams and columns
Pillars needed to be dry before fixed with nailed metal straps?
sinking them into the ground. Roof Are corner bracings used at all corners?
Walls Are metal straps used to connect the roof truss to
Walls were all reinforced with the beam?
diagonal bracing. Are all connections between members made with
The floor beam was mortise and tenon joints?
strengthened with corner Is the joint of the ridge and the truss diagonally
bracing. reinforced with bracings?
A wall plate tied the wall and Are all four corners of the roof beams braced with
roof structure together. diagonal timbers?
All connections were Upgrade items Are metal straps used for wooden connections?
strengthened with metal straps Are ropes used for connections?
or strong rope. The roof was Is wood in contact with the ground treated with an
securely connected to the wall. oil and petrol mixture?
Mortice and tenon joints were
used to connect timbers.
A.19 Nicaragua – 2007 – Hurricane

Keywords: Non-displaced, Construction materials, Core housing construction, Site planning,
Case study: Training.

Nicaragua
Project location: – Project completion
12 ½ months –
North Atlantic Autonomous 12 months – – Dedication
Region ceremony
Disaster:
Hurricane Felix
Disaster date:
4th September 2007
Number of houses damaged
destroyed:
7,895
7 months – – Construction begins
35,100
Project outputs:
150 core houses
50 community members trained – Surveys completed;
5 month –
in carpentry skills. salvage of materials
Auhya Pihni
Occupancy rate on handover: complete
100 per cent Puerto Cabezas
2 months – – Project participation
Shelter size: agreement signed
30m² with community
Materials cost per shelter: elders
US$ 1,400 – Project start
Nicaragua 1 month –
US$ 2,000
4 September
Project description
This project was implemented in the context of a poorly funded response and recovery operations for the
2007 hurricane in Nicaragua. The organisation chose to focus its limited budget on providing improved shelter
conditions for nearly the entire population of one of the most affected villages. The project included physically
re-planning the settlement, building 150 new core houses, and training community leaders and work crews.
Strengths and weaknesses only one village, the village selection process became
9 The engineering design of shelter solution was of pressured.
high quality. - There was a strong pre-existing level of organisation
9 The reconstruction project enabled the settlement in the affected community, facilitating communication.
layout to be rationalised and improved. - The project entailed the use of ethnic language
9 The carpentry skills training component provided interpreters, leader orientation techniques, knowledge-
livelihood opportunites. transfer and community training methodologies to
9 The shelter project addressed almost 100 per cent avoid cultural and language barriers with the target
of shelter needs in the Auhya Pihni community. population.
8 The project scale was limited (less than 2 per cent of - Having a local church partner with insider
the affected population) and did not address the needs knowledge of local power-brokers was crucial. The
of the majority of affected people. fact that the church was respected helped to legitimise
8 The project did not include improveming sanitation the introduction of labour agreements between the
facilities. Some families rehabilitated latrines in implementing organisation and the beneficiaries.
parallel to the shelter intervention but many remained
substandard.
8 By choosing to concentrate all of the support in
62
Tarpaulins were provided as part of a response that had very limited funding both nationally and internationally.
Photo: Mario Flores
Before the hurricane holds did not have access to basic Selection of beneficiaries
Auhya Pihni is a settlement water and sewage disposal, while Surveys were conducted by
located 55km north-west of Puerto nearly half of households had no a group of Nicaraguan NGOs to
Cabezas, comprising indigenous toilet and 7 per cent shared their select a community. Meetings were
Miskito people. The settlement latrines with other families. also held with local authorities,
is within the autonomous RAAN influential church leaders and elders
region, and has its own tribal laws. After the hurricane from the Miskito community.
On 4th September 2007,
The project village is one of ten Hurricane Felix, a category five The community of Auhya Pihni
Miskito communities who possess storm, hit the north-east coast. was chosen, and 150 families were
approximately 150,000 hectares of Winds of 260 km/h caused wide- selected as beneficiaries out of a
land. Traditionally Miskito people spread devastation. In the provincial total of 167 households.
live in large houses with their capital city of Puerto Cabezas, the
extended family. The size of the Implementation
hurricane caused severe damage
extended family has reduced over Following previous organisa-
to houses and services, cutting off
time, with the village community tions’ unfulfilled promises of assis-
all communications. The hurricane
playing a similar role. Kinship in the tance, the community were initially
caused over 160 deaths.
community remains matrilineal. distrustful. However, once the first
Nearly 8,000 houses were two model houses were built, trust
Lands are communal, and a destroyed and smaller settlements improved.
Council of Elders acts as a decision- such as Auhya Pihni were complete-
making body. The person with To build the shelters, local
ly obliterated. The day after the
the position of Justice within the labour was hired from the affected
hurricane the government declared
council served as the chairman of community to support skilled car-
a ‘’national state of calamity’’.
community leaders. These leaders penters brought in from Puerto
have the final say in decisions Neither the government nor the Cabezas and other towns.
affecting the community. international community were able
A processing centre for
to commit the funding to respond
Most of the inhabitants survived timber was established within the
to the overwhelming shelter needs.
on unstable, sporadic sources of community, and homeowners
income, and many of the families Some tarpaulins were provided were paid to support the produc-
were living in extreme poverty as emergency shelter by relief tion and fabrication of construction
before Hurricane Felix struck. agencies to enable families to elements, including posts, wood
stay in their houses and prevent frames and rafters.
The water table at the site is very further displacement. Many
close to the surface and the part of Through its external networks,
families combined the sheeting
the settlement located close to the the local church was able to provide
with reclaimed materials to make
river is prone to flooding. power tools. Community-owned
shelters.
timber processing equipment was
According to the 2005 popula- also made available.
tion Census, 65 per cent of house-
Completed houses in the village.

Photo: Mario Flores
Workers involved in the project elevating houses on stilts to Logistics

were trained in carpentry and reduce the risk of flooding Approximately 10 per cent of the
home construction by the organi- cross-bracing supporting posts timber was salvaged from damaged
sation. Some were trained in the metal straps to reinforce houses. The quality of materials was
production of wooden doors and connections of wooden approved by the homeowners and
windows, which included the use elements the project engineers to ensure that
of woodworking machines. a strengthened structural design their use would not weaken the
use of twisted roofing nails to houses.
To build the houses, the organi-
better secure roofing sheets.
sation established six construction The community initially rejected
teams, each led by a community The goal of these incremen- the use of pine to build shelters,
leader. tal changes was to build stronger because they wanted a more
structures with better tied down durable housing solution.
Coordination roofs.
Due to lack of funds and the The forest had long been a sig-
limited number of organisations Sanitation nificant source of livelihoods for this
operating, there was very little in- Before the hurricane, most community. Before the hurricane,
ter-agency coordination. Overall, households either had or had pre- a community organisation had
local coordination with the church viously used latrines. The organisa- been established with the support
and local authorities was good. tion did not get involved in latrine of an international organisation to
construction, but another organisa- manage the local forest resources.
Technical solutions tion was able to build latrines for 35 As a result, the forestry resources
The house design followed per cent of the households. were well managed for the use
a local design, using familiar of this project, minimising any
There was some disagreement
materials: timber for most of the negative environmental effects.
between the two organisations
house components and corrugated
about the low level of coverage, but The organisation was able to
galvanized sheeting for the roof.
budgetary constraints prevented purchase the timber at a discounted
The final design and covered further work. price from this community organi-
area (30m²) was agreed with the sation who put the money into a
community representatives through community fund for community
a process assisted by architects from projects.
another local organisation.
All timber came from within
The design included three simple 10km of the village and was trans-
disaster risk mitigation features that ported by river.
were new to the community:
64
A.20 Pakistan – 2010 – Floods

Case Study: Keywords: Advocacy, Training, Guidelines, Coordination.

Pakistan
15 months – – 100,000 one room
Disaster: shelters built
Floods
Disaster date:
July to September 2010 – 1 million house-
7 months –
No. of houses damaged: holds provided with
1.8 million houses damaged or emergency assis-
destroyed tance
No. of people affected:
More than 20 million people – National shelter
Project outputs: coordination work-
shop
Coordination established – Revised inter-agency
3 months –
nationally and in 7 provinces response plan
6 weeks – – Flooding continues

in Sindh. Standing
Pakistan Water remains
7 days – – National shelter

cluster team in place
27 July 2010 – – Flooding in North-

ern Pakistan
Project description
The organisation established national coordination across 7 provinces in response to large scale floods, with
the purpose of addressing gaps and increasing the effectiveness of the humanitarian response. The organisation
established a national coordination team that managed a wide range of issues through a system of Strategic
Advisory Groups (SAGs) and Technical Working Groups (TWIGs). It also appointed different organisations as
lead coordinators in the different provinces. District level coordination proved difficult and slow to establish, but
lessons were leant for the following 2011–2012 floods.
Strengths and weaknesses response within Pakistan, they were largely outside of
9 The lead coordinating agency shared responsibility the cluster system.
for coordination with different organisations as focal - Effective coordination requires a physical presence
points for different provinces. at national, provincial and district levels. Although
9 The lead organisation was able to establish a effective coordination must remain focused on output,
reasonably clear division of responsibilities between the certain aspects of coordination are process-focused.
coordination team members and its own operations. Jointly creating a sectoral strategy and shaping
9 The coordination process resulted in detailed gap advocacy positions are two such examples.
identification in the response in the south of Pakistan - There are increasing obligations relating to the
at village level. contingency planning process, as more efforts are put
8 Coordination was slow to be established at district into preparedness and disaster risk reduction. Using
level in the 2010 floods. Lessons were learnt for the clusters to conduct sector planning during ‘peace time’
2011 and 2012 floods. may be good value for money.
8 There were challenges in reaching consensus on - At national level, coordination focuses on ensuring
line management responsibilities due to multiple lead a harmonised and adequate response, policy and
organisations across provinces. resource mobilisation. At provincial or district level the
- Altough local organisations, foundations, focus is on issues of assistance delivery and partnership
philanthropists, and private sector actors have an building. The practical value of coordination increased
increasingly important role in preparedness and the closer it was conducted to the affected population.
The coordination team tracked commodities, both in pipeline and distributed as a core service.
This allowed gaps to be identified and resources sought to fill unmet needs.
Coordination and Clusters Management Authority in 2006 twice per week, but decreased in
Following a review of interna- and the implementation of the De- frequency as the emergency pro-
tional responses to humanitarian centralisation Act of 2010, which gressed into the recovery phase.
emergencies in 2005, the cluster devolved significant, although not
approach was proposed as a way always clearly defined, authority to SAG and TWIGs
of addressing gaps and ensuring the provinces. Multiple groups had to be estab-
responses were more effective. lished to coordinate the response,
2010 floods and most effectively use the time
By clarifying organisations’ roles Floods in 2010 affected 20 of the different parties involved,
and responsibilities, the cluster million people and destroyed 1.7 including donors, government
approach helps ensure predictabil- million houses throughout the officials, NGO and UN partners, and
ity and accountability, and creates country (see A.22 shelter Projects others.
a more structured, accountable and 2010). They struck all 7 provinces
professional system. A Strategic Advisory Group
of Pakistan with 29 districts being
(SAG) was formed to discuss and
See www.sheltercluster.org classed as severely affected.
propose rapid agreement on
Distances were large, and with national strategic issues such as
Shelter Cluster in Pakistan
some locations taking days to travel advocacy positions that the cluster
The Shelter Cluster in Pakistan
to. The scale was such that no one should take and which projects
coordinates shelter activities in
organisation could effectively co- should be promoted for funding.
response to disasters and specific
ordinate on its own, and it was To ensure accountability, SAG
conflicts. The Cluster comprises the
necessary to set up coordination members were elected by all cluster
government, lead organisations and
mechanisms at both the national members with agreed numbers rep-
all of the organisations engaged in
and provincial level. resenting different types of organi-
shelter activities who wish to coor-
sations and donors. The SAG’s rec-
dinate. National Coordination ommendations were submitted to
Before the 2010 floods, the The cluster lead organisation plenary meetings and disseminated
Shelter Cluster had been activated agreed to represent the Shelter by email for final agreement.
in Pakistan after the 2005 South Cluster at the national level. It es-
tablished a team of nine people that Technical Working Groups
Asian earthquake, the 2007
worked relatively separate from the (TWIGs) were formed to deal with
cyclone, the 2008 earthquake
operations of the hosting organi- specific technical issues, such as the
in Baluchistan, and the complex
sation. This independence allowed composition of a winterisation kit
emergency peaking in the 2009 IDP
the team members to represent the or common specifications.
crisis.
“cluster“ and not their host organi-
Coordination challenges within Provincial coordination
sation.
Pakistan include multiple languages The lead organisation agreed to
and the changing institutional The team consisted of a cluster coordinate nationally and in Punjab
roles and relationships within the coordinator, a technical advisor, an and Sindh provinces. Three other
humanitarian community and the information manager (with two as- organisations agreed to coordinate
government. Different types of sistants), a Geographical Informa- the other four provinces.
disasters and conflicts all require tion Systems team (two people)
Sharing coordination respon-
different responses and different and an administrator. As the team
sibilities with other organisations
management of the responses. members needed to visit field and
that had experience and competen-
hub locations regularly, it relied
The governance structures of cies in the shelter sector proved to
heavily on the logistics support of
Pakistan relating to disaster assis- be an effective way to ensure that
the host organisation and other
tance have changed significant- coordination was rapidly extended
cluster members.
ly since 2005. Two of the most throughout all of the flood-affected
important changes have been the The team held regular meetings areas.
creation of the National Disaster in Islamabad. Initially these were
66
Coordination team
National Islamabad
coordination Coordinator
Technical advisor - Information management - GIS - ERF Focal point
Province S Sindh Punjab N .Sindh Baluchistan KPK Kashmir Gilgit/Baltistan

coordination Hyderabad Multan Sukkur Quetta Peshawar Muzafarabad Gilgit
District
coordination
Functional organisation chart for the coordination team in November 2010. Each colour represents a different organisation.
District level coordination was just starting and district focal points were being identified in several of the provinces.
Each role might require several staff, either full time or part time. For example, at national level, the information manage-
ment role was fulfilled by an information manager and two assistants.
A disadvantage of having ensure that support was reaching During the recovery phase,
different organisations taking the the most vulnerable rather than the three agencies acted as district
lead at provincial level was that most vocal. focal points and each covered two
every organisation had a different districts. The same district focal
The nearer coordination took
interpretation of the role of a coor- points were transferred to Northern
place to the affected people, the
dinating agency. This led to some Sindh after the 2012 flood. Their
challenges faced become less
sticking points between national experience and knowledge ensured
focused on policy and resource mo-
and provincial coordination. the rapid establishment of coordi-
bilisation issues and more focused
nation in the newly affected areas.
Some of these organisations on issues of delivery of assistance
were new to the role of cluster co- and partnership building. Practical Non-emergency activities
ordination, so the national team issues included working with local The importance of prepared-
had to spend some time clarifying government officials to facilitate ness was emphasised by the gov-
what the cluster lead role entailed. access to communities, mitigat- ernment, humanitarian organisa-
ing potential conflicts in resource- tions and donors alike. Pakistan’s
Common reporting formats and
scarce areas and identifying the recurrent natural disasters and
digital filing structures were agreed
most vulnerable people affected by ongoing complex emergency make
on in the first weeks of the response
the disaster. preparedness crucial. Consequent-
but this was not sustained and
different versions were later used in The setting-up of district focal ly, cluster leadership obligations
different provinces. This made data points was a slow process requiring expanded beyond response to
consolidation more difficult. specific resources and funding. For include contingency planning.
each district, a capable partner had The shelter cluster prepared
A national workshop was held
to be identified, and memoran- contingency plans for 2011 and
for coordinators and information
dums of understanding needed to 2012 in coordination with cluster
managers a few months into the
be signed to clarify roles, responsi- members, other clusters, and
response to discuss and share expe-
bilities and cost recovery issues. various levels of government. Stock
riences between provinces, and to
synchronise systems. lists were compiled to show stock
2011 and 2012 floods
levels before the monsoon season.
Provincial coordination teams In the 2011 and 2012 flood
A summary of capacity in terms of
varied in size, from a dedicated responses national and internation-
human resources was made, with
coordinator and two information al non-governmental organisations
lists of trainers and experts who
managers to a single coordinator were responsible for coordination
could support emergency distribu-
who had other operational respon- at the district level. The role of the
tions and assessments.
sibilities. district focal points was to monitor
and support the shelter cluster After the 2012 flood, the gov-
District focal points members in the implementation ernment of Pakistan did not request
It quickly became apparent that of their programmes, liaise with international humanitarian support.
coordination would be required local government and keep them Instead it requested for relief stocks
outside provincial capitals and aware of relevant issues, provide from the existing contingency plans
hubs. Many parties recognized that technical and trouble-shooting developed by the clusters to be
the practical value of coordination advice and maintain an overview of distributed to complement its own
increased the closer it occurred to who was doing what, and where response.
the affected populations. they were working. This allowed a
close and thorough monitoring of
District and sub-district coordi-
the response and resulted in a more
nation was essential for organisa-
informed coordination at national
tions entering an area for the first
and district level.
time to avoid duplication and to

Case Study: Keywords: Non-displaced, Tools, Core housing construction, Cash, Infrastructure, Training.

Pakistan
Disaster:
2010 floods
Disaster date:
July to August 2010
23 months - – Evaluation
destroyed:
1,744,471 households damaged 21 months – – Shelter construction
in total (876,249 households
damaged in Sindh province) 19 months – – Material distribution
Project outputs: and training
5,350 shelters constructed
17 months – – Material procure-
61 construction trainings ment
7,638 households cash-for-work
for shelter construction
Occupancy rate on handover: 12 months – – Beneficiary selection
92 per cent 11 months – – Pilot projects com-
Shelter size: pleted
Pilot shelter: 20m2
Materials cost per shelter: 7 months – – Project start
US$ 710 for the shelter Sindh
materials and labour
US$ 983 for the shelter
component of the project
July to August
Project description
The project provided shelter, food security and disaster resilience assistance to flood-affected communities in
Sindh province. 5,350 families were provided with materials, labour and trainings to enable households to rebuild
their shelters. The project design was designed on community-based Disaster Risk Reduction (DRR) principles, but
the constraints of a short project timescale and high target numbers made this challenging.
Strengths and weaknesses beneficiary selection process and piloting leaving

9 Investment of time in a pilot project showed what families with a delay before shelter support was
would work at scale. available.
9 The project aimed to make a lasting impact despite 8 The high cost of the shelter in comparison to local
short term funding by incorporating permaculture, houses reduced the likelihood of replication.
DRR principles and food security. 8 Relatively high cost of beneficiary contribution may
9 Use of locally available materials and skills as well as affect timely financial recovery.
a strong technical training component created a shelter - This project was part of a multi-sectoral approach
design that could be replicated by other families. that included WASH, shelter and food security
9 In depth vulnerability assessment helped improve programmes, implemented in the same target areas.
targetting. - Highly sensitive security situation in target areas lead
8 Large-scale direct procurement was complicated to a need for self-help and a 'do-no-harm' approach.
by scarcity due to high post-flooding demand for - The project led to many discussions in Pakistan on the
materials. benefits of introducing horticulture and permaculture-
8 Some community based DRR activities were hard inspired principles into recovery programmes.
to complete due to tight timeframes and the need to - Flooding in the same areas in 2012 meant that the
construct quickly and at scale. DRR elements of the project were tested and can be
8 The project was slow to start due to the extensive evaluated.
68
The project was designed on community-based disaster risk reduction and permaculture principles. However it found It
difficult to maintain these principles and effect the social change required given the scale and donor time frames.
Photo: ACTED
After the floods Selection of beneficiaries Implementation

(See A.22, Shelter Projects 2010 Union Councils are the local The shelter components of this
for background) administrative unit for humanitar- project comprised four key activi-
ian coordination in Pakistan. The ties:
The 2010 floods reached
Union Councils to be supported
northern areas of Sindh Province piloting of various shelter
were selected on the basis of flood
in August 2010. Comparatively designs to enable the
damage and a gap analysis of
few humanitarian actors were in a identification and replication of
responses planned by other actors.
position to respond to the scale of innovative best practices
the disaster. A 15 minute questionnaire was provision of shelter materials
completed for each household in and toolkits
According to an assessment in
each Union Council. Over 24,000 provision of training on shelter
the target area, up to 60 per cent
families were interviewed to construction incorporating DRR
of households had lost their shelter
identify the most vulnerable house- principles
entirely, while nearly all rice and
holds. This survey and data analysis shelter construction using cash-
vegetable crops were damaged or
took four months. Finally 5,350 for-work.
destroyed. The crop damage was
families were selected, meeting the Throughout the project, the
a particular problem in northern
following criteria: organisation conducted extensive
Sindh since agricultural livelihoods
provided the primary source of Households headed by community mobilisation activities,
income. vulnerable people such as including hazard mapping and
elderly, female or disabled village planning.
Access to shelter and livelihoods
people.
were reported as a priority need Pilot phase
Families with a significant
under the Relief Response Plan In the initial stages of the
proportion of children under five
launched on 1st March 2011 by the project, the organisation purchased
years of age, elderly, pregnant
Government of Pakistan. compressed earth block machines,
and/or lactating mothers and
trained community members in
Where possible, shelters were malnourished children.
their use, and built several pilot
to be constructed using locally Basic low socio-economic
shelters.
available building materials. In characteristics, including a lack
addition to the provision of shelter of income, assets, and bread- However the community and the
materials, organisations were en- winners in the family, and/or organisation’s engineers expressed
couraged to promote ‘appropriate chronic debt. the following concerns about the
technical assistance and support Local community-based organi- use of compressed earth blocks:
revitalisation of the supply chain of sations were identified, or estab- Production was slow and labour
key materials’. Using social mobili- lished. They were responsible for intensive, especially during
sation and mass communications verifying the accuracy of informa- extreme summer temperatures.
strategies, beneficiaries and their tion provided and ensuring that no The local soil type was not ideal
communities were to be mobilised vulnerable families were excluded. for creating the blocks and a lot
to directly participate in the con-
Houses were rebuilt in the same of training was required to get
struction process, either through
locations as before the floods, the right mix of clay and sand.
material or labour contributions.
either on their own land or with the
agreement of their landlord.
There was a lack of acceptance Implementation soil conditions, and the required
of mud houses, as in the local The project was run as two scale and speed of the project made
language mud houses are projects, each funded by separate this part of the project challenging.
described as “katcha” (bad) international donors. Implementa- Despite this, some villages
houses and brick concrete tion varied between the projects, greatly appreciated the trees, and
described as “pucka” (good) though in both cases beneficiar- kitchen gardens were well tended.
houses. ies provided half of the unskilled The organisation was able to
It was very difficult to transport labour. Skilled masons were use the lessons learned from the
the blocks for more than a few provided by the organisation. disaster risk reduction components
hundred meters as they were
On-site training was given to the in its response to the 2011 floods in
easily broken in transit.
masons, focusing on shelter design southern Sindh.
Given these obstacles and the and quality control of brickwork
size of the project, the organisa- and foundations. Materials list
tion decided not to continue with Materials Quantity
compressed earth blocks. Instead Coordination Shelter toolkit
they provided fired bricks and Coordination between other
cement for the lower portions of humanitarian actors working in Wheel barrow 2
Kassi (Trowel) 5
the walls. The beneficiaries contrib- the area and the local authorities, Spade 5
uted sun-dried or fired bricks from including the provincial disaster Sall (plummet) 1
the windowsill level up to the roof. management authority, enabled the Steel pan 5
Block making frame 4
In many cases people were able organisation to share lessons learnt Mask 4
to find or purchase fired bricks for from the innovative techniques and Cotton gloves 4 pairs
their contribution. approaches piloted through this Payodin (injury cream) 1
Band aid (rolls) 2
project. Land rights issues were Water level 1
The one-room shelter design
addressed through working with Iodine balm 1
went through several adjustments
other shelter partners, facilitat- Shelter construction materials
based on the feedback from the
ing constructive engagement with
organisation’s engineers and the Fired bricks (Size 8.5''x4.''x3'') 3,228
landlords. Mud blocks (Size 6''x8''x12'') 1,115
community. These included the
Cement 7 bags
location of the door, the number of DRR / permaculture Sand (wastage not included) 85 ft3
windows, the type of ventilator, the Northern Sindh is highly vulner- Stone crush 10 ft3
number and spacing of columns, Brick ballast 46 ft3
able to future flooding, particu- Mud 218 ft3
the type of construction material, larly as the 2010 floods damaged Bhoosa for mixing mud plaster 80 Kg
and the procurement method for drainage and floods defences. The and roof
bricks. Steel girder 27 ft.
inclusion of DRR principles in shelter (13.5'x 3.5''x7.5'')
designs and mobilisation activities Bamboo (19.5' length 254 ft.
was a strong focus of the project. average dia 2.5")
woven mats (size 19.5x13.5) 263 ft2
Improved disaster - resilient raw straw 108 ft3
Polythene sheet 263 ft2
construction techniques included Galvenised iron spout 2' length 1
raising platforms for shelter con- Wooden door 3'x6' with frame 1
struction, and improving roof 3" x 3"
Wooden window (size 2' x 3', 1
drainage. frame 3"x3")
Wooden door lintel 2
DRR trainings were provided to (3"x4.5"x4.5')
target communities as a whole, not wooden ventilator lintel 2
just direct beneficiaries. Locations Wooden window lintel 4
(3"x4.5"x4.5')
for construction were agreed Bitumen for damp proof course 1 Kg
following hazard mapping by the
community. Cash for work projects
were conducted to repair embank-
ments and some flood defences.
The initial concept was to
combine tree planting, kitchen
gardening and permaculture prin-
ciples to capture waste water and
Compressed mud blocks were improve the village environment
abandoned after the pilot stage due and food security. The extreme
to a lack of community acceptance
summer temperatures and saline
and slow production speed.
Photo: ACTED soil in this part of Sindh, variable
70

Keywords: Non-displaced / returns, Core housing construction, Cash, Training, Guidelines and
Case Study: training materials

Pakistan
Project location: 14 months – – Construction ends
920 villages in south Sindh for phase I
Disaster: – Field work and
trainings start for
2011 floods and intensive rains phase II
Disaster date:
September and October 2011 11 months – – Construction starts
for phase I
destroyed:
750,000–950,000
Project outputs: 7 months – – Field work and
4,624 shelters at end of 2012 - trainings start for
ongoing phase I
55,914 villagers trained 6 months – – Technical training
for implementing
Occupancy rate on handover: partners
100 per cent
Shelter size:
Recommended area 21m2
Materials cost per shelter: Pakistan 2 months – – Vernacular construc-
US$ 300 tion survey finished
1 month – – Project start
US$ 514
September and
October 2011 – – Disaster date
Project description
The organisation worked with 27 implementing partners to deliver shelter at scale. The project provided cash
to households to build their own shelters. It aimed to increase the resilience of communities by increasing the
quality of technical input, incorporating more disaster risk reduction (DRR) components, monitoring to ensure
compliance, and supporting the construction of safer shelters to catalyse self-recovery. This was achieved through
knowledge and cash transfers to enable households to make choices based on their needs and priorities.
Strengths and weaknesses dependent on local markets.

9 The project promoted self-reconstruction and 8 The construction process output was directly
strong beneficiary participation. proportional to the household input so the quality of
9 Distributing cash to beneficiaries stimulated the reconstruction was not uniform across the project.
local economies. Households managed the funds that 8 Not all beneficiaries were interested in complying
they received in instalments. with the recommendations provided, leading to
9 Technical trainings for partners, interested variations in shelter quality.
village members and beneficiaries focused on safer - Before the project the majority of flood affected
construction practices. households had a limited understanding of why their
9 Vernacular shelter typologies were recommended previous shelters failed.
and promoted. These were affordable and maintainable - The vernacular construction typologies differed
by low income families. widely, even within a single village.
9 Created awareness of flood resistance principles - Reintroduction of lime in the traditional construction
using traditional or simple technologies such as lime. process was perceived to be an innovation.
8 Households had to divide their time between daily - Beneficiaries who witnessed the 2011 shelters with
tasks and construction. Implementation dependended improved DRR techniques surviving the 2012 rains were
on crop cycles, leading to delays for donor deadlines. very motivated to learn from the technical trainings,
8 Material quality was variable because it was and to implement the recommendations.
Women plastering their one room shelters. The project provided cash to build shelters and training on safer construction
with incremental improvements such as stronger plinths and footings to walls. Households chose their own shelter design.
Photos: IOM ORS
Before the floods and the social coping mechanisms who was responsible for receiving
Many people in Pakistan could were stretched to the limit. and distributing cash to a group of
be classified as vulnerable before up to 25 households. The organisa-
the floods. 27 per cent of Pakistan’s Villages to intervene in were
tion transferred the first tranche of
population lived in severe poverty selected in phases:
funds as an advance for the benefi-
and 23 per cent lived on less than Phase I was based on the list of ciaries to construct the base of the
US$ 1.25 per day. The flooded most affected union councils house. Once all the members of
region is one of the poorest areas (administrative districts). group had finished their plinths the
in Pakistan. Phase II was based on the list organisation transferred the second
In 2010 there had been major of union councils with the most tranche for the construction of the
flooding and the organisation had unmet needs. walls.
supported households to build over Village committees were set This process of advance
38,000 shelters (see A.24 Shelter up to identify the most vulnerable and milestone construction
Projects 2010). Evaluations of the households among those whose was completed with the third
response indicated that extra action shelters had been completely tranche payment once the roofs
on trainings would enhance the destroyed. Vulnerable households were finished. Joint construction
impact and longevity of disaster risk included: provided positive peer pressure and
reduction interventions. Female-headed households encouraged collaboration.
After the floods Households with no adult male During the entire construction
Monsoon rain in 2011 led to Households with an elderly process, implementing partners
the collapse of many houses due to member (over 60 years old) and project staff provided practical
the weight of waterlogged roofs, Households with a disabled or technical trainings in the villages.
or failed due to foundations being chronically ill member This aimed to ensure that safe
compromised by rising water. An Households with extremely low practices and cost-effective disaster
estimated 1.2 million people were income and no livestock risk reduction techniques were in-
displaced throughout Sindh and Households with a dependency corporated at all stages of the con-
Balochistan provinces, without rate above 60 per cent. struction.
shelter, access to safe drinking Implementation
water, health services or food. Monitoring
The organisation worked with
To monitor the distributed
It was estimated that 35 per 27 implementing partners. Each im-
funds and construction progress
cent of the communities affected plementing partner agreed to build
the project team scrutinised a
in 2011 were also affected by the 500 one-room shelters or support
minimum five per cent of the total
2010 floods. This indicates that 23 villages. Each one had an average
shelters committed. Households
more than a million people affected of four field staff members, two
were chosen to be monitored at
by the 2011 floods had barely social mobilisers and two technical
random from the project benefi-
recovered, or were still trying to assistants. The organisation also
ciary database.
recover from the 2010 flooding. provided its own project staff to
support the implementing partners, The aim was to ensure that the
Beneficiary selection assisting them in the project and on monitoring process was evenly dis-
Working through implement- solving all questions and challenges tributed amongst all groups. Moni-
ing partners and focusing on the that arose. toring plans were devised in such a
most severely affected districts, as- way that they guaranteed a visit to
Village committees, in coor-
sessments were made to identify each village. The process continued
dination with the beneficiaries,
villages where more than 20 per throughout the project, starting
appointed a village focal person
cent of the houses were destroyed from verification of beneficiary
72
A homeowner using lime to prepare stabilised adobe Some homeowners decorated their houses. This was encouraged
bricks for the walls of his house. as a way of building pride in traditional architecture.
Photo: IOM ORS Photo: IOM ORS
selection, to construction oversight free telephone hotline was set up A core component of the
and cash distribution. to record any complaints. Colour project was to train flood affected
posters and business cards with households on how to build back
Coordination key messages and phone numbers safer using disaster risk reduction
The project activities and imple- were distributed. Awareness raising techniques. The objective of the
mentation locations were coordi- sessions were provided to all bene- trainings was to build the resilience
nated with the national interagency ficiaries, implementing partner rep- of affected populations enabling
coordinating body (see the case resentatives and focal points before them to cope with future disasters
study on coordination in Pakistan, the first payment. on their own.
A.21). This reduced duplications
and maximised coverage in line Training Each implementing partner
with agreed priorities. The coordi- In order to identify the most conducted four trainings per village
nation team also supported organi- cost-effective local construction during the construction process.
sations to liaise with the authorities methods, a build-back-better Affected people could also request
and donors, creating a platform for survey on vernacular construction additional trainings. By December
information collection and sharing was conducted in the six priority 2012 the programme had delivered
amongst all shelter actors. districts with support from a local 2,071 trainings that where attended
technical implementing partner (see by over 55,900 villagers.
Regular progress reports were
case study A.23). In September 2012, there
publicly shared. Close coordina-
tion was maintained with the The survey was designed to were two weeks of intense rain.
disaster management authori- record existing conditions in the At the time, all beneficiaries and
ties at the district, provincial and flood-affected districts with a focus implementing partners where very
national levels to ensure a coherent on local self-built techniques. It worried that all the constructions
approach towards shelter recovery. assessed the strengths and weak- were going to be washed away.
nesses of existent vernacular con- However, plinths and platforms
Accountability struction practices. The main aim were minimally damaged and
A complaints telephone hotline was to record the different types of people could repair them, and
was established to encourage structures that survived, the tech- continue construction within a
transparency and to provide ben- niques and practices that largely short time.
eficiaries with a direct link to a withstood the flood waters and the Although not yet quantified,
complaints and feedback process. ones that led to house collapse. there are anecdotal examples of
It also provided beneficiaries, imple-
Once the best construction villages where families who did not
menting partners and project staff
methods were identified and receive the cash grants copied the
a channel to report irregularities
improved they were compiled into construction techniques because
and challenges. Trained staff, fluent
a construction manual used for they had free access to the village
in local languages, responded
practical and theoretical capacity trainings.
to queries and recorded com-
plaints in a database. At the end building trainings for affected
of every week, the complaints were households. The programme also
forwarded to the project manager provided a Training of Trainers
and followed-up by the field teams. course which had theoretical and
practical sessions. It was based
Shelter staff in the field on the construction manual and
informed all beneficiaries and their educated all technical and field staff
implementing partners about the working with the communities.
complaints referral mechanism. A
74

Keywords: Non-displaced, Core housing construction, Housing repair and retrofitting, Training,
Case Study: Guidelines and training materials, Advocacy.

Pakistan
Project location:
Sindh Province
15 months– – Technical Support
Disaster: Programme 17,200
Floods One room shelters
Disaster date: starts
2011 to 2012
Number of houses damaged / September
destroyed: 2012 – – Floods in Sindh
2011: 750,000–950,000
2012: 275,000 10 months – – Model “Eco village”
Project outputs: complete
887 shelters,
Training attended by 55,914
7 months – – Technical support
villagers, 60 artisans and 160 programme 5,500
implementing staff shelters
Changes to national
reconstruction policy
100 per cent
Pakistan
Shelter size: 1 month – – Build Back Safer
16.2m2, 13.8m2, 20.88m2 with Vernacular
Sindh Methodologies –
Project cost per shelter: field survey
This was primarily a training,
assessment and advocacy
project September
2011 – – Rain and floods in
Sindh
Project description
The organisation provided research, training, assessment, design, technical assistance and construction
monitoring and mentoring support to 7,500 households (to an additional 17,500 later) following the 2011 floods.
Based on the organisation’s experience in disaster-affected areas since the 2005 earthquake, the project focused
on developing improved vernacular construction through the use of low-cost sustainable building materials and
training. The organisation provided technical guidance based on its programme “Build Back Safer with Vernacular
Methodologies”, leading to stronger and safer structures that have withstood hazards.
Strengths and weaknesses due to extreme poverty levels and lack of access to
9 The organisation was able to shift national shelter microcredits.
policy by rapidly implementing pilot projects with 8 Lack of funds to promote trained builders into
viable project models. building entrepreneurs or technical advisors for large-
9 The focus was on improved low-cost construction scale self-sustaining shelter programme.
using traditional construction materials and methods, 8 Failure to promote bamboo farming on a large
fostering pride in familiar materials such as mud. scale.
9 Involvement of student volunteers in various stages 8 Challenges in convincing a large number of
of construction and monitoring developed a spirit of other organisations of the efficacy of the project
giving and of unity. methodology.
9 Training in safe eco-building techniques was - Further work is required to improve quality and
provided to NGO personnel, social mobilisers, reach of low-cost technical support.
architects, engineers, students and master artisans. - Disaster Risk Reduction compliant community
8 Lack of testing of stuctural elements due to lack of structures needed to be built in large numbers to
specific funding for the purpose. provide safety of life, water, food, livestock, and
8 Households were unable to enlarge constructions livestock feed etc.
Disaster Risk Reduction components included the use of lime, The project was supported by student volunteers. One of
reinforced plinths and strong accessible roofs. the project aims was to promote traditional construction.
Photo: Sohail Z. Lari, Heritage Foundation Photo: Mariyam Nizam, Heritage Foundation
Before the disaster or did not have access to any other to develop the eight shelter typolo-
The rural communities in Lower land. As a result, shelters were gies which are being constructed in
Sindh province suffered from high mainly rebuilt on old plots. internationally-funded projects (see
levels of illiteracy, lack of access to A.22)
primary healthcare, and were disad- Implementation
vantaged and marginalised. Most The organisation began its 2011 The damage to the different
people worked in fields as tenants response by conducting a survey of types of houses was recorded and
for low wages. There were limited housing typologies and damage. used to analyse the reasons for
other livelihood opportunities. structural failures. These were used
During the assessment phase, to help communities understand
There are major variations in the organisation divided its teams the technical failings of their homes
construction technology, materials, into two groups, a survey group and how to build back safer with
climate and hazards across Pakistan, and a construction group, each with designs developed by eminent ar-
and even between adjacent villages. 12 student volunteers. The survey chitects and engineers associated
team was lead by an experienced with the organisation.
Following the 2010 floods, the
architect, the construction team
organisation provided assistance Many mud walls that had been
was led by the field coordinator.
to over 400 households. Following partially damaged were rehabili-
Teams worked in rotations of three
the 2011 floods, it built on these tated after an engineering review.
weeks in the field after which the
projects to extend their impact. The rehabilitation included repairs
data was compiled and analysed in
the head office by an experienced on walls, bases and plasters, along
After the floods
technical team. with the use of accessible roofs
Following the floods of 2010 and
using bamboo.
2011, the affected communities During the project, more than 80
were in a much worse state. After volunteer students were involved. Demonstration shelters were
two successive years of floods, they Students were mainly from archi- constructed in some affected
had lost all their reserves, and there tectural colleges in their third or villages.
were some signs of aid dependency fourth year of study. Social sciences
among affected households. Construction
students were also involved in some
The initial projects responding
parts of the project. The organisa-
Selection of beneficiaries to the 2011 floods focussed on
tion also engaged professional
Initially villages were chosen on households rehabilitating the walls
architects, horticulturists, artists,
the basis of damage and of existing of their shelters, and the organisa-
textile designers and product devel-
relationships by the organisation tion supporting in the retrofitting
opers for different project stages.
with major landowners and author- of new strong roofs. In the pilot
ities. Later in the project, choice of The organisation normally had project, complete shelters were
location was informed by a detailed up to four people on site, with built for people with disabilities and
housing damage survey. support from its head office. for female-headed households. In
these initial projects, no money was
Depending upon the project Survey directly given to the householders.
location, the organisation worked The organisation assessed ver-
from lists provided by the Provincial nacular architecture, surveying 170 In two locations, the organisa-
Government, on its own assess- homes in 35 tehsils (sub-districts) in tion established workshops to make
ment data or from lists of priority eight priority districts. bamboo joists. In other locations,
needs provided by its donor organi- where different construction
sation. The results were used to develop methods were used, fabrication of
a database of vernacular construc- bamboo roofs was done by local
Most people did not wish to tion typologies in the province of artisans trained by the organisation.
move from their place of origin and/ Sindh. This database was later used
76
After a trial phase, the organi- Members from each household be well cured after application.
sation signed an agreement with attended trainings, by the end of Use salinity-free soil and clean
an international agency to provide which they were able to construct water for stronger construction.
technical support to other organisa- shelters themselves. Use bamboo reinforced lime
tions who were collectively aiming concrete beams and bamboo
The organisation trained 60
to build 22,750 shelters. lintels above openings.
craftsmen in the following con-
Encourage roof projections to
Monitoring construction was struction skills: layout, excava-
improve drainage and to protect
challenging due to lack of internet tions, masonry, bamboo fabrica-
top of mud walls.
and communication, on-site dif- tion, mud-brick making, layered
Use lighter but stronger
ficulties, harsh climate and long mud construction, plastering and
materials for roof construction,
distances. finishing and the use of lime.
such as bamboo joists.
With the increase in scale it was DRR components Maintain a slope on floors and
difficult to check all stages of con- The project had a very strong roofs to drain water.
struction. Each area of work also focus on improving resilience
had its own set of problems that
Project impacts
of communities. As there was The project had significant
had to be dealt with differently. To repeated heavy rain and flooding impacts on the overall national
resolve these challenges, the or- in Sindh, shelter performance was response. By rapidly mobilis-
ganisation worked out a system of monitored. All of the improved ing skilled volunteers to conduct
forms, and a strict set of rules for structures had withstood the technical assessments, the organi-
organisations to control construc- flooding and rains. sation was able to produce accurate
tion and material quality. This was
The methodology became and usable information as well as
overseen by visits from its monitor-
locally known as ‘katcha kot’ or proof of concept pilot projects.
ing and mentoring field teams.
“unfired clay fortress.” The method was rapidly adopted
As it scaled up, the organisation
Technical solutions as a key component of the national
had to increase the number of ar-
The assessment made differing “Pakistan Initial Floods Response
chitects working on-site.
conclusions for different types of Plan” and in grant applications to
Training houses. These became key training donors. Thus, a relatively small but
The training programme messages: experienced technical organisa-
consisted of eight shelter typology tion was able to have a significant
modules for Disaster Risk Reduced Shelters should be constructed impact beyond the scale of its own
construction. The trainings were in a hazard-free location projects.
adapted to the different building wherever possible.
Houses should be orientated Many beneficiaries spent time
methods in each location. to decorate and beautify their new
north-south to reduce heat
Training of trainers sessions for gain. Openings on opposite homes, showing pride and giving
implementing partners were ac- walls improves air circulation. each house an individual character.
companied by mentoring during Houses should have a strong
the construction phases by the or-
Logistics
base to protect the walls. Lime Most materials were procured
ganisation’s technical teams. can stabilise mud walls, plasters locally reducing transportation costs
and renders. Renders need to and stimulating the local economy.
Materials
Below are the key materials for
the different shelters types built.
Materials Quantity
Flat roofs – accessible roofs 18’x10’
Bamboo 628ft.
Lime (50 kg/bags) 350kg
Mud bricks 14,580
Layered mud/adobe 1,170ft3
Loh-khat walls + roof 18’x10'
Bamboo 918ft
Lime 350kg
Mud bricks 47ft3
Loh-khat (reed filling for 464ft2
walls)
Conical Chaura roof Size 16’ diameter
Rehabilitated house with accessible roof which can withstand the weight of
Bamboo 224ft
fifteen people. Such roofs provide an escape location in case of floods. Lime 350kg
Photo: Heritage Foundation Layered mud/adobe 1380ft3
A.24 Peru – 2012 – Flooding and Land Slides

Case Study: Keywords: Unplanned / Unmanaged camps, Household NFIs, Tools, Emergency shelter, Training.

Peru
Project location: 6 weeks – – Project ends.
Central Peru and Lima
Disaster:
Floods and Landslides
Disaster date:
November 2011 to May 2012 1 month – – First family shelter
kits are distributed
Number of people displaced: to families.
November 2011 to May 2012
278,800 people made homeless
409 families
Project outputs: 2 weeks – – Family shelter kits
409 tents and non food item kits arrived in Lima
100 per cent
Shelter size: 4 days– – 224 family shelter
Peru kits requested.
18.5m2
Materials cost per shelter: 1 day – – State of emergency
US$ 280: Tent declared
US$ 455: Non food item kit Lima 17th March
2012 – – Assessment team
Excluding transport and personnel arrives
November
2011 – – Unusually heavy
rains start
Project description
Tents and non-food items were provided to families who had lost their homes as a result of landslides. The
tents and family kits were shipped into the country from international pre-positioning locations in coordination
with the local disaster management authorities. The entire distribution project lasted 6 weeks.
Strengths and weaknesses with limited services.

9 Regionally prepositioned stock overcame the time- 8 Time was lost in resolving national and regional
restraints of procurement lead-times. import regimes.
9 Tents and blankets provided a rapid shelter 8 Materials were imported before clarifying customs
solution that provided protection from the elements. and handling fees. This led to delays and protracted
Fuller construction would have been difficult as the negotiations.
construction season had passed. 8 The small size of the project team restricted the
9 The emergency shelter was distributed with various ability of staff to actively participate with beneficiaries
non food items, including a tool kit. and monitor the response.
9 Families were allocated safer land by the authorities. - Although tents and non food item distribution is not
9 Given the challenges of access, the portability of the sole solution in emergency response, and may not
single-family kits was useful. directly support recovery programming, there are clear
8 Relatively high per household cost for an emergency times when they meet a humanitarian need.
intervention compared to other relief operations in - The organisation relies on small teams of expat
response to the floods. volunteers to oversee distribution so as to keep running
8 The organisation provided tents and non food costs low. However, this combined with the wide
items as emergency shelter, but did not actively geographical area of the distribution left the team little
engage in water sanitation and hygiene promotion or time to train people in how to erect the tent or to use
other needs. There were few other actors involved in the water purification equipment.
the sites leaving a newly formed but small settlements
78
Mud slides came after months of rain. The project provided imported tents and non-food items,
It was cold in the mountainous areas. creating temporary settlements of up to 60 families.
Photo: John Cordell Photo: ShelterBox
Before the floods / The floods most severely dependent validation. Initially,
landslides affected people living in the region selection criteria included particular
Peru is prone to natural of Loreto, which was also one of vulnerabilities (specifically the very
disasters, including droughts, fires, the poorest areas in the country. As young and old, pregnant women
floods, landslides and avalanches, a result, coordination of humanitar- and people with physical and
extreme temperatures and earth- ian activities mainly focused on this mental disability). Healthcare pro-
quakes. region. fessionals were consulted on what
were likely to be the most vulner-
Since 2008, Peru’s economy The landslides affected homes
able groups in specific geographic
has grown at around 9 per cent in the south eastern part of Peru.
areas of need, taking in account
annually, mainly as a result of its In the case of mud slides, both the
climate and altitude.
natural resources. However, there damage and the needs were very
are huge economic disparities localised. However, as more accurate
throughout the country, with more needs assessments were compiled,
The circumstances of the people
than half of the population living it became clear that the organisa-
displaced by the landslides varied.
below the poverty line. tion would not be able to provide
Some stayed with families, others
emergency materials for the entire
The government disaster man- received a (temporary) stipend from
displaced population in the Andean
agement agency, INDECI, consists the government to live in guest
states. At this stage the initial
of a federal body and departmental houses. Others were living in tents
selection criteria were dropped.
bodies that have different amounts of variable quality that were on loan
Distribution was then coordinated
of resources at their disposal, from the government disaster man-
with the elected representatives
dependent on the wealth of their agement agency, or lived under
of camps for displaced people.
department. plastic sheeting that was open
Generally, these representatives
at both ends. The temperatures
Construction practices vary were connected to the project staff
ranged from 3 to 25 degrees centi-
across Peru. In the central highlands, by the local Civil Defence. Occa-
grade depending upon location.
the majority of houses are built sionally the project staff were ap-
from mud blocks with corrugated Following the landslides, many proached directly.
iron roofs. The construction season people slept outside or in simple
Additionally, locally elected relief
is dependent on the amount of makeshift shelters.
representatives produced detailed
water available which is very limited
Selection of beneficiaries lists of people requiring support.
during the dry season.
Working with the local INDECI
Site selection
After the floods / office, the assessment team
Approximately ten percent
landslides decided to provide some support
of the affected households were
La Niña, a macro weather phe- to 173 households in Loreto district
staying with host families. For these
nomenon, combined with local and 409 households affected by
groups, tents had been erected on
systems, caused the Amazon River landslides in the central Peruvian
individual plots of land adjacent to
to reach a historically high level. highlands. The majority of this case
host families’ houses.
This led to a State of Emergency study will refer to the intervention
to be declared on 18th March in in the highlands. For the majority of households,
18 of the 24 departments in Peru. the land that their houses were on
The main selection of locations
This state of emergency was later had been destroyed by the land-
and beneficiaries was through the
extended for an additional 60 days. slides, and new sites had to be iden-
local INDECI office with some in-
Families made modifications and upgrades to their shelters. Left: in the highlands, Right: in the lowlands.
Photo: ShelterBox
tified. These new sites were identi- There were very few other organi- for the import and handling fees.
fied and permanently allocated by sations operating in the area, and In the end, the international or-
the authorities in agreement with no organisations took on sanitation ganisation covered the handling
the hosting villages. These sites or hygiene promotion activities at fees to prevent further delays, and
varied in size, the largest site of one the sites. the disaster management agency
hectare initially housed 76 house- covered all other costs.
holds. Training
As the disaster affected people The shelter kits were brought to
Implementation had never lived in tents before their final locations in 4 wheel drive
The organisation did not have the teams conducted some basic vehicles, or carried by hand. The
any staff in Peru before the disaster. trainings in erecting and maintain- strong packaging made this task
However, a team had travelled to ing tents. The main focus was on much easier.
the country to perform a needs as- tightening guy ropes sufficiently so
Adaptation of tents
sessment in the country in March that the tents did not blow about,
Three months after the distribu-
2011 following reports of flooding. but not tightening them too much
tion, an evaluation found that the
As a result the team was able to so that the tents ripped. There was
majority of families were still living
draw upon pre-existing contacts, also a short training on locating and
in the tents that had been provided.
such as INDECI, who became the orientating tents and the distance
There were no projects in process
key project partner. between them.
to build more durable shelter or
The organisation imported As the organisation also dis- housing.
kits containing emergency tents tributed some water purifica-
Many families had built a kitchen
and household non-food items. It tion equipment and conducted a
and a shaded area as an extension
delivered these to families who had training on its use. However, this
out of timber and tarpaulins. This
been removed from their homes by may not have been fully adequate.
was observed both in the highlands
landslides.
Logistics and supply and in the lowland locations.
The organisation deployed three All items had been transported
consecutive expatriate volunteer Materials list – if relevant
by sea to Panama where they were
teams over the course of six weeks, pre-positioned. From Panama they Materials Quantity
starting the day after the mud were flown to Lima. Within Peru Box 1
slides. The first team of two people Relief tent 1
the boxes were transported by Blankets 5
conducted a needs assessment. This air, truck and/or boat. 69 out of Ground sheets 2
was followed by two four person the boxes that were sent to the Landtrade mosquito nets 2
teams who oversaw the distribu- Kitchen set 1
Loreto region were flown as gifts- Water carriers 2
tions and trainings on the erection in-kind. Transport for staff was paid Family water filtration unit 1
and maintenance of tents. Three for through gifts-in-kind. Toolbag with hammer saw, pliers 1
months after the end of the project, and rope
a monitoring and evaluation team INDECI offices at the depart-
travelled to the affected region. mental level acted as the consignee
for bringing tents and shelter kits
In the newly established sites, into the country. This created sig-
the disaster management authority nificant delays as negotiation was
provided some clean water, con- required between federal and local
tainers and some very basic latrines. offices to agree who would pay
80
A.25 Philippines – 2011 – Cyclone

Overview:
Summary
In late 2011, over 39,000 houses were damaged and over 400,000
people were displaced by winds, floods and landslides following tropical
storm Washi (also known as Sendong). Collective centres were established
and non-food items were distributed in the first phase of the response.
After the emergency phase of response, transitional sites were
established and programming shifted to include reconstruction on newly
identified relocation sites (see A.27), transitional shelter programming in
existing urban areas (see A.26), and repair and rehabilitation of damaged
houses. After one year, 7,800 people remained in 38 different evacuation
centres.
Background After the cyclone established in northern Mindanao

The Philippines is a middle-in- Tropical storm Washi, (also by the Office of Civil Defence. It
come country, with a well-educated known as Sendong), hit the worked closely with international
population and engaged local and Mindanao region of the Philip- organisations, and established co-
national authorities. The Philippines pines from the 16th to the 18th of ordination groups for shelter, camp
regularly faces natural disasters December 2011. The storm brought management coordination and for
and the country has had previous strong winds and heavy rain that led non-food items.
experience of coordination with to flash floods, landslides and pro- Approximately three quarters
the cluster system. This helped to tracted flooding. 624,600 people of those people affected by the
manage the response efficiently. were affected, 430,000 people storm lived at or below the poverty
were displaced and 39,000 houses line with limited means for self-
Many low income families had
were damaged or destroyed. The recovery. Of the partially damaged
settled in particularly vulnerable
primary impacts were in Cagayan houses, nearly half had no struc-
locations on river banks and other
de Oro City and Iligan City. tural damage but needed to be
marginal land. In large parts of
Mindanao there had not been any In the immediate aftermath of cleaned before families could move
major disasters in recent memory. the storm, people found shelter back in.
in evacuation centres, with host Two months after the storm,
In rural areas, families commonly
families, in rented accommodation, moderate to heavy rains fell over
lived in amakan type shelters (with
in makeshift shelters at the site of parts of Mindanao and Visayas
woven bamboo walls) with frames
destroyed houses or in damaged islands, triggering some flooding
made from bamboo and other
houses. and landslides. Although no
varieties of wood.
The government immediate- flooding was reported in the areas
For urban areas, people living
ly mounted a major emergency affected by the tropical storm, the
at or below poverty line, lived in a
rescue, evacuation and response rain worsened the conditions in
mixture of raggedly constructed
operation. Coordination was rapidly temporary shelters.
shanties and semi-concrete houses.
Before the cyclone, many families were living in locations that were vulnerable to storms and flooding, but that had access
to livelihoods. The government declared that some of these were “no build” zones, and new sites had to be identified.
Photo: Wan Sophonpanich
The government established a

reconstruction policy that included:
the establishment of no build
zones
permanent housing
material supplies
site upgrading for informal
settler families
housing loans for families in
Heavy rain caused over 400,000 people to be displaced. Most people made formal settlement sites.
temporary repairs to their houses or moved in with host families.
In practice, the only no-build
Photos: Anna Pont
zones that were officially declared
Evacuation centres Host families were in Isla de Oro and Cala-cala.
A total of 119 evacuation Despite the early focus of relief These highly damaged settlements
centres were established, housing activities on collective centres and were directly in the path of the
100,000 people (20,000 families). the comparative ease of deliver- river. No official declaration was
Initial response mainly focussed ing large scale assistance to these made regarding other high risk and
on meeting the needs of people centralised sites, the majority of the medium risk areas.
in these often crowded evacuation affected population found accom-
centres. Camp management com- modation with host families. After Land
mittees were established in many of 2 months, 260,000 people were One of the major constraints
the sites. living with host families. The main in the provision of temporary and
support that these families received permanent shelter was the lack
By the end of 2012 many evacu- of available land. Identifying land
was through emergency distribu-
ation centres had closed, leaving and preparing transitional and
tion.
7,800 people (1,700 families) in 38 permanent relocation sites took
evacuation centres. Recovery many months.
An interagency shelter assess-
Tented camps
ment based on secondary data
Some tented camps were es-
sources was conducted within the
tablished to decongest some of
first month of the storm, but took
the most overcrowded evacuation
some time to be finally published. It
centres, and to provide shelter for
provided numbers of damaged and
people living in evacuation centres
destroyed houses that were used as
which needed to be returned to
planning figures.
their previous use (such as schools).
Following these results, the
Transitional sites and shelter organisations collectively
Relocation sites agreed to prioritise support to the
Where temporarily available most vulnerable 65 per cent of
land could be found, transitional people whose houses had been lost
sites were established as a more or damaged:
durable solution to camps (See
A.26). families/occupants of the
13,850 structurally damaged Camps were established for people
When land for construction houses who were at or below living in closing or overcrowded
could be negotiated on a long evacuation centres.
the poverty line Some of the camps were very dense.
term basis, relocation sites were families from all the 11,427 Photo: Anna Pont
established (See A.27). After four totally destroyed houses.
months, seven relocation projects
were underway, with a planned
capacity of nearly 6,000 houses for
households whose land was unsafe.
By the end of 2012, nine
permanent relocation sites had
been established by the local gov-
ernment working with NGOs.
3,147 shelters were complete,
2,943 of which were handed over.
Some transitional sites were established as
359 more permanent shelters were more durable solutions than camps.
being built. Photo: Anna Pont
82

Case Study:

The Philippines
12 months – – 1,823 t-shelters
Project location: completed
Mindanao
Disaster:
6 months – – 675 t-shelters in
Tropical Storm Washi (Sendong)
relocation sites,
Disaster date: – 194 on-site
December 16th 2011
destroyed:
39,000 3 months – – 8,000 cash for
work days worked
30 per cent of the 600,000
population of Cagayan de Oro
City 2 months – – First transitional
settlement centres
Project outputs:
occupied
30 transitional settlement sites
with services Philippines – Provided water to
1,823 t-shelters 10,000 people in
evacuation centres,
distributed over
92 per cent Mindanao 2,000 WASH kits
Shelter size:
18m2 for family of five 3 weeks – – Project start
US$ 410 for relocation sites 16th December
US$ 550 for on-site 2011 – – Disaster date
construction.
Project description
The organisation implemented an urban transitional settlement programme building 1,823 transitional
shelters. Many complex issues arose, including land and property rights, zoning issues, high-risk settlements and
providing shelter solutions to those without land rights. This programme demonstrated the importance of and
challenges to acquiring land for transitional settlements.
Strengths and weaknesses on-site shelter support. Additional targeting criteria

9 The transitional shelter (t-shelter) design cost and stricter decision-making timeframes would have
US$ 410, including labour. This was cheaper than improved beneficiary selection.
emergency tents (US$ 800-1,000, including airfreight). 8 The project was unable to support some of the
9 The t-shelter design and was inspired by the local most vulnerable affected populations, notably people
vernacular architecture. Shelters could be maintained in ‘high-risk zones’ (due to official objections) and
and materials could be re-used. people with ambiguous land tenure.
9 The integration of WASH and shelter was 8 An alternative shelter design for people with
emphasised from the beginning of the program. disabilities should have been developed.
9 The agency put a great deal of effort into persuading - An ill-defined ‘no-build zone’ policy created
land owners to release their land. challenges. A number of landowners remained in
9 The agency successfully negotiated the free ‘limbo’ because their homes were within no-build
installation and use of water and electricity for two zones, and new land was not allocated.
months for 7 relocation sites. - Different stakeholders, such as the church and local
8 There were questions around how disaster-resistant government, had different approaches to beneficiary
the t-shelter design was. selection and prioritisation.
8 The organisation would have benefitted from hiring - Some affected households refused to move into
a liaison officer to better understand the political a transitional settlement because they thought this
system and accelerate the project. would impact on their right to promised permanent
8 There were difficulties in verifying beneficiaries for housing.
Before the cyclone

(See overview A.25 for back-
ground.)
Until 2011, there had been no
major floods in the area since the
1950s. The population of Cagayan
de Oro had spread along risk
areas, such as river banks and delta
areas. In Macasandig, one of the
most affected areas, there was a
mix of commercial and residential
buildings. Residents ranged from
poor in shanty areas to middle-class
in apartment buildings.
Despite the well-developed local
administration, the complexities
of addressing housing, land and Emergency shelters such as schools and gymnasiums quickly became
property issues in an urban transi- overcrowded in the aftermath of the storm.
tional response presented real chal- Photo: CRS/S.Hirano
lenges in supporting the most vul-
nerable. risk of flooding or landslide Selection of beneficiaries
access to roads Relocation
After the cyclone access to water (either There were only two organisa-
The flash floods caused by groundwater or pipe tions who responded with transi-
Tropical Storm Washi destroyed a connection) and electricity tional shelter projects in the Phil-
large portion of the city centre of costs of travelling into the ippines. As a result, there was
Cagayan de Oro. Macasandig and city from the site were not considerable pressure from gov-
Isla de Oro were the worst affected prohibitively expensive for ernment officials, church leaders,
urban barangays (the smallest ad- beneficiaries camp managers and other NGOs to
ministrative boundary, equivalent the proximity of public facilities prioritise certain evacuation centres
to a village). such as schools, health centers or specific beneficiaries.
Poor families residing in and markets.
The government prioritised
makeshift shelters by the river Different types of agreement closing evacuation centers and tent
banks suffered the most. Many were required with different land- cities before assisting community-
middle-class households who owners. In most sites, there was based IDPs as the evacuation centres
rented or owned apartments were a guarantee that land would be were costly and water and sanitation
also affected. returned to owner. Overall 30 sites services were over-stretched.
As the emergency response were established. Meanwhile, organisations working
unfolded, the government on education issues advocated
The types of agreement are
launched their permanent housing for emptying schools to address
summarised in the table below.
programme. The agency proposed protection concerns associated with
a two-tier transitional shelter owner type of agreement endorsed having displaced people living on
programme to plug the gap by school grounds.
between emergency shelter and City Verbal agreement for Mayor
temporary use. Other Families who wanted to return
permanent housing. conditions included to their places of origin were given
requests for certain lowest priority on the permanent
Land Acquisition shelter recipients or, in
one case, early closure housing waiting list.
The following criteria were used
of the site in order for
to verify the suitability of land: the land to be used for The organisation faced the chal-
permanent shelter. lenges of determining whether
clarity of land ownership
Private Written MoA between Landowner informal settlers had really lost
land is donated rent-free for up the Archdiocese of their homes in the storm. There
to 2 years Cagayan de Oro and the
landowner with terms were some cases of ‘opportunists’
land owner clearly understands
and conditions. trying to use the system to receive
the purpose and the nature of
Church Verbal agreement after Archbishop a shelter although their home
transitional settlements request of Archbishop. remained intact.
land is well drained and is not at
84
Transitional shelters could be placed

Transitional shelters could be relocated. on available plots of land.
Photo: Charisse Mae Borja / CRS Photo: Seki Hirano / CRS
The organisation aimed to retain Implementation DRR components

community social structures as far To address the range of needs Drainage, sewage channels and
as possible when relocating benefi- the agency offered two transitional other essential infrastructure were
ciaries in the most affected areas. shelter options: construction on provided where necessary. This was
This was not always possible due either the original site or in one of to ensure the protection of both
to variations in site location, timing 15 relocation sites. the people living on the land and
of response, and the number of the land itself.
shelters available on each site. Transitional shelter design
Transitional shelters erected On-site transitional shelters
On site Construction on relocation sites needed to were constructed using a reinforced
Affected households whose be moveable and make minimal concrete foundation enabling the
houses had been totally destroyed, impact on the land. shelter to be securely anchored,
and who lived in low to medium preventing it from being upturned
risk zones, were offered flood- The agency worked with a local
by flood or strong winds.
resistant transitional shelters sited architect and local engineers to
in their original neighbourhood. design an adaptation of the tradi- The design featured a raised
Water and Sanitation facilities tional Amakan (bamboo or palm floor to provide flood protection,
were organised within community leaf weave) house. facilitate ventilation and to keep
groups and elevated septic tanks out vermin.
Amakan houses have been built
were constructed. for centuries and are well adapted Logistics
Informal settlers were often to the tropical climate of the Phil- Drying timber and limited road
without official land or house ippines. They can also easily be access were the biggest logistical
tenure papers. This meant it was repaired or rebuilt. The design used issues, affecting delivery time and
difficult to confirm whether they locally available amakan (palm was costs. One truck could carry enough
had lost their home during Washi used) for the walls and coco lumber, timber for 28 transitional shelters,
or if they had lived elsewhere. which is durable and inexpensive, meaning that over 75 truckloads of
for the structural frames. timber were required for the whole
To identify households for
The design was based on the project.
on-site rebuilding, the organisation
conducted a community mapping following design criteria: Materials list
process. This involved visiting Culturally appropriate: Provides Materials Quantity
former housing locations, verifying privacy, uses local materials and Portland cement(40kg) 5 bags
the damage to houses, verifying the provides protection from rain Mixed gravel 1 bags
lack of shelter, interviewing neigh- 10mmx6.0m re-bar 12m
and heat 8mmx6.0m re-bar 3m
bours and verifying lists of names Relocatable: Can be carried by Coco Lumber 4”x4”x12’ 64 ft.
with ward leaders and community 20 persons or easily dismantled Coco Lumber 2”x3”x12’ 128 ft.
leaders. This ward specific approach Coco Lumber 2”x4”x8’ 128 ft.
Speed of construction: Can be Coco Lumber 2”x2”x8’ 75 ft.
was taken helped to retain the built in 2-3 days Coco Lumber 2”x4”x8’ 32 ft.
community structure. Economical 2” umbrella nails 1kg
Bamboo slats 3 bundle
It was challenging to identify Flexible: Design can be adjusted Nails 9kg
those most in need. As time passed, for relocated families or those Plywood ¾”x4”x8” 6 sheets
returning to original sites Plywood 3/16”x4’x8’ 6 sheets
a number of people had begun re- Amakan 4’x8’ 13 sheets
building, making it difficult to verify Upgradeable: Can be upgraded Sealant 1 pint
the original level of damage. to a permanent home.

Case Study: Keywords: Resettlement, Household NFIs, Construction materials, Core housing construction,
Housing repair and retrofitting, Site planning, Infrastructure, Training.

The Philippines
Project location:
Mindanao
Disaster:
11 months– – 70 per cent of
Tropical Storm Washi (Sendong) planned 6,000
Disaster date: houses completed
16th December 2011
destroyed:
39,000
Project outputs:
5,000 emergency shelter kits
6,000 permanent core houses
(90 per cent complete)
70 per cent occupancy
Shelter size:
21m2 - permanent core house 4 months – – First 500 perma-
Philippines nent core houses
Materials cost per shelter: constructed
US$ 50: emergency shelter kit Mindanao
US$ 2,750: permanent core
house 1 month – – Distribution of non-
food items starts
Project cost per shelter: 2 weeks – (5000 HH)
US$ 3,100
16th December
Project description
The organisation distributed 5,000 shelter repair kits and built 6,000 housing units for displaced families. It
built the houses with services on new relocation sites using contractors, volunteers and working with partners.
It deployed three construction mobilisation units for the repair and restoration of houses and communities
damaged by the storm.
Strengths and weaknesses negative impact in the project and the local economy.
9 Good relations were established with local 8 Relocation introduced the need to develop new
authorities. As a result, land for relocation sites and networks and community relations among the
resources for site development were readily available relocated population. These activities had very little
from the authorities. funding support from the project.
9 Quick development of family selection criteria and 8 Delays among other organisations providing
process. As a result, displaced families could be offered infrastrcuture and services to the sites meant that only
a clear path to recovery in a relatively short time. 70 per cent of the houses were occupied by the end
9 Good management of construction activities in of 2012.
multiple sites with a variety of contractors contributing - Strong coordination with other organisations
to a steady delivery of permanent shelter. through national coordination and local interagency
9 The project has allowed the development of block- group meetings was needed to avoid duplication of
making, welding and carpentry skills among the material distributions. Several organisations provided
affected populations. similar products, such as repair kits.
8 Due to limited availability of local construction - At the end of 2012, Typhoon Bopha (Pablo) hit
materials and high prices, advance scouting became Mindanao. Previously, Mindanao was seldom hit by
necessary to order from suppliers. This created some cyclones and typhoons, as a result preparedness was
backlog in implementation. lower than elsewhere.
8 Price hikes of 30 per cent and more created a
86
The organisation rapidly completed 70 per cent of a planned 6,000 houses within 11 months of the storm on safer
permanent relocation sites.
Photo: Mikel Flamm
Before the cyclone The government issued a decree Implementation

See Section A.25 for back- to prevent re-settlement and recon- The organisation distributed
ground. struction of houses in some high 5,000 emergency shelter kits con-
risk areas. As a result, families were taining construction materials
Families were settled along the
displaced into camps set up by the (timber, corrugated galvanised
river banks of the Cagayan de Oro
local authorities and international sheets, nails, etc.) and basic tools
river and other minor streams in
humanitarian organisations. to support emergency repairs on
northern Mindanao. The locations
The Government of the Philip- damaged homes.
are extremely hazardous and in
high-risk for flash floods. While pines made an early decision after Staff made an initial damage
being high risk areas, these locations the disaster to relocate affected assessment in affected neighbour-
were well located economically, families who had been living in the hoods and issued vouchers. The
being near the cities’ commercial river banks of the Cagayan de Oro distribution was made out of a
districts where most families found river. Their homes were completely centrally located warehouse.
support for their livelihoods. washed away by the floods.
In coordination with local and
After the cyclone Local government entities national authorities, the organi-
Rain from the severe tropical provided land for temporary sation conducted assessments
storm Washi (Sendong) created camps in the outskirts of cities, and planned to construct 6,000
flash floods. Most houses located to accommodate the displaced permanent shelters in 10 relocation
by the river banks were completely until permanent shelter could be sites in Cagayan de Oro City and
destroyed. Homes in safer locations secured. Iligan.
were damaged by high winds. Government agencies provided
land from pre-existing land banks
and facilitated planning resources
and heavy machinery for site devel-
opment. The organisation was put
in charge of overall programme co-
ordination and the construction of
the permanent shelters.
Selection of beneficiaries
The Philippines’ natioanl Depart-
ment of Social Welfare and Develop-
ment, conducted a thorough survey
and census of affected families.
It used this to determine eligibility
for assistance and shelter support.
Families prevented from resettling
in high risk areas were placed in
tented camps and selected for re-
location to the nearest site where
permanent shelter was being built.
New relocation sites were planned in locations with lower cyclone risk.
Photo: Mikel Flamm
Non-food items and housing repair kits were Construction was implemented using contractors, vol-
distributed to 5,000 households. unteers and by working with partner organisations.
Photo: Leonilo Escalada Photo: Mikel Flamm
Implementation The permanent core houses were

The organisation used 22 small structurally designed by engineers, “At the beginning, we were
construction groups as external incorporating strapping and rein- doubtful we could be in a
contractors. These worked in com- forcements and were approved by permanent house so soon
bination with its own staff, volun- the relevant authorities. The sites after Washi. We are happy
teers and implementing partneror- were provided with drainage infra-
that we could move out of
ganisations. structure and roads, and walkways
were built to manage erosion. the tent into a permanent
Family participation in project house.”
activities was limited to unskilled Before families moved into
tasks and attendance to skills de- their new homes, as part of the
velopment training (carpentry, induction to the new settlements,
welding, and concrete block-mak- they received an initial training
ing). induction on disaster preparedness. A new housholder at the
This was coordinated with the local Calaanan site, Cagayan de
Coordination emergency management agency. Oro City
From the beginning of the
response, it became clear that Technical solutions
there would be a division of labour The core house was built from Logistics
between humanitarian organisa- concrete blocks, with a reinforced On account of its scale, the
tions responding to the disaster. masonry design. There were steel project presented many logistical
reinforcement bars, both verti- hurdles related to the supply of
While some organisations
cally and horizontally. The roof construction materials.
invested efforts in tents and transi-
structure was made of metal trusses
tional shelter in camp settings, this The organisation purchased
and purlins, with a cover of zinc/
organisation was keen to embark cement, reinforcement bar and
aluminium sheeting. Doors and
on a permanent shelter construc- other materials in bulk to minimize
windows used metal frames, and
tion programme to allow for the the price rises following the disaster.
the floor was covered with ceramic
next stage in the recovery. Coordi- These materials were then distribut-
tiles.
nation was key in helping to clearly ed to contractors as required by the
define these roles, and to provide a Each shelter unit had a multiple progress of construction.
pathway to permanent shelter for purpose room, an attached sanitary
affected families. unit (toilet and bath area) and a The project benefitted from
small kitchen area. The height of skilled and experienced manage-
DRR components the buildings allowed a mezzanine rial staff coming from the organi-
The different relocation sites level to be built by occupants to sation’s central office in Manila, as
were located in low-risk areas, with create a raised sleeping area. This well as newly hired staff.
reduced natural threats. These relo- could potentially increase the living
cation sites were safer than families’ space from 21m2 to 36m2.
original plots by the river.
88
A.28 Somalia – 2011 – Famine / Conflict

Keywords: Urban neighbourhoods, Household NFIs, Construction materials, Transitional shelter,
Update: Site planning, Infrastructure, Coordination.

Somalia
Project location:
Mogadishu
14 months – – Construction phase
Conflict / disaster:
July 2011 Famine and
Continuing conflict 12 months – – Mapping of Zona K
11 months – – Dedicated Tri-
Number of people displaced: Cluster Coordinator
200,000 IDPs in Mogadishu arrives
Project target population: 10 months – – Selection of 16
Approximately 36,000 projects to be
implemented by 14
Project outputs: agencies
3,645 housing units complete
WASH and health facilities 8 months – – Strategy formed and
dedicated coordina-
Occupancy rate on handover: tor agreed
100 per cent - November 2012 7 months – – First funds allocated
Shelter size:
15.8m2 (3.6m x 4.4m) – Non-food items and
shelter kits distribu-
Materials cost per household: tion
US$ 420 July 2011 – – Famine declared
Somalia
Early 2011 – – Displaced people

start to arrive
Mogadishu
Project description
The Tri-Cluster project is a coordinated group of 16 projects implemented by 14 partners across the sectors of
shelter, WASH and health. Zona K in Magadishu was chosen as the target area as it had the densest concentration
of IDPs and was the least likely IDP settlement to be evicted once Mogadishu stabilised and developed. The
project goal was to improve the protection for displaced people living in Zona K through improved settlement
planning and the provision of integrated services from multiple sectors.
Strengths and weaknesses fact that many people were already settled within the
9 Regular coordination meetings achieved a common settlement meant that it was not always possible to
understanding of aims and objectives amongst all follow site plans and meet minimum standards.
partners. 8 Communal spaces have been eroded by an increase
9 By integrating services the project was able to act in the numbers of people living in Zona K.
more efficiently to provide shelter, access to water and - As the sectors work at different levels (shelter
sanitation and basic health serives. with households, WASH with groups of five families
9 Settlement planning has enabled organisations per latrine and health with the whole community)
to have better access and the beneficiaries have an synchronising activities required complex work plans.
enhanced sense of community. Displaced people - Mapping all the stakeholders in the process was
were involved in the development of context-specific difficult, and their influence changed over time.
planning standards which helped manage expectations. - The project had a high profile, putting implementing
8 Underestimation of the impact of other projects partners under pressure to produce results quickly,
funded through other sources active in the same compromising planning and construction quality.
project area. - The Tri-Cluster coordinator took on many of the
8 Although eviction is unlikely in the short-term, there camp management and camp coordindation duties.
is no clear ownership of land and so displaced people
are vulnerable to the Somali ‘gatekeepers’.
8 A weak community structure combined with the
An urban area of Mogadishu was re-planned and many organisations worked together working in three sectors of interven-
tion. These shelters have been upgraded by inhabitants who have built their own external shaded and cooking areas.
Photo: Richard Evans
Before the displacement 70,000 IDPs living in make-shift Cluster advocated that organisa-
Mogadishu has hosted displaced shelters called buuls (traditional tions should provide humanitarian
people from conflicts since 1991. Somali thatched shelter). These assistance to the locations where
However, as drought worsened in were constructed by the IDPs them- IDPs had self-settled. This has been
late 2010 and famine approached selves from scavenged materials the approach in Somaliland and
in early 2011, more and more and items received from humanitar- Puntland where the conditions and
Somalis were driven away from ian organisations. access are more favourable.
rural areas to Mogadishu looking Any attempt to coordinate set- The mechanics that control
for assistance and safety. tlements in Mogadishu would the creation of new camps were
Displacement was compounded have directly interfered with the deemed too complex and unpre-
by the ongoing conflict in Somalia. economic relationship between the dictable to encourage new sites.
host population and the IDPs. As a
After the displacement result, no formal camp coordination Implementation
Upon arrival in Mogadishu, the mechanism was established. Under the umbrella of the Tri-
Internally Displaced Persons (IDPs) Cluster there were five shelter
settled on any unoccupied land. This As a response to the influx of projects, with a total value of US$
process of self-settlement meant IDPs into Mogadishu, a three-phase 4 million.
that there was no site planning. strategy was developed in July
2011: The first project focused on
Services such as water and sani- mapping the existing settlement,
tation, and access to the 100 or Provide all displaced people producing settlement plans, and
so settlements were sporadic. As with a non-food item packages creating access roads and storm
the number of sites closer to the Provide transitional shelter drainage.
centre of town reduced and as Al- solutions
Shabaab’s influence lessened, many Provide site planning to improve This mapping was followed by
IDPs settled into the area which living conditions and access consultations with the beneficiary
became known as Zona K. to other basic services such as community and landowners to
WASH and health. ensure that people would not be
Zona K’s mixed ownership, evicted once work was completed.
between the government, the uni- The shelter coordination did not
versity and some private individuals, advocate the creation of new set- One organisation chose to work
meant that it was one of the least tlements for the IDPs. This strategy through long-standing partner or-
likely sites to be evicted. By the end was attempted in Puntland (see A.8 ganisations while the other con-
of 2012, the site covered an area in Shelter Projects 2008) but was tracted the work to local construc-
of over 3km2 with an estimated not very successful. Instead, the tion companies.
90
Where possible the implement-

ing organisations followed the site
plans, but they were often forced to
deviate from them. Reasons for this
included the need to accommodate
new demands from stakeholders,
the construction of new permanent
structures that had been built after
the initial mapping, and the need to
accommodate a larger population.
Once the shelters were
completed, two local organisations
provided non-food items, including
blankets, kitchen sets, jerrycans and
fuel-efficient stoves. Beneficiary
lists were provided by the main
shelter partners,and distributions
were undertaken once the shelters
The project integrated shelter WASH, health and site planning.
were handed over.
Photo: Richard Evans
Selection of beneficiaries Coordination and communi- Once the framework and
The whole area was sub-divid- cation was needed with the local common understanding on coor-
ed into 25 zones, and settlement authorities ensured that they were dination was created, it became
planning was based on the displaced aware of the project and its impli- feasible to add additional sectors
population at the time of mapping. cations, and that they approved the and projects.
The two main organisations started temporary development plans. As
in different zones and completed all The Tri-Cluster approach came
the final shelter solution was semi-
the construction before moving on about because the Humanitar-
permanent (5 to 10 year lifespan),
to the next. Every IDP that was reg- ian Coordinator considered that
the urban planning undertaken as
istered received a shelter and non shelter, WASH and health were the
part of the Tri-Cluster, will influence
-food item kits. The other Tri-Clus- most pressing needs for the IDPs.
the development of this part of the
ter partners provided sanitation and At the time there was surprise that
city. Access roads created now,
water points in the locations identi- other sectors were not also included
will be the main roads for years to
fied during the planning process. in a multi-sectorial approach.
come.
However, the coordination of just
Coordination Technical solutions three sectors was difficult enough,
Effective coordination was The shelter actors worked with and in retrospect the presence of
crucial for success, as there were 16 the main partners to identify a additional partners and targets may
projects operating in a very concen- unified shelter typology. Initially, have reduced the effectiveness of
trated area. In addition, there were US$ 80 shelter kits were planned the entire intervention.
many actors who were already as the land tenure was not known. Generally, once an organisa-
working in Zona K. Therefore, a Later, a ‘hybrid’ between plastic tion secured funding, the focus was
dedicated Tri-Cluster coordinator sheeting and corrugated galvanised immediately on implementing as
was brought in to act as a focal iron (CGI) was adopted during quickly as possible in order to meet
point for the 16 projects. the planning stage. This provided project targets. To combat this
Initially there was reluctance a better quality shelter while also “tunnel vision” amongst organisa-
from some of the implementing keeping a light footprint. The tions, the successful multi-agency
partners to work under the same design was developed further just approach invested heavily in com-
umbrella. The WASH and health before the construction phase into munication and consultation. This
partners did not want to wait for the a full corrugated iron model, partly always takes time.
mapping process to be completed, due to donors and partly due to
protection concerns. Starting with just 3 sectors
and wanted to implement projects
enabled a culture of coordination to
immediately, regardless of the
Future be ingrained. Only once the coordi-
output from the planning phase.
The Tri-Cluster project was nation was working with a few key
Over a series of meetings, the expanded for 2013 to include partners was it possible to expand
importance of coordinating activi- education and protection focused to the full array of humanitarian
ties was emphasised and a plan was projects. It was planned for an ad- services.
developed where some activities ditional shelter agency to join the
could be carried out at the same existing two partners, and 3,000-
time as the mapping. 4,000 more shelters were planned.
Site planning for the urban areas of Mogadishu. Different potential plans were shared with focal groups. In the end, row planning was chosen
because92people could understand it better and could clearly mark the extent of their ‘land’. This would make it easier for people to know what
belonged to them and help to avoid conflicts.
A.29 Republic of South Sudan – 2011 – Conflict

Keywords: Returns, Resettlement, Construction materials, Core housing construction, Site plan-
Update: ning, Infrastructure, Training.

Republic of South Sudan
Conflict: 17 months – – Project completion
Post-war reconstruction – 8,300 shelters
16 months –
Conflict date: completed
1983 to 2005
2,000,000
Project target population: 13 months – – Compressed mud
70,000 (includes beneficiaries of blocks introduced
12 months– – 6,800 shelters
quick impact projects) complete
Project outputs:
8,300 shelters
2,200: Compressed mud blocks
8 months – – First construction
6,100: Bamboo / wattle and daub phase
95 per cent
Shelter size: – Procurement and
5 months –
16 m2 - up to four people delivery of shelter
24 m2 - five people or more Republic of building material
Materials cost per shelter: South Sudan 3 months – – Agreement with
Republic of
partners
US$ 400 - 600: poles and bamboo South Sudan
US$ 800 - 1100: compressed mud March 2011 – – Project start
blocks
Labour: US$ 260
Average: US$ 1,100
US$ 600-1,200 2005 – – Peace agreement
Project description
The project supported reintegration of returnees. It constructed 8,300 shelters on new land plots provided by
the government. Basic urban services such as school buildings and boreholes, were constructed through parallel
programmes. Two shelter designs were employed: bamboo and thatched-roof shelters (6,800) that could be
built quickly to respond to large-scale returns and compressed mud block shelters with CGI sheet roofs (1,500)
to provide more durable structures.
Strengths and weaknesses cent due to rising costs and delays in block production.
9 Communities participated in the selection of 8 Construction using compressed mud blocks required
vulnerable households and in designing shelters. a highly-skilled lead builder. In some early cases, skills
9 Good coordination prevented returnees from being were lacking and build quality was poor.
sited in areas too far from transport or services. 8 Due to unexpectedly slow block production, the
9 Shelter construction was linked to projects to deliver number of mud block shelters was cut by 800.
basic services and livelihood opportunities. 8 Plans to use drainage activities to supply the mud
9 The project was able to respond to input from required for blocks failed due to the lack of organisation
authorities and change the shelter design. at the community level.
9 Training of affected populations improved their 8 The project was too big and created unsustainable
construction skills. demands for materials, leading to concerns over the
9 Partners were required to submit phased progress destruction of national forests.
reports for each household to keep the project on - Compressed mud-blocks needed to be plastered
schedule. with burnt oil, sandy soil and Arabic gum.
8 Communities demanded incentives for their - As the compressed mud-block technique was new
involvement in the construction phase. to some areas, its performance over time remains
8 The target number of shelters was reduced by 35 per untested.
A.29 Conflict
Approximately 25 per cent of the shelters were built using compressed soil blocks. The technique represented a financial and
environmental improvement, but was slower, requiring significant efforts to introduce as a new technology.
Photo: Fernando Murillo
Before the conflict The project was coordinated by Beneficiary lists were then verified
South Sudan became independ- an international agency (with two by the main agency’s field staff.
ent from the Republic of the Sudan technical and two administrative
The beneficiaries came mostly
in 2011 following two civil wars staff), and implemented by partner
from the returnee community but
(1955-1972 and 1983-2005). The NGOs and community organisa-
10 per cent of shelters were con-
Comprehensive Peace Agreement tions. Construction teams were
structed for families from the host
in 2005 signalled the beginning of made up of nine people, including
community.
a return process. engineers, construction supervisors,
masons and carpenters. Associated projects such as
In 2011 Sudan, (north and south borehole and school construction
combined) had an Human Devel- Materials were procured by
benefitted both groups. Land allo-
opment index of 0.408 placing it the main agency on behalf of the
cation was made through a govern-
in the “low human development” partners. The materials were dis-
ment lottery process.
group. South Sudan is relatively tributed as self-construction kits.
less developed than the north and Experienced masons and carpenters Households with special needs
faces considerable challenges in were identified to provide “on-the- had their veranda, kitchen or oven
terms of infrastructure develop- job” construction training for young built for them.
ment and poverty reduction, with people from both the returnee and
many people unable to access social host communities. Coordination
services or education. Coordination was critical since
Construction progress was so many actors were involved. The
After the conflict monitored by giving each shelter coordinating agency not only had
The conflict between The one of four statuses: to ensure coordination within the
Republic of Sudan and South Sudan To be done: Beneficiaries not project in terms of working with
stunted development in the South yet identified implementing partners but also had
and most returnees had no shelter In progress: Beneficiaries to work closely with national and
or land to return to. identified and land title received state authorities who were devel-
Under construction: Structure oping their planning and building
2011 marked the peak in return regulations from scratch. Despite
as it coincided with the deadline and roofing complete
Finished: Beneficiaries have many delays the land allocation was
for southern Sudanese to leave completed in time for the shelters
Khartoum, where the majority of moved in.
to be constructed.
IDPs had fled to during the war. Selection of beneficiaries
There was also a significant return Project areas were determined Beneficiaries and host communi-
of the diaspora in neighbouring by the agency in collaboration ties were also involved in prioritising
countries, Europe and the USA. with the Ministry of Humanitarian the type of quick-impact projects to
Affairs. The shelters were distrib- be implemented.
Implementation
uted according to the proportion of Hazards
The project built 8,300 shelters
returnees in each county. There were a number of site
(6,800 in 2011 and 1,500 in 2012)
and more than 42 community Individual beneficiaries were hazards, including severe flooding,
buildings (mostly schools) across selected jointly by the implement- that prevented access to some
the 10 states of South Sudan. Land ing partner agencies and the gov- areas. Introduction of significantly
was allocated by the Ministry of ernment. Criteria included house- stronger compressed mud block
Housing and Physical Planning. holds that were headed by children foundations helped to mitigate the
or women, households with individ- flood risk in shelters. Beneficiaries
The project also implemented with technical supervision, volun-
uals with disabilities and those who
quick-impact projects and liveli- tarily dug site drainage channels to
had no visible means to support the
hood schemes. reduce flooding risks.
construction of their own shelter.
94
”Return back home is easy

when someone supports
you to build your shelter.”
Beneficiary, Central Equatoria

State.
New settlement, in Central Equatoria state, showing a bamboo, wattle and daub shelter (far left), and compressed mud
blocks (right).
Photo: Fernando Murillo
Technical solutions Compressed mud blocks that mud-blocks were a relatively

Shelters had a single slope for Government representatives efficient material in terms of water
the roof to improve water harvest- were aware of a project in the use, led to a more positive view of
ing. This design was replicated Republic of Sudan which used the mud-blocks.
by other returnees who were not stabilised soil blocks (SSB) and The holes left behind by the
beneficiaries of the programme. A expressed an interest in this alterna- production of mud blocks were
small water tank, that could later tive. SSBs had been used for public an issue in some areas, and more
be upgraded by homeowners, was buildings but were too expensive effort could have been made to
provided with every shelter. for domestic purposes. combine drainage digging with
The shelter could be expanded Using the same press, and mud block production to facilitate
with a veranda and an external mostly black cotton soil, it was a more efficient use of both labour
kitchen to reduce the health risks of possible to make compressed mud and soil.
smoke from cooking indoors. blocks without a cement stabiliser. In the first year of using com-
Sample shelters were built for It was possible to produce 400 pressed blocks, 500 fewer shelters
the community to examine and compressed blocks a day. While the than planned were built, and the
comment on. Following feedback, technique is slower than traditional project had to return to the bamboo
shelters were plastered with burnt mud brick production (1,000 per design instead.
oil, Arabic gum and sandy soil. day) it used much less water.
Logistics
Different foundation designs The government was positive Bamboo and compressed mud
were developed for different and felt that the technique created blocks were procured or produced
ground conditions. In poor soil a new type of industry. locally. Plastic sheeting and iron-
areas, wider foundations were built mongery were imported.
Mud-blocks were less prone
on top of large stones.
to attack by insects compared to Materials list
Bamboo model bamboo, and enabled construction
Materials Quantity
Initially, shelters were built using of strong, load-bearing walls.
CSB (mud) blocks (foundation) 414
poles and bamboo wattle and daub They were cool by day and warm Polythene sheet (1m wide) 15m
walls. These were relatively quick to by night, and did not have to be CSB (mud) blocks-walls/columns 1034
build but required significant pro- transported over long distances. Corrguated iron sheets x 4m 8 pieces
Timber 125mm x 50mm x 4m 4.2 pieces
curement of timber and bamboo. The project also demonstrated Timber 100mm x 50mm x 4m 2 pieces
to each community how blocks Timber 100mm x 50mm x 4m 4 pieces
Bamboo-based structures Timber 75x50mm x 4m 11 pieces
required “mudding” to complete could be used for energy efficient Timber 50x50mm x 4m 4 pieces
and seal the walls. In a number ovens. Galvanized drainage zinc 2m 2.5 pieces
Hoop Iron (50m roll) 20m
of cases beneficiaries used plastic The introduction of compressed Nails 4" 2kg
sheeting for walling instead. mud-blocks in 2012 resulted in Nails 3" 2kg
Nails 2.5" 1kg
Shelter costs rose during con- different reactions from communi- Galvinized spiral roofing nails 3" 2kg
struction due to rising bamboo ties. Hinges and bolts 5+1 pieces
Chicken wire 1 piece
prices and unplanned transport In some areas, people already Cement (plastering) (1/6 2 Bags
costs of soil and water for mudding. built using dried mud-blocks. In cement/soil)
Soil/sand for mortar 1m3
Due to the local environmental other areas the technique was new. Anti termite treatment 2 litres
impacts of using timber, and new In some cases there was resistance
conditions set by the government to the use of the blocks, as produc-
to protect timber sources, it was tion involved considerable heavy
decided to switch away from these labour. The introduction of the
materials. block presses and the realisation
A.30 Thailand – 2011 – Bangkok Floods

Keywords: Non-displaced, Collective centres, Hosting, Urban neighbourhoods, Guidelines and
Overview: training materials, mass communications.
Summary
During the 2011 floods in Thailand, social media became a crucial
tool for information-sharing and decision-making, both for those
affected by the floods and for agencies responding to needs.
The use of social media presents challenges in terms of
filtering useful information from misinformation, the reliability
and accountability of those distributing message, and identifying
communication channels and strategies which will reach specific
target groups. Some people may not use social media at all.
This overview draws particularly on two publications: “The role of Twitter

during a natural disaster: Case study of 2011 Thai Flood,” in Technology
Management for Emerging Technologies (PICMET) and “Flooding in Thailand:
flee, fight or float”, Forced Migration Review No. 41, by Wan Sophonpanich.
This animated video explained the floods, and whether people should stay or evacuate, using whales to help explain the
volume of flood waters. It has received over one million internet hits.
Images: Roo Su Flood
Background Response options of flooding has been lost. Those

A combination of a heavy rainy As the floods slowly moved caught up by the flooding can
season and tropical storms caused towards Bangkok and its surround- be categorised into the following
the worst flooding Thailand had ing areas, people began to make groups:
seen for fifty years. Over five per contingency plans. Precautionary displaced:
cent of the country’s land was under
Despite the scale of the floods People sealed-up their houses
water by November 2011 and the
and the number of people affected, and garages and moved away
flooding had affected 13 million
the capacity of the Thai authori- from risk areas until the water
people and caused 813 deaths.
ties, national NGOs, community levels dropped.
A novel way of thinking about
groups and individuals to deal with Emergency displaced: People
the volume of water that had
problems meant that international forced to move to collective
accumulated and needed to be
organisations played a relatively centres or friends once the
dispersed was presented by the
small role in the response. flood swamped their homes.
animation group Roo Su Flood
Stayed with simple
(Know, Fight, Flood). Flooding does not automatically
precautions: People living in
The billions of litres of water lead to displacement. In fact, Thai-
areas where flooding is more
was calculated to be the equivalent land’s traditional building designs
frequent were able to withstand
of 50 million blue whales, and Roo historically coped with floods by
flood heights of two to three
Su Flood made a popular online allowing water to flow through the
metres, with minimal assistance
animation which explained the bottom floor of a house while the
needed to replace their
impact of the floods in terms of family retreated upstairs to wait for
temporarily lost livelihoods.
these millions of whales slowly the water to disperse.
Stayed with advanced
trying to make their way out of
However, in many urban areas precautions: People with
the country and into the Gulf of
of Thailand the traditional cultural considerable resources
Thailand.
capacity to mitigate the effects built flood-defence walls,
(www.youtube.com/roosuflood)
96
Livelihoods were most affected for those who chose to relocate. For most people, daily life continued despite the flood
waters.
Photo: Thanchanitch Suttichote/IOM Thailand
sandbagged entrances, installed competing for attention. Those retweeted the most were not
water pumps or bought motor- necessarily those who tweeted the
In some cases, for example, pol-
boats. People in this group most or had the most followers.
iticians offered different advice and
often helped out in their
assessments with political point- Those with the most retweets
neighbourhoods.
scoring in mind. included:
Stayed with high level of
need: People who chose not to “We are not only being Thaiflood / kapookdotcom:
move but lacked the ability to These accounts tweeted
flooded by floodwaters, but
cope with the consequences of information from the private
the flood and relied on external also by information.” sector site thaiflood.com.
assistance. Thaiflood.com became a major
People who relocated sometimes source of information, with an
found that they had under-estimat- Reliable information? active community and facebook
ed the impact of the floods and Twitter usage in Thailand soared page, and also collaborated
were forced to stay away much by 20 per cent between September with Google’s Thailand Floods
longer than they initially planned. and October 2011. A research Crisis Response site.
This had knock-on effects for their paper published in 2012 analysed SiamArsa: An account belonging
livelihoods. the most prolific tweeters and most to one of the largest volunteer
re-tweeted tweets. groups. It used Twitter and
Some of those moving to col- Facebook to share information
lective centres were displaced for The study showed that the
about flooding and volunteer
a second time when the centres content of tweets with the hashtag
work.
themselves flooded. ‘#thaiflood’ overwhelmingly
GCC_1111: The account
concerned situational announce-
belonging to the official
Information flood ments and alerts (39 per cent).
government website for the
Information was available from Support announcements made up
Flood Relief Operation Center
a huge number of different sources: ten per cent, requests for assistance
(http://floodthailand.net) which
the private sector, print and online accounted for eight per cent of
also facilitated the posting of
media, the government, NGOs and tweets and requests for information
assistance requests.
informal social media. five per cent. 37 per cent of tweets
were categorised as “other”. The Lessons to learn
The founder of the animation
study found that the majority of the Using and monitoring social
group that produced the Roo Su
situational and location-based infor- media is an important part of disaster
Flood series, explained how the
mation was tweeted by members of response in today’s world. An active
animations were a response to the
local communities. analysis of the data can help pri-
difficulty in picking out useful infor-
oritise communication channels
mation from misinformation. To identify which Twitter users
and displacement patterns, while
were seen as providing reliable in-
Information was not only being coordinated messaging can reduce
formation the study looked at the
communicated by a multitude panic and misinformation.
number of retweets users received.
of different actors but was also
Some people moved to evacuation centres, where emergency support was available, often from
volunteer group. However the majority of people decided to stay.
The two reports in the summary

note the following learnings:
Verification: It was not always
possible for people or agencies
to easily identify misinformation.
Accountability: Those actors
giving advice did not always
consider how they might be
accountable for the messages
they sent out.
Rights and responsibilities:
Knowledge and understanding
of humanitarian principles and
codes of good conduct was
often overlooked.
Simplicity: The popularity of
Roo Su Flood demonstrated
that there was an appetite for
easily understandable messages
communicated in novel ways.
Context and target audience:
The audience for the messages
should be made clear. For
example, providing information Collective centre in a university.
on how to seal up a door may be
technically correct for low-level
flooding but inappropriate and
dangerous in high-risk areas.
Of course, not all the electronic
information is available to everyone,
and communities with little or no
access to the internet not only had
less access to information, but were
also less able to vocalise their needs.
This is particularly true of highly-
excluded groups, such as migrant
workers. The migrant workers not
only had less access to electronic
information due to langues issues,
but may also have had less access
to the support available to Thais. Most people decided to stay in their houses with various levels of precautions
against the flood waters. Some vulnerable people did not relocate and did not
There were reports migrants were have access to the electronic information, and required special assistance.
denied access to some collective Photo: Wan Sophanpanich
centres and relief items.
98
A.31 Tunisia – 2011 – Conflict in Libya

Update: Keywords: Planned and managed camps, Resettlement, Household NFIs, Emergency shelter.

Tunsia
Conflict:
20 months – – 1,200 refugees
Conflict in Libya / asylum seekers
Conflict date: await resettlement
February 2011 - October 2012
1,000,000
200,000
8 months – – Conflict in Libya
Project outputs: ends
Camp with a capacity of 25,000
200,000 people pass through site
during project timeframe
4 months – Camp rebuilt
3 months – Camp destroyed

Tunisia
Shousha
camp – Migrants received at
2 weeks Shousha camp
3 days – – Migrants are hosted

Libya in urban settings
2 days – – Influx at the border
15th February starts
2011 – – Conflict starts
Project description
A transit camp was established to assist refugees and migrants fleeing the conflict in Libya. The camp was
rapidly established in partnership with the Tunisian authorities and housed a population with more than 60
nationalities mostly for only short periods. The camp management worked closely with organisations providing
support for the repatriation of displaced people to ensure that people had a smooth transit from the camp to
return locations.
Strengths and weaknesses 8 Latrines, showers and water taps were not readily
9 The organisation was able to work together with available during the initial phase of the emergency.
the authorities to rapidly establish camps to cover 8 The lack of a rapid shelter solution that was more
emergency needs. durable than tents greatly hampered the ability of
9 The camps dealt with the complexity of sheltering the organisation to assist beneficiaries in a timely and
people from different nationalities by establishing efficient manner.
separate sectors for the major nationailities and an - construction of durable shelter solutions could not
overflow sector for minority groups. be considered given the temporary nature of the transit
9 The organisation worked with fourteen other camp.
national and international organisations to provide
assistance.
8 Tents initially provided had a very short lifespan and
were difficult for people to assemble. They were also
poorly suited to the climate.
A.31 Conflict
Shousha camp was established in Tunisia near the Libyan border. It had a capacity of 25,000 people, with tents provided as
shelters. Most of the camp residents were foreign nationals. The majority travelled onwards to their home countries.
Photo: A. Branthwaite / UNHCR
Background authorities. Neighbouring countries population rebuilt Shousha camp

The conflict in Libya, began in like Algeria and Egypt refused to set with a much more organised sepa-
mid February 2011. It caused a mass up camps within their own borders. ration of nationalities and ethnici-
exodus of migrants and refugees ties in order to reduce conflicts and
from Libya. The majority of fleeing Site planning challenges to cultural sensitivities.
Libyans found refuge in Tunisian In the first days of the
emergency, the military liaison Humanitarian assistance and
homes and public institutions.
officer and the international organi- camp services were provided to
The first groups of non-Libyan sation’s field unit jointly conducted each community separately, with
nationals sought shelter in Tunisian the site planning. During the first each community allocated its own
public institutions. However, the few days, Shousha camp hosted food distribution points, water
majority the Tunisian authori- more than 20,000 migrants, pre- points and sanitation facilities. Dis-
ties and civil society groups stated dominantly single men from various tribution points were also strategi-
that a refugee camp setting would nationalities. No WASH facilities cally placed to reduce conflicts and
be more suitable for providing were available during the first days to ensure that adequate humani-
necessary humanitarian assistance. of the crises. tarian assistance was provided in a
secure environment.
Site selection In the first 24 hours, attempts
The Tunisian military set up an were made at separating groups Not every nationality and
emergency field hospital 9km from by nationalities. However, the ethnicity could be accommodated
the border with Libya when the attempts failed and Shousha camp in a separate sector and therefore
conflict erupted. The hospital was accepted large numbers of single sector E was created to host
as a result of concerns that a large men without much organisation. minority groups. Communities were
number of war wounded individu- given the option to have a separate
At this early stage, Shousha section for families in their sector.
als would be crossing the border camp did not conform to in-
into Tunisia. ternational camp management Site construction
The Tunisian authorities standards. However, emergency The site was initially construct-
requested that the United Nations tents, water, medical assistance and ed by the military who levelled
establish a transit camp (later food were provided. the ground and provided some
named Shousha camp) next to the lighting. The erection of the tents
As a result of the mixed popu-
field hospital in order to host and was completed by the military, the
lations, numerous problems arose
assist thousands of predominantly two international organisations and
amongst camp residents. Coming
migrant workers fleeing Libya. The the camp residents. Eventually, a
from very distinct cultures, religions,
displaced would stay in this camp local company was contracted to
ethnicities and lifestyles, the camp
while waiting to be repatriated to erect tents.
residents frequently bickered over
their countries of origin. Interna- space and access to humanitarian Partners and other internation-
tional organisations did not have a assistance. The most visible proof al organisations contracted local
say in the location of the site. of the tensions were the frequent companies to build sanitation in-
As Tunisia had itself experienced conflicts that arose between com- frastructure and the water network
a revolution, the political situation munities during food distributions. in the camp. International and local
was volatile. The large number of organisations provided food.
In May 2011, a major fire
displaced people entering Tunisia burned down most of Shousha Additional camps were built
meant that the United Nations had camp. The camp management or- by other organisations at nearby
no choice but to accept the available ganisation, operational and imple- locations between March and April
option of establishing the camp at menting partners and the camp 2011.
the site designated by the Tunisian
100
The camp was originally built using light-weight tunnel tents, but these had a limited lifespan in the hot and windy environ-
ment, and were replaced with heavier canvas tents. The camp later burned down and was replanned to take into account
the population’s different nationalities.
Photos: Left: A. Duclos / UNHCR, Right: A. Branthwaite / UNHCR
Coordination Some convoys were also A third type of tent was later
During the first week of the organised from Libya into Tunisia. introduced, and performed much
crisis, the United Nations Disaster Migrants and refugees were mostly better in the harsh conditions,
Assessment and Coordination team received in Shousha camp. Once though they remained technically
supported daily field coordination the other camps were established, difficult to erect.
meetings in the camp. The organi- they also received people fleeing
Core relief items such as
sation also led daily coordination Libya.
blankets, quilts and jerrycans were
meetings in Zarzis, about 1.5 hours An arrangement was estab- adequately pre-positioned and dis-
drive from Shousha camp, where lished to receive migrants from tributed. Mattresses also distributed
all international stakeholders were specific nationalities in the different and proved to be very useful.
located. camps. However, this arrangement
The organisation found itself
After the first week, various did not fully succeed given the
obliged to set up a very costly
working groups were formed. limited capacity of the other camps,
electricity grid in the camp which
Because the response was based and there was a frequent overflow
continues to be difficult to manage
in a camp, all working group rep- back into Shousha camp.
since the network is constantly
resentatives were present during Once their return had been tapped into by camp residents.
camp coordination meetings. As organised, camp residents were
the crisis subsided and the camp driven to the airport to be repatri- Exit
population diminished, coordina- ated. All camp residents received By the end of 2012 around
tion meetings were reduced to humanitarian assistance. 1,200 refugees and asylum seekers
once per week and then once per remained in the camp. The majority
month. Shelter solutions were awaiting resettlement, some
This emergency response Initially, lightweight white within Tunisia. In addition, around
involved an exceptionally high level tunnel tents were used. These tents 200 rejected asylum seekers
of cooperation with local authori- proved to be too complicated to remained in the camp. The or-
ties in general, and the Tunisian construct in a very fast evolving ganisation was in discussion with
army in particular. The Tunisian emergency with thousands of the Tunisian authorities to find
army acted as the main humanitar- migrants and refugees entering a solution for this group since it
ian interlocutor, and, in addition to the camp during the first days and was outside of the organisation’s
providing security, had a key role nights of the emergency. mandate to assist them.
in the building of the camp and in The tents were also very fragile,
the provision of humanitarian assis- breaking very easily. They did not
tance (food, shelter and health). have any exterior shading and were
blown away by the wind. After a
Population movements few weeks, the white tunnel tents
In the first two weeks of the were replaced by heavier green
emergency, migrants and refugees canvas tents. These tents were
were mostly transported from easier to build and a little more
the border to the camp by public robust. However, the roof pole
transport buses mobilised by (horizontal beam) was weak and
the Tunisian authorities and civil regularly broke.
society. Later international organi- Border crossing early in the
sation rented buses to carry out this These tents were also blown response. The camp was established
work. Some migrants were forced away by strong winds and did not near the site of an emergency field
to walk to the camp during the days have sufficient shading. hospital 9km from the border.
where the influx reached its peak. Photos: Left: A. Duclos / UNHCR
Vers Ben Gue
rdene 23 Km
102
A.31
C04
E
C03 C02
OS
Vers Ras Djd
ir 7 Km
M
Nigeria
OS
C04
sector UNHCR IOM
Military point Military
6.2 hect. Kitchens
2.1 hect. Base
NGO WC Military
compound ICRC Douche Hospital
NGO compound
Douche Douche
Military point NGO
6.9 hect. compound
Douche WC
Douche
SECTEUR Church
Douche A SECTEUR
SECTEUR 6.9 hect. E
B WC Nationale Garde
Douche
SECTEUR 8.3 hect. WC
WC
C Douche
Dou
c
9 hect. Douche Military point
he
SECTEUR WC
D WC
Military point Douche
EXTENSION Military point WC Point d’eau
Route
Child and Youth center Conduite d’eau de SONED
Conduite d’eau d’OSMOSE
Conduite d’eau de rejet
Military point Tented area
1 hect. = 10,000m2
Camps Réfugies Choucha

Plan Réseaux
Ben
Guerdane. TUNISIE
0m 30m 100m 200m 300m ` 400m
created to host minority groups.

As there were over 60 nationalities present in the camp, not all groups could have their own sector, and Sector E was
Site plan for rebuilt Shousha camp after the fire. Infrastructure and water supply networks are superimposed.
Conflict
DWG 002-2/11 23-04-2012

A.32 USA (Chicago) – 1871 – Fire

Keywords: Non-displaced / returns, Hosting, Collective Centres, Household NFIs, Construction
Case Study: materials, Transitional Shelter, Rental support, Cash, Guidelines.

USA 1874 – – Society officially
Project location: finishes its work,
Chicago having received
US$ 5m in
Disaster: donations
Great Chicago Fire 20 months – – Jubilee week to
Disaster date: celebrate the city’s
recovery
8th to 10th October, 1871
Number of houses destroyed: 8 weeks – – 15,000 families
18,000 buildings provided with food,
fuel or shelter
100,000 left homeless (a third of 6 weeks – – Estimate of relief
the population of Chicago) costs for the first
Project outputs: 6 months of ap-
proximately US$
45 per cent of the homeless were
4m (US$ 800m at
assisted by Chicago Relief and Aid today’s prices)
Society
8000 one-room “isolated houses” 1 month – – Of 6,259 applica-
6,000 free rail tickets out of town tions for isolated
houses, 4,564 had
Four barracks housing 1,000 Chicago been approved
families 2 weeks – – Barracks erected.
Shelter size: Distribution of ma-
17.8 m2 for 3 people or less terials for “isolated
USA houses” begins
29.7 m2 for more than 3 people
1 week – – Responsibility for
Materials cost per shelter: relief transferred to
US$ 125 for materials and basic the “Chicago Relief
non-food items (approximately 8-10 October and Aid Society”
US$ 2,375 at today’s prices)
This is one of an ongoing series of “historical” case studies to show how thinking on humanitarian sheltering has evolved.
(see also Section D of Shelter Projects 2008, Section C of Shelter Projects 2009 and Section C of Shelter Projects 2010)
Project description
The response included non-food item distribution, the building of barracks and one-room shelter construction.
The response was administered by the Chicago Relief and Aid Society, a voluntary body, first established with
the aim of supporting the poor in areas that the local authorities could not or would not support. The Society
used a “scientific charity” method, employing paid professionals to carry out the policies of the executive board,
emphasising the importance of public health issues and encouraging self-reliance amongst recipients of aid.
Strengths and weaknesses considered. Many temporary homes were still used but
9 Use of a non-political, voluntary body as relief were in poor condition years later.
administrator reduced potential for corruption. 8 The Society held on to a surplus of funds through
9 Prioritisation of key-worker support ensured periods of severe economic difficulty, including when
reconstruction work was not hampered. protestors marched on the Society’s offices in 1873
9 Support of beneficiaries in having their own shouting “Bread or death!”.
shelters, rather than living in collective centres, kept - Many of the challenges faced by the Society are
the “dependent” population low. similar to challenges faced today.
9 Use of relief as economic stimulus meant recovery - Chicago benefitted from being economically
was reasonably fast. important and reconstruction began quickly. Over 200
8 Moralising approach to shelter response. Beneficiary stone buildings were under construction within two
selection was based on class and "worthiness" and not months of the fire
necessarily on need. - The relief response was aided by the swift arrival of
8 Long-term sustainability of shelter solutions not fully donations from other States and even overseas.
View after the Chicago Fire.

Photo: Lovejoy&Foster
Before the fire However, some important people to relocate was quickly
The city of Chicago had a popu- parts of the city remained intact, stopped as the Society felt that too
lation of around 300,000 people. including its heavy industries and its many skilled workers necessary for
The city was a growing manufac- rail infrastructure. reconstruction work were leaving.
turing centre. Chicago was known
for its leadership and innovation in Coordination Selection of beneficiaries
public health issues, largely due to In the immediate panic, during Selection of beneficiaries was
the activities of the Chicago Relief and after the fire, there were informed by class, skillset, ability to
and Aid Society. concerns about public order. The work and vulnerability.
mayor imposed martial law.
Many of the working classes Although the vulnerable were
of Chicago lived within the city Five days after the fire had a priority, the Society felt that the
limits. The city centre was a mix begun, the mayor handed over re- “workshy” did not fall into this
of expensive stone buildings and sponsibility for the administration category. In a Society report the
cheap wooden-frame houses. of relief and reconstruction to the “St. Paul’s Rule” is mentioned:
Chicago Relief and Aid Society.
Fires were common in Chicago. “He who does not work, neither
The City Fire Marshal reported The Society immediately divided shall he eat.”
earlier in 1871 that the previous 12 itself into eight committees forming
The two main shelter solutions,
months had seen 600 fires. a structure not dissimilar to today’s
barracks or isolated houses
coordination systems.
(one-room timber-frame shelters),
After the fire were allocated more or less on
The Great Chicago Fire started Response
The Society offered different the basis of class. Those from the
on the evening of Sunday 8th lowest social groups were housed
October 1871. After two days, the types of shelter and non-food item
support: in barracks.
fire burnt itself out and rain helped
douse the flames. 6,000 free rail tickets to those The isolated houses were
wishing to leave the city reserved for skilled workers or the
The fire destroyed nine square “respectable” labouring poor,
kilometres of Chicago, making a 8,000 one-room shelters,
“isolated houses” who were needed for reconstruc-
third of the population homeless tion work and who the Society felt
and destroying 18,000 buildings Barrack accommodation with
furniture for 5,000 people needed a home of their own to
and 200km of wooden pavements. maintain their pride and prevent
Around half of the damage was Rental payments at US$ 10/
month (US$ 190 equivalent their morals from slipping.
insured, but as several insurance
companies were burnt out, perhaps today) Beneficiaries had to “make an
only half this was ever paid out. Furniture, stoves and fuel to application” to receive the materials
families that did not require to build. In most cases, the isolated
Those people living in cheap or other shelter support. house was donated by the Society.
rented accommodation lost every- In cases where they believed the
The early policy of offering free
thing. family to have sufficient financial
train passes out of Chicago to allow
104
Barracks inhabitants were: property market.

“Under the constant and careful The fire led to the suburbani-
supervision of medical and police sation of Chicago. New buildings
superintendents, their moral and and fire-safety codes led to many
sanitary condition is unquestionably working class families moving
better than that which has hereto- to new plots in the suburbs. The
fore obtained in that class.” building of the “isolated houses”
outside the city limits was encour-
Implementation aged by lower land prices and
The Society divided the city into a policy that exempted wooden
districts and then established a buildings outside the city limits
series of charity bureaus responsi- from taxes.
ble for assessing beneficiary claims.
The supply depots were connected As only richer families could
by telegraph. afford the stone buildings that
conformed with new fire safety
Beneficiaries were constantly regulations, the centre of Chicago
monitored, and each person had his became the business and commer-
or her own “ledger” completed by cial district, with accommodation
The Chicago Relief and Aid Society
a Society representative. Assessors for the wealthy. The Society did
coordinated the response. The
quotes here are taken from their reported in turn to their superinten- not appear to find this situation
own reports. dent, who reported to the General problematic at the time, but several
Superintendent. Included in each years later the new suburbs of small
resources, recipients were asked to ledger was the amount of relief wooden houses had in some places
pay back 75 per cent of the value of supplied and “whether they are idle become overcrowded tenements.
the shelter over one-year, interest- or industrious”.
free. Technical solutions
After five weeks, the Society
The Society provided two types
A “Bureau of Special Relief” was had a “clerical force” of 498 people
of house: a larger 20’ x 16’ (29.7
set up to seek out those too proud at district-level, 111 people working
m2) model for families of more than
to ask for assistance. A mechanism on warehousing and distribution
three people and a smaller 12’ x 16’
was set up to enable this class of and 34 people working on the
(17.8 m2) model.
beneficiaries to: planning committees.
The house design was a wooden
“Seek relief where publicity In the end, the Society supported
structure with a double iron
could be avoided, and the shock be around 45 per cent of those
chimney. Walls were lined on the
lessened to their sensitiveness and affected by the fire with shelter.
inside with thick felt paper.
reserve.”
Beneficiary feedback Along with the materials for the
Barracks The Society set up a mechanism house, the Society provided chairs,
The Society decided that the for beneficiary feedback by placing a table, a bed, bedding, a stove
number of people living in barracks an advertisement in all newspapers and kitchen equipment. The total
should be limited: for people to contact the Superin- cost of the materials and non-food
tendent of the Chicago Relief and items was US$ 125 (approximately
“So large a number, brought Aid Society with details of cheating US$ 2,375 at today’s prices).
into promiscuous and involuntary or overlooked beneficiaries.
association, would almost certainly The houses were designed to be
engender disease and promote DRR and suburbanisation erected by the families themselves,
idleness, disorder, and vice.” New fire regulations were possibly with the assistance of pro-
proposed in November 1871 largely fessionals that the family would
The Society was worried that
banning wooden houses within the pay for. If the beneficiaries were
housing people in barracks would
city limits, meaning that only the “widows, the infirm or otherwise
reduce their incentive to work,
wealthy could construct houses in helpless persons” then the Society
endangering the moral wellbeing
the city centre. Despite protests, built the house.
of the whole city. This meant that
they were eventually passed but
schools and churches that had been The Society assumed that the
without effective penalties for
used as temporary collective centres houses would be improved on at a
breaking the rules.
were closed as soon as possible. later date (some families upgraded
Chicago faced further notable with second stories or extensions),
The cost for housing a family of
but smaller fires in both 1873 and and provided a screen to allow for
five people in barracks, including
1874. Stricter safety codes were the division of the house into two
the provision of furniture, was cal-
not put in place until insurers rooms as the family wished.
culated as US$ 80 for six months,
threatened to boycott the Chicago
not including food and fuel.
Perhaps the earliest recorded drawing and bill of quantities for a one room shelter.
Source: Report of the Chicago Relief and Aid Society of Disbursement of Contributions
for the Sufferers by the Chicago Fire, 1874
Logistics In the first five days after the Materials list for a shelter
65 million linear feet of timber fire, 330 rail carriages of goods Materials Quantity
were destroyed by the fire. Demand were received as donations. None Studs 2"x4" (8 ft lengths) 52
for rebuilding was high, and the of the cars arrived with way-bills. Joists for 2 floors 2"x6" 18
cost of timber rose dramatically. At this point “the law of humanity (12 ft lengths)
Rafters 2"x4" (8ft lengths) 10
The Shelter Committee, led by a was paramount to the laws of Sills 2"x6" (16ft lengths) 2
businessman in charge of one of commerce” and most items were Plates and Ridge 2"x4" (16ft 3
Chicago’s biggest lumber firms, distributed without being recorded. lengths)
Girders 2"x4" (16ft lengths) 4
anticipated this and pre-ordered Initially, mainly second-hand Sides (8ft boards) 500ft.
large quantities of timber in the first summer clothing was available.
Floor(16 ft boards) 300ft.
few days of the response at 80 per Floor Attic (16 ft boards) 200ft.
This could not provide sufficient Roof (8ft boards) 500ft.
cent of the market price that was protection for the winter. The Battens 66
reached just two weeks later. Society supported a number of
Door and Frame
Two Windows and Frames
The first load of timber was “Ladies Societies” to produce Door Trimmings
delivered on the same day that the winter clothes. This employed Nails 10.d 30lb
Nails 20 d. 5lb
final flames were extinguished. many women who were otherwise Nails 8d. 5lb
without work after the fire.
106
Opinions... Shelter Projects 2011–2012 B
SECTION B
Opinions
This section contains short discussion documents on various issues in shelter, written by individuals with
a specific interests in each subject.
B.1 The History of Three Point Five Square Metres 104

A review of the origins of the use of 3.5m2 as an indicator for covered living space,
contrasted with the multiple other indicators that exist.
B.2 Bankers and Builders: 106

A review of cash transfer programming in shelter responses.
B.3 Livestock Sheltering in Humanitarian Situations 112

An overview of the background and issues in livestock sheltering in humanitarian response
B.4 A Reflection on the Importance of Settlements

in Humanitarian Shelter Assistance 115
A discussion of the neighbourhoods approach
B Opinions...
108
Opinions Shelter Projects 2011–2012 B.1
B.1 The History of Three Point Five Square Metres

Of all the numeric indicators commonly used as reason given for these shelter spatial requirements, is
guidelines in humanitarian shelter response, it is the air ventilation, rather than other possible concerns such
indicator for covered shelter space that is perhaps the as climate control, privacy or storage of belongings).
most often quoted – three and a half square metres
By 1979, with the overwhelming numbers of
per person. However, a lack of awareness of where
refugees crossing the borders from Cambodia into
this and other indicators came from has played a part
camps in Thailand, and institutional fears for the
in limiting discussion on the appropriate use of this
breaking of the principle of “Do No Harm” by another
indicator across all forms of shelter and reconstruction
wave of new field organisations, UNHCR regional
response.
offices asked Cuny and others to facilitate a series of
The development of principles and designs for hu- workshops, with the express purpose of making hu-
manitarian shelter started in the early 1970s, when manitarian response globally more efficient. Major
failures to provide adequate support to displaced outputs from these 1980 workshops included the
people in camps resulted in public-health catastrophes, creation of the system of UN ‘lead agencies’ for each
and the reduction in disease-related fatalities was seen major sector of humanitarian response, the drafting
as the key improvement to be prioritised above all else. of the first specific book published for emergency
Wanting to avoid repetition of disasters in Biafra and responses (UNHCR’s Handbook for Emergencies) –
Bangladesh, Fred Cuny and others working for a variety and the adoption of a system of numeric minimum
of NGOs, referred back to their own personal experi- standards as control mechanisms, including a number
ences in minimum-existence standards for low-cost of the standards borrowed from the WHO publication,
public housing in Europe and north America, as well as which included the standard of 3.5m2.
emerging research in non-emergency sites-and-service
During those workshops, there was a debate about
slum-upgrade projects in Latin America.
the conundrum of applying global standards, when
Lessons learnt from the first attempts at commu- situations and needs were often so vastly different. The
nity-focused camps in India and Nicaragua, demon- analogy used in these debates was once again from the
strated that the designs must remain very localised, medical perspective – how can doctors have a rigorous-
in order to be culturally acceptable to the inhabitants. ly universal recommended dosage of medicine on the
At the same time though, there remained the life- one hand, and yet still be able to successfully adapt that
and-death challenge of ensuring that everyone in a dosage to each patient’s needs on the other? Despite
camp, or in need of shelter support, had the equitable these reservations, the numeric standards were incor-
minimum sufficient necessary to actually live. The need porated into the Handbook for Emergencies in 1981,
to solve this problem was seen as more urgent because and fifteen years later, despite opposition from some
with every major humanitarian crisis from the 1970s NGOs, had even wider adoption with the first draft
onwards, the failures of response were made worse publication of Sphere (see www.sphereproject.org).
by the exponential increase in newly-formed NGOs
coming into the field, and the diminishing prospects
of giving any sort of personalised guidance to inexperi-
enced organisations or managers.
Cuny and his associates found a short booklet,
published for the World Health Organisation in 1971,
and written by an under-secretary for the Ministry of
Health in Iran, called “Guide to Sanitation in Natural
Disasters”. Here was a seemingly ready-made list of
minimum numeric standards specifically for shelter and
camps, and with the overriding objective of ensuring
adequate public health in disaster situations. For the
most part, the booklet does not deal with shelter spe-
cifically, but in its list of standards for shelter, lies the
standard for covered shelter space – 3m2 per person
for tents in tent camps, and 3.5m2 for buildings. The
booklet also offers other shelter standards which have
not been adopted more widely since, including one for
the actual cubic metres of volume space (rather than
just flat floor area), but the author gives no references
or evidence to support the shelter numbers. The justi-
This document written in 1971 is possibly the first
fications given for each standard in the booklet come published reference to numerical planning figures for
consistently from a public health perspective (the one humanitarian emergencies.
B.1 Opinions
In the latest 2011 edition1, the Indicators as a

section, with all their exact numbers, have been moved
further down the page, and their prominence has been
replaced by a new section called ‘Key Actions’. For the
standard on Covered Living Space, both the Key Actions,
and for the first time the Guidance Notes, highlight the
livelihoods potential provided by adequate shelter. The
editions of Sphere subsequent to 2000, have highlight-
ed the qualitative aspects of the standards (compared
to what Sphere terms the numeric indicators), and
have increasingly emphasised both the incremental
process, and the need for localised adaptation of these
standards. However, the numeric indicators, despite
being pushed progressively further down the page, still
exist, and still exist in the project proposals and evalua-
tions for many humanitarian organisations.
The situation becomes more complex if, as in many
cities in both developed and less developed regions, the
Sphere minimum standards are better than local living
conditions. A further complication, and one which high-
lights the gap in humanitarian standards in general in
not going beyond the individual household, is illustrat-
ed by the post-tsunami response in Aceh. Adherence to
3.5m2 was a contributing factor in a rapid post-disaster
spread of the urban area, through the construction of
low-density shelter settlements, into marshlands which
would have otherwise provided much of the natural
protection for the city, from floods and tsunamis.
“...In the Shelter section [of The Sphere Handbook 2011]
of the chapter on Shelter, Settlements and NFIs there are In order to determine what is essential it would be
a total of 32 Key Actions, and 10 Indicators, but it is the more appropriate to apply measures that reflect what
3.5m2 (which is after all now just part of a Guidance Note)
shelter does other than just contribute to public health.
which still gets more attention..”
Photo: Joseph Ashmore Shelter can have more impact on areas of humanitarian
intervention such as protection and livelihood creation,
The first draft of Sphere, in 1998, has only one and indicators that reflect these aspects of shelter
standard for individual shelter, called ‘Housing Standard impact need to be developed. It would be sensible to
1: living quarters’ for which the first indicator, and the reflect these aspects of shelter provision in guidelines
only indicator with a numeric measurement, was, ‘The for shelter living space. It is telling that for the six Core
covered area available per person averages 3.5-4.5m2.’ Standards in Sphere 2011 there are a total of 62 Key
As Sphere now clarifies, sometimes this indicator will Actions, and 27 Indicators, and in the Shelter section of
not be appropriate. There will be situations where 3.5m2 the chapter on Shelter, Settlements and NFIs there are
per person cannot be met, for example when there are a total of 32 Key Actions, and 10 Indicators, but it is the
insufficient resources to provide this amount of living 3.5m2 (which is after all now just part of a Guidance
space. In such circumstances a pragmatic decision may Note) which still gets more attention.
need to be taken to provide a basic level of shelter for
many, rather than meeting the minimum standard for
only some. Providing shelter in a cold climate presents
another dilemma - 3.5m2 per person can be difficult
to heat at a time of scarce fuel, or environmentally Jim Kennedy
damaging in an area where using timber for constru- Charles Parrack
cion leads to deforestation. By the 2004 edition of
Sphere, there was a major shift in emphasis, with the
guidance notes expanded to include ‘Duration’, under
which Sphere stated that 3.5m2 may be appropriate
in the first instance, that 3.5m2 may be incremen-
tally achieved over time, and that an argument may
be made for providing less than this, based upon the
shelter norms of the affected or neighbouring popula-
tions.
1
Sphere Project, Sphere: Humanitarian charter and minimum standards in humanitarian response, 2011
110
Reported covered living areas of the shelters in the case studies in this book.
m2
70
20 x 3.5m2
15 x 3.5m2
Minimum provided by the project

Maximum provided by the project
Covered area of shelter in multiples of 3.5m2
m2
40
m2
36
10 x 3.5m2
m2
30
m2
28
m2
m2
7 x 3.5m2
22
m2
m2
m2
m2
22
m2
21
21
21
21
20
6 x 3.5m2
m2
m2
m2
m2
18
18
m2
18
18
m2
5 x 3.5m2 m2
16
16
16
m2
4 x 3.5m2
m2
12
10
3 x 3.5m2
2 x 3.5m2
1 x 3.5m2
–
Co ara – 2 –
d’ a – 2 –
Bu – H – 2 –
A ina ti – 0 –
A .28 Sud – 20 –
m –2 –
.8 gas – 2 –
–
2
07
07
10
2
12
11
1
12
2
12
11
te no 201
01
01
01
A 2 – iop 201
.2 stan 01
.3 ine 01
S 201
01
20
20
20
Ph ista 20
in 20
fg Pe 20
Ph ista 20
18
op 20
20
2
2
a –
–
– s–
–
Le bia
1 ire
ti
Pa es
ru
So n
lia
Et ar
ia
Pa A
r
te gu
i
– Fas
a
ai
ai
i
c
a
.2 kist
– ’Ivo
A Ivo
m
U
pp
pp
ki
hi
k
an
lo
h
b
–
–
Pa
d
Et
a
ili
ili
4
Co
–
c
2
h
ad
i
.1
.1
rk
N
–
–
–
–
A
A
–
M
Co
.9
3
A
6
A
.3
.2
.2
.2
.1
A
–
–
7
A
–
–
A
.1
A
A
.2
.2
8
.1
.6
.5
.2
A
.1
A
A
A
Chapter number and case study
Smaller shelters are often constructed after an assessment of local and host population standards, as well as what
is practically possible. Shelter size is not necessarily a good indicator of the quality of a shelter programme, and
reflects a diversity of issues, including varying needs, permanency, budgets, logistics constraints,
host standards, and official policies.
Note: Covered areas are often reported based on external wall dimensions and not the internal usable space.
For example, a 6mx3m shelter with 20cm thick mud block walls will often be reported as being 18m2. Practically the
usable covered living space will be lower (5.6mx2.6m = 14.5m2).
B.2 Opinions
B.2 Bankers and Builders

The coming of age for cash and shelter projects
Basic principles Cash transfers for shelter
The professionalisation of the shelter sector as a part Across all sectors, the direct provision of goods
of humanitarian assistance is often dated to the early and services to affected populations – in kind assistance
1970s and the work of researchers-cum-practitioners - remains the most common form of delivering human-
like Fred Cuny and Ian Davis1. These grand doyens of itarian aid. The drivers for in kind assistance among
shelter after disasters helped establish a number of prin- agencies and donors are the same across all sectors: the
ciples for the sector that remain true today, including: need for highly visible relief operations; the desire to
reduce suffering and disease through quickly launched
The aspirations and capacities of affected
humanitarian responses; and achieving economies of
populations must be at the heart of all settlement
scale and value for money. There is a growing recog-
planning and shelter reconstruction activities.
nition within the humanitarian community that direct
The majority of people displaced by disasters figure cash transfers to disaster-affected people can help
out their own shelter solutions, often through the agencies, donors and governments fulfil their mandates
involvement of other relatives, neighbours, or the and meet public expectations.
host community at large.
As noted in the shelter case studies in Shelter
To help regenerate livelihoods and provide income Projects 2011-2012 and previous volumes, there are
to affected households, preference should be two main types of cash transfer methods used in shelter
given to the use of local labour and local building programmes:
materials for construction activities.
Cash for Work:
In the 40 years since these principles of community Direct cash payments to beneficiaries for their
driven shelter programmes were first espoused, these labour on debris clearance, shelter construction
ideas have been accepted as axioms by the shelter sector or other community focused infrastructure
as a whole. Yet many of the current debates about the projects;
most appropriate shelter solution for affected popula-
tion are led by architects and builders, not community Conditional Cash Grants and/or Vouchers:
mobilisers or anthropologists. The first two editions Direct cash payments to beneficiaries or landlords
of these case study reports, Shelter Projects 2008 and for services defined by agencies or governments;
Shelter Projects 2009, are heavily weighted towards e.g., participating in training programmes;
“expert driven” shelter options. Of the 81 case studies rebuilding homes according to pre-defined plans
in these two volumes, less than 15 per cent refer to or construction stages; or rental support;
shelter projects that included a component that offered
affected people a greater choice and responsibilities: A paper, token, or debit card voucher that can
the provision of cash transfers directly to beneficiaries. be exchanged or redeemed at pre-selected
vendors for a pre-determined quantity or value
The case studies in Shelter Projects 2010 highlight of construction materials or services.
how far the shelter sector has come in considering cash
transfers as a tool for shelter responses – almost 50 A third type of cash transfer mechanisms used in hu-
per cent of the projects cited have a cash component, manitarian responses is Unconditional Cash Grants,
including an early use of shelter-related cash grants and where direct cash payments are made to selected
loans for disaster affected people in 19062. beneficiaries (usually the highly vulnerable or poorest)
without conditions or requirements. While post distri-
bution monitoring of household expenditures suggest
that food, health care, or loan repayments are typical
purchases made with unconditional cash grants, there
is some evidence to suggest that under certain condi-
tions beneficiaries will choose to spend the money on
shelter materials. For example, in Pakistan in 2005, over
95% of earthquake affected households who received
a small cash grant (US$ 40) spent the funds on shelter
construction or material transport3.
1
See C.1 Shelter Projects 2009, and D.1, Shelter Projects 2008
Case studies illustrate how far the shelter sector has 2
See case study B.2, Shelter Projects 2010
come in considering cash transfers as a tool for shelter
3
responses. See A.1 for an example of a programme that See Humanitarian Exchange Magazine, Issue 34, “Respond-
has moved towards using cash support. ing to shelter needs in post-earthquake Pakistan: a self-help
Photos: Jake Zarins approach”
112
Scepticism on cash and shelter Strengths and weaknesses of cash for

Accompanying the increase in interest by agencies shelter
and donors in cash transfers for shelter support pro- Regardless of the sector, the success of cash
grammes is scepticism from some shelter specialists transfer interventions is highly dependent upon assess-
on what is seen as “cash evangelism”. Many of the ments and a thorough response analysis. Key elements
doubts focus on concerns and perceived risks around to be considered in shelter programme design using
unconditional cash transfers and self built reconstruc- cash as tool are:
tion. How can we ensure, ask the sceptics, that people clarity on the programme objectives, and what the
won’t rebuild using inappropriate designs, poor quality shelter programme is trying to achieve within the
materials and unsafe construction techniques if we limits of budgets and time frames
just give them cash? Fortunately, most mainstream
agencies and shelter professionals recognize that cash proper targeting of affected households who are
transfers for shelter projects must be accompanied both most likely to benefit from and take advantage
by technical advice and support, or given in tranches of cash transfers programming
based on a phased approach. Like all humanitarian as- an understanding of household and community
sistance, however, post distribution monitoring of cash economic activity that help inform how cash
or in-kind assistance is essential to ensure that project injections can complement and enhance recovery
goals are met and that the aid given “does no harm” after disasters
to its recipients.
a market analysis with a sufficient level of detail
A second set of concerns on cash transfers for to know how the disaster or conflict has affected
shelter relates to the high cost of safe or safer shelter building material supplies, skilled labour, and
after disaster. By restricting the number of families rental markets, and what might be the negative
who receive cash and shelter assistance, isn’t there a (inflationary) impact of injecting cash into local
risk that social tensions within or between communities economies
will be exacerbated? Does the liquidity of cash poten-
tially increase conflicts between neighbours? While the a robust monitoring and evaluation system in place
answers to these questions are possibly yes, humanitar- to measure impact and gauge the effectiveness of
ian assistance in all sectors grapple with these questions cash transfers as a programme tool.
in each and every response. To date, the best way to
avoid these potential conflicts is through coordination,
ongoing consultation, and robust accountability mech-
anisms in place to address community and beneficiary
concerns.
As evidenced in the increasing number of cash transfers

“The success of [cash] interventions is highly dependent in case studies in recent volumes of Shelter Projects, it
upon a detailed response analysis.” can be expected that cash will become a more frequent
Ampara, Sri Lanka. component in humanitarian-driven shelter responses.
Photo: Jerry Galea-Oxfam Photo: Ivan Muñoz, Intermon.
B.2 Opinions
Strengths Weaknesses
Cash for Work Provides a temporary income Physical labour projects might
to affected people, especially exclude women and less able-bodied
vulnerable groups who may not persons;
have other sources of income; May create expectations
Motivates people to participate that NGOs will pay people for
in relief and recovery operations; participation in all relief work;
Generally high level of Could compete with private sector
acceptance by authorities. labour;
Cash for Work activities selected
may not have strong coherence with
shelter activities.
Opportunities Threats
Can help address disaster waste Potential health and safety issues;
Cash for Work programme at
Tapis Rouge camp, in Port au management concerns; Quality of work might be poor;
Prince, Haiti Phot: Ivan Muñoz, Can link with training If too high, the wages might
Intermon programmes to improve building discourage normal livelihood
skills and livelihoods. activities; if too low, participation
may be reduced.
Strengths Weaknesses
Conditional Cash Grants Contributes to economic Less able bodied and more
(including vouchers) recovery of local markets; vulnerable groups may not be able
Cash grants can support access to meet conditions, and will require
additional support;
to rental accommodations; Time commitment by beneficiaries
Generally high level of to meet conditions may conflict with
acceptance by authorities. seasonal livelihood activities.
Opportunities Threats
Voucher based programmes Increased demand for shelter
require close contact with local materials may outstrip local supplies;
suppliers, which can help ensure Possible inflationary impact on
Conditional cash grant used in
housing reconstruction. Deah
quality in shelter materials; prices by creating high demands for
Baro village, Banda Aceh, Sumatra Setting targets and transferring shelter materials.
Indonesia. cash in tranches can support safe
Photo: Jim Holmes, Oxfam onstruction practices
Cash grants to host families can
reduce their financial burden and
facilitate camp decongestion.
The Future of Cash and Shelter

As evidenced in the increasing number of cash While bankers may not have the skill set that
transfers in case studies in recent volumes of Shelter agencies and donors look for to help guide cash and
Projects, it can be expected that cash will become a shelter programmes, the architects, engineers and
more frequent component in humanitarian-driven builders of the shelter community would be wise to
shelter responses. With the proliferation of mobile include cash transfers as a potential instrument in their
phone access throughout the world and the increase tool box. As with all innovations, however, care must
in security of mobile banking transactions, future cash be taken to avoid cash transfers as the default option
and shelter programmes are likely to be more digitally for all shelter programmes. Builders and bankers alike
oriented than what we see now. In other humanitar- know the truth to the old adage: if the only tool you
ian sectors such as Water, Sanitation and Hygiene have is a hammer, all problems look like nails.
Promotion (WASH), mobile phones are increasingly
used to reach wider audiences with key messages and
as project monitoring tools. For a sector such as shelter,
Rick Bauer,
where hazard reduction principles and “building back
Engineering Adviser and trainer in the use of cash
better” are the new axioms, the potential of combining
and market assessment tools,
mobile phone technology and cash programming are
Oxfam GB
yet to be explored.
114
B.3 Livestock Sheltering in Humanitarian

Situations
Background expense of human needs or a lack of consideration
Livestock have been sheltered within household in- of livestock shelter needs at all. Planning for livestock
frastructure for hundreds of years. Vernacular buildings shelter need may not incur an additional cost, and will
in many less developed countries still contain provision help mitigate issues arising from livestock and people
for livestock. Fences and bushes within a household plot living in close proximity.
of land are also traditionally used to shelter livestock. A few examples do exist, such as the response to the
For example in Gujarat, India, thorny fences of Acacia 2005 Pakistan earthquake, where shelter for livestock
Arabica are used to protect the buffaloes. In Sri Lanka, has been built by external organisations. More common
fences of wood and wire are used alongside sheds are accounts of how disaster-affected people make pro-
made of wood or bamboo, roofed with grass or leaves. visions for their livestock themselves, often using the
In less developed countries, livestock are often materials that organisations have intended them to use
people’s largest capital asset and keeping them within for sheltering their family.
the household plot is the most obvious way of pro- In cold climates, such as northern Pakistan, there is
tecting that asset. Wherever people find shelter, they greater consideration of livestock shelter needs as part
attempt to make provision for their animals to live close of a sustainable livelihood solution. In warm climates,
to their dwelling, especially if their livelihoods depend
upon them.
Why consider livestock in a shelter

response?
Livestock are important to people for a broad range
of reasons. They are a key wealth asset acting as a bank
account in areas where people have no other means
of storing financial capital. Livestock are also important
for livelihoods dependent on animal produce and
labour, transporting goods and people, providing milk
and meat, cultural activities and personal security.
In less developed countries where humanitarian as-
sistance is required post disaster, up to 70 per cent of
resource-poor people rely on livestock for their liveli-
hoods. This dependence, coupled with the potential ad-
vantages of linking emergency responses with recovery
and development programmes, gives a window of op-
portunity for emergency shelter actors to incorporate
a sustainable livelihoods approach at the emergency
phase. Assessments of livestock shelter needs could
be carried out as part of broader shelter needs assess-
ments.
Designing livestock shelters is one of the simpler
parts of an overall shelter response. Support for
the construction of sheds, covered areas or secured
external spaces can be provided or enough space can
be left in settlement planning for people to build these
themselves. Livestock sheltering should include consid-
eration of access to grazing, fodder production, envi-
ronmental impact, vaccines and quarantine. Failure to
do so can lead to weak shelter that can be damaging
to animal and human health, and is not locally sustain-
able.
“Livestock have been sheltered within household infra-
Livestock in shelter responses structure for hundreds of years.“
Top to bottom: European farmhouse with space for cattle
Shelter solutions for livestock are seldom seen as underneath. Goat shed attached to main house in Sri
a priority by responding organisations during the first Lanka. House in Vietnam with pigs sheltered under the
stages of an emergency. This is often due to the as- raised floor.
sumptions that livestock sheltering is a cost at the Photos: Top Julia Macro, Below Tim Leyland
B.3 Opinions
“A few examples do exist, such as the response to the 2005 earthquake in Pakistan, where shelter for livestock has been
built by external organisations”. These communal stables were built to shelter livestock in a large camp.
such as in Haiti post-earthquake 2010, the need to

shelter livestock from the elements was reduced. In
very hot climates, however, there may be a need for
shade which people often take their own steps to find
or create.
The argument for sheltering livestock can also be
made when there are potential security threats from
livestock thieves. If the livestock asset is safely secured
and/or out of sight, this can reduce opportunis-
tic livestock raids and reduce the vulnerability of the
livestock owner.
Sheltering issues relating to livestock

For displaced and non-displaced disaster-affected
people who are keeping livestock, or where people
have migrated with their livestock, there are several
shelter and settlement issues related to livestock:
Spread of trans-boundary disease: As people
migrate with their livestock due to conflict, drought
or other natural disasters there is an increased risk of
spreading infections. This risk can be mitigated through
working with national and regional disease surveillance
projects (where available) and planning a locally ap-
propriate response with regional veterinary and public
health organisations.
Access to grazing / fodder: There may be reduced
grazing land or fodder available or access may be
constrained due to insecurity, difficult terrain or host
community relations. Competition between host and
displaced communities over grazing land and watering
points can often cause conflict and is therefore a key “Accounts [are common] of how disaster-affected people
consideration in settlement planning. make provisions for their livestock themselves, often using
the materials that organisations have intended them to
Location of livestock: When displaced people in use for sheltering their family”.
a camp or temporary settlement have their livestock Above: A child’s coat is given to a calf after the Pakistan
with them, space for livestock should be included in earthquake, and a shelter is built for the livestock whilst
site planning. Considerations include living space for the family are living in a similarly makeshift shelter.
the animals, access to grazing, exit routes and a secure Photos: Joseph Ashmore
Camels provided with blankets in a camp in Kandahar province Afghanistan

Photos: Joseph Ashmore
116
and protected location. Cultural norms may prevent

communal grouping of livestock as people may prefer
to keep their livestock separate for fear of identification
problems or sharing disease. For non-displaced people
community livestock sheltering may be possible, if
identification and disease control are considered. In the
majority of cases, people prefer to keep their livestock
assets within their own household plot, and considera-
tion of this should be made when planning.
Environmental impact: Livestock can over-grave,
pollute water sources and cause local erosion.
After Cyclone Aila in Bangladesh, with limited space
Water availability: Households with livestock available on road sides or embankments, many families
require extra water for the animals. Household water prioritised secure shelter for their livestock.
shortages are sometimes due to a failure to include
animals’ drinking needs. Separate water points for
livestock and people should be planned where possible
to reduce contamination of human water sources.
Faeces disposal: The collection and disposal of
livestock faeces should be included in a shelter response
to prevent health and hygiene problems. Dried dung
can be an important, cheap source of fuel.
Host community / Government: Along with
firewood and water collection, livestock grazing can In Dadaab, Kenya, there are restrictions on livestock, but
place extra demands on fragile natural resources. Ar- they remain an important livelihoods asset.
rangements for livestock should be discussed with local Photo: Joseph Ashmore
communities and government.
Conclusion
As international organisations seldom incorpo- The need to consider livestock-based livelihoods
rate livestock considerations into their emergency during the emergency stage of shelter responses is
shelter programmes, problems arise that, with greater beginning to be identified by some agencies. Simple
awareness and integrated planning, could be mitigated actions, such as assessing livestock-based livelihoods in
against or eliminated completely. a disaster affected community, can enable responding
The Livestock Emergency Guidelines and Standards shelter agencies to decide whether there is a need to
(LEGS) can be used to support decision making and factor in livestock considerations.
planning for livestock shelter and settlement interven- In the majority of cases, conflict and disaster
tions. (For further information see www.livestock-emer- affected people make provisions for their livestock
gency.net) shelter without support or resource allocation from
external organisations. Greater awareness of livestock
issues by responding organisations could enhance the
value of livestock to displaced people by reducing levels
of animal ill-health caused by inappropriate sheltering.
Seeing the sheltering of livestock as part of a liveli-
hoods solution may provide organisations with better
opportunities for integrating livestock-based inter-
ventions. Increasingly organisations want to provide
people with the tools they need for self-recovery,
where affected people choose how and where they
will rebuild, which may involve livestock as a necessity.
Improved support for livestock-based livelihoods may
be part of an umbrella of interventions to improve re-
silience.
Julia Macro
“Dried dung can be used for fuel”.

Dung drying in a camp near Kandahar, Afghanistan, 2003.
B.4 Opinions
B.4 A Reflection on the Importance of Settlements

in Humanitarian Shelter Assistance
The convergence of disasters and urban areas has that provides a positive momentum to return to normal.
propelled the need to address “settlements” as an The second factor is that settlements, in the urban
important component of disaster assistance. But, urban context, become more important than the shelters,
areas are complex physical and social environments because the settlement sets the conditions for, and fea-
which have forced a considerable increase in the com- sibility of, the shelter/housing response and the other
plexity of our humanitarian response and the difficulty qualities necessary for sustainability and resilience. In
of recovering after the disaster. Shelter and Settlements short, if the settlement does not take hold or is inad-
are inextricably linked and can no longer be treated as equate, recovery cannot begin, and decent, permanent
separate units or responses, but must be managed as a housing will not start. Ultimately, if the settlement fails
single, indivisible programme undertaking. the housing will fail. Alternatively, if a good settlement
start is established, housing will likely commence and
The settlement is the framework and platform
develop on its own. The third factor is that urban settle-
(physically and socially) that establishes the living space
ments are often too big to deal with all at once. For this
of which shelter is but one component. There can be
reason the best approach is to work with smaller set-
many types of settlements: good and bad, big and little,
tlement units or neighbourhoods. Neighbourhoods are
temporary and permanent.
the building blocks for urban recovery. A larger urban
It is not the purpose of this reflection to address disaster, seemingly overwhelming, can be effectively
the features and nuances of the potential types of set- addressed one neighbourhood at a time.
tlements. What we consider here is the situation of a
In the urban setting, compounding issues which
destroyed, primarily residential, urban area and the
take on elevated importance in the shelter and settle-
general awareness that should be considered in an
ments dynamic include:
effective humanitarian response that will both assure
good habitable space and properly envision recovery Livelihoods: In cities people are depended upon a
and future growth. job to provide the resources necessary to buy food,
services, housing and other needs. Removing them
In a large disaster, the primary trigger to action is
from their neighbourhood usually removes them
displacement of people from their living space and the
from their work/living.
attending problems that accompany such displacement.
Dense population: Living space is much more
Although there may be many intermediate objectives,
controlled, confined and limited; buildings are much
transitional goals or temporary situations, the universal,
closer together and “affect” one another; habitable
overriding goal is to get those displaced back to their
space goes up rather than out; and people are more
living space (settlement, housing, etc.) in a timely
often renters that owners. All these factors make
manner and to a state where the family asset base
sheltering and recovery difficult.
and the settlement service base can sustain community
Land conflicts: Land use is more regulated;
recovery with minimal outside assistance. This goal
services to land (water, electricity, pathways,
must be our prime objective and is the necessary ingre-
etc.,) are essential and transcend land ownership
dient for sustainability and resilience.
boundaries; land related activities, particularly
There are three important “habitable space” factors where construction is involved, often have to be
to note in dealing with displacement. The first factor is coordinated and phased with similar activities of the
that a person is not in “recovery” until that person is owners and users of surrounding land.
at least in a stabilised, non-transient living environment
“...Shelter and
Settlements
are inextricably
linked and can no
longer be treated
as separate units
or responses...”
Muzafarabad, Pakistan 2005 after an earthquake.

118
“A
neighbourhood
is a settlement
which has a
socially defined
closeness of the
inhabitants in the
common aspects
of their lives”
Gonaives, Haiti 2008, following extensive flooding.

Social issues: Political and economic issues; land

and property rights; and the interacting impacts of
all these issues become important, are germane,
and need to be programme components.
Poverty: Urban areas often have large
concentrations of impoverished families who
live in poor housing, hazardous conditions, and
infrastructure and service deficient communities, all
of which makes these populations highly vulnerable
and difficult to assist.
The operational unit – the
Neighbourhood
A neighbourhood is a settlement which, in addition
to having a “territory”, has a socially defined closeness
of the inhabitants in the common aspects of their lives.
This closeness may reflect economic, physical, cultural,
ethnic, religious, political, administrative and other
characteristics. It is important to prioritise these factors
in terms of their importance in any given situation.
The neighbourhood is the rough urban equivalent of a
village in the rural setting, but significantly more com-
plicated.
All the necessary elements of the settlement
(discussed below) become interrelated, interconnected,
and interacting with shelter/housing at the centre. The
neighbourhood approach becomes the mechanism
in which we optimise all the competing factors and
create the transformative platform to re-establish, and
hopefully improve, the “living space” of the population
after the disaster. When linked together through an
emergency urban plan, neighbourhoods can serve as
the basis of recovery in extended or complicated urban “Debris removal is decisive... Drainage is essential...”
areas. Photo: Joseph Ashmore
B.4 Opinions
Elements defining a neighbourhood b) Affinity relationships

What constitutes a “neighbourhood” for disaster The neighbourhood must have a reasonable level of
assistance shelter and settlements purposes is driven social cohesion.
first by context and then by need, to provide a safe, The cohesion requirement is necessary for community
functional, and appropriate living space, in accordance involvement and participation in the neighbourhood
with our universal humanitarian principles. There are design, planning, and implementation functions, all
no “magic bullet” factors (although “driving” factors of which are important factors in our participatory
usually emerge). The ultimate specification of the set- shelter and settlements programming.
tlement is an optimisation process that will define the Major consideration should be given to
neighbourhood and which may involve the balancing neighbourhoods where compatible organisations
of competing factors. This article does not address how (also potential partners) have already been working.
to weigh factors, how to identify “driving” factors, Any neighbourhood programme will require the
or how to use the resulting neighbourhood attributes use of local assistance to get the job done. In this
in the design function. Those are topics for separate regard have available community organisations and
articles. Also the process will differ if one is working group that can mesh with the programme can be
in an existing neighbourhood as distinguished from very valuable.
creating a new neighbourhood. In this reflection we c) Minimum elements and considerations
consider the situation of working in an existing neigh- Land mapping, social occupancy mapping and
bourhood. rights
Mapping and a community, participatory, land
The neighbourhood is defined both by physical at-
boundary identification process is critical. This
tributes and affinity relationships. Factors important
approach is necessary to: properly establish land
to the determination of the neighbourhood for our
rights for the project; have the agreed parcel
disaster assistance response purposes include:
boundaries before demolition; and, provide a basis
a) Physical attributes for the settlement design, particularly in improving
The neighbourhood should have reasonably defined roadways, housing accesses, drainage and land use.
and determinable natural/physical boundaries. The mapping process must be done is such a way as
Boundaries may be set by streets, railroad lines, to fill the gaps in the country’s civil land system and
etc.; by building types (residential vs. business); by to be compatible with that system as it develops.
natural features (ravines, waterways, etc.); and/or Debris removal, land stabilisation and drainage
legal requirements (zoning, land use policy, etc.). Debris removal is decisive. Even relief response
The important aspect is that the area determined cannot effectively start without rubble removal,
is reasonably compact, can be serviceable without much less recovery or development. Drainage is
complex, complicated or involved physical or social essential because it will dictate the quality of the
infrastructure. storm water (which affects floods) and waste water
For most shelter and settlement programs, the systems (which affects sanitation). Drainage will be
neighbourhood should be largely residential in tied to a mitigating topology which is also important
character, since sheltering care is the primary focus because most of the places in which responders
of our humanitarian response. serve involve hazardous lands.
Neighbourhoods, many times, will have elements Pathways and roadways
of business, industry or dominant uses. The more Pathways and roads must be one of the first design
these elements are brought into the programme elements because they are hard to establish after the
design the more complicated and conflicted the fact. These elements, have a big impact on drainage
situation can become. One should consider edge as so must be considered contemporaneously
matching to these factors rather than bringing them with land topology work and drainage. Road and
in as major elements. These elements, if important pathways also play an important role in access and
may have to be considered as part of any livelihoods egress in the case of emergencies and evacuations.
component. Economic viability and livelihoods
As noted, livelihoods are a critical element in
urban disasters, so the settlement response must
take account of minimal economic considerations,
including the long term economical viability of
the settlement location and the context after the
disaster.
LeGrand L. Malany, P.E.

Independent shelter and Settlements Advisor,
USAID/OFDA.
The neighbourhood is defined both by physical attributes
and affinity relationships.
120
Annexes Shelter Projects 2009 Country Index
Annexes
The Annexes contain a list of case studies in the four editions of this report to date, tables of
conversion factors and further reading. The further reading includes some of the public docu-
mentation on which Shelter Projects 2011-2012 was based on.
Annex 1 - Index - by country 123

Annex 2 - Conversion Tables 126
Annex 2 - Further reading 127
Country Index Annexes
122
Annex 1 - Index - by Country
This index is to help readers find case studies of shelter projects from Shelter Projects 2008, Shelter Projects
2009, Shelter Projects 2010 and Shelter Projects 2011-2012. It is sorted by country and by date. Projects are colour
coded as follows:
- Case studies
- Updates - Follow up reports on existing case studies, and technical analyses of individual shelter designs.
- Historic case studies - Case studies of projects before 2000
Afghanistan, 2012, Conflict Returns .................... Case study .............................. Shelter Projects 2011–2012
Afghanistan, 2010, Conflict ................................. Update .................................... Shelter Projects 2010
Afghanistan, 2009, Conflict ................................. Case study .............................. Shelter Projects 2009
Afghanistan, 2002, Conflict ................................. Case study .............................. Shelter Projects 2008
Afghanistan, 2002, Conflict ................................. Update .................................... Shelter Projects 2009
Azerbaijan, 1992, Conflict ................................... Case study .............................. Shelter Projects 2008
Algeria, 1980, Earthquake ................................... Case study .............................. Shelter Projects 2009
Bangladesh, 2009, Cyclone .................................. Overview ................................. Shelter Projects 2009
Bangladesh, 2007, Cyclone .................................. Overview & Case study ............ Shelter Projects 2009
Bangladesh, 1975, Conflict .................................. Case study .............................. Shelter Projects 2008
Burkina Faso, 2012, Conflict ................................ Case study .............................. Shelter Projects 2011–2012
Chile, 2010, Earthquake ...................................... Case study .............................. Shelter Projects 2010
China, Sichuan, 2008, Earthquake ....................... Case study .............................. Shelter Projects 2009
Colombia, 2010, Floods ....................................... Case study .............................. Shelter Projects 2011–2012
Côte d’Ivoire, 2010–2011, Post-election Crisis...... Overview & 2 Case studies ...... Shelter Projects 2011–2012
DRC, 2009, Conflict ............................................. Case study .............................. Shelter Projects 2009
DRC, 2002, Volcano ............................................ Update .................................... Shelter Projects 2011–2012
DRC, 2002, Volcano ............................................ Case study .............................. Shelter Projects 2008
Eritrea, 1998, Conflict .......................................... Case study .............................. Shelter Projects 2008
Eritrea, 2004, Conflict .......................................... Update .................................... Shelter Projects 2009
Ethiopia, 2012, Conflict and Drought................... Case study .............................. Shelter Projects 2011–2012
Ethiopia, 2011, Sudanese Conflict ....................... Case study .............................. Shelter Projects 2011–2012
Georgia, 2008, Conflict ....................................... Case study .............................. Shelter Projects 2009
Grenada, 2010, Hurricane.................................... Case study .............................. Shelter Projects 2010
Guatemala, 1976, Earthquake ............................. Case study .............................. Shelter Projects 2009
Haiti, 2010, Earthquake ....................................... Overview & 3 Case studies ...... Shelter Projects 2011–2012
Haiti, 2010, Earthquake ....................................... Overview & 6 Case studies ...... Shelter Projects 2010
Haiti, 2010, Earthquake ....................................... Technical ................................. Transitional shelters: 8 designs
Country Index Annex 1 - Index - by Country Annexes
Haiti, 2008, Floods ............................................... Case study .............................. Shelter Projects 2009
Haiti, 1982, Hurricane .......................................... Case study .............................. Shelter Projects 2009
Honduras, 1998, Hurricane .................................. Case study .............................. Shelter Projects 2008
Honduras, 1974, Hurricane .................................. Case study .............................. Shelter Projects 2009
India, 1977, Cyclone ............................................ Case study .............................. Shelter Projects 2009
India, 1977, Cyclone ............................................ Case study .............................. Shelter Projects 2008
India, 1971, Conflict ............................................ Case study .............................. Shelter Projects 2008
India, 2001, Earthquake ....................................... Case study .............................. Shelter Projects 2008
Indonesia, 2009, Earthquake................................ Overview & 2 Case studies ...... Shelter Projects 2010
Indonesia, 2009, Earthquake................................ Technical ................................. Transitional shelters: 8 designs
Indonesia, 2006, Earthquake................................ Overview & 2 Case studies ...... Shelter Projects 2008
Indonesia, 2004, Earthquake/Tsunami .................. Case study .............................. Shelter Projects 2008
Indonesia, 2004, Earthquake/Tsunami .................. Technical ................................. Transitional shelters: 8 designs
Ingushetia, 1999, Conflict .................................... Case study .............................. Shelter Projects 2008
Italy, 2009, Earthquake ........................................ Overview & Case study ............ Shelter Projects 2009
Japan, 2011, Earthquake and Tsunami ................. Case study .............................. Shelter Projects 2011–2012
Kenya, Dadaab, 2011, Famine / Conflict .............. Case study .............................. Shelter Projects 2011–2012
Kenya, Dadaab, 2009, Conflict ............................ Update .................................... Shelter Projects 2009
Kenya, Dadaab, 2008, Conflict ............................ Update .................................... Shelter Projects 2009
Kenya, Dadaab, 2007, Floods/Conflict.................. Case study .............................. Shelter Projects 2008
Kenya, 2008, Conflict .......................................... Case study .............................. Shelter Projects 2008
Kenya, 2008, Conflict .......................................... Update .................................... Shelter Projects 2009
Kyrgyzstan, 2010, Conflict ................................... Case study .............................. Shelter Projects 2010
Lebanon, 2007, Conflict ...................................... Case study .............................. Shelter Projects 2011–2012
Lebanon, 2011, Conflict ...................................... Case study .............................. Shelter Projects 2011–2012
Liberia, 2007, Conflict.......................................... Case study .............................. Shelter Projects 2008
Liberia, 2007, Conflict.......................................... Update .................................... Shelter Projects 2009
Sri Lanka, 2009, Conflict ...................................... Case study .............................. Shelter Projects 2010
Sri Lanka, 2007, Conflict ...................................... Update .................................... Shelter Projects 2009
Sri Lanka, 2007, Conflict ...................................... Case study .............................. Shelter Projects 2008
Sri Lanka, 2004, Tsunami ..................................... Case study .............................. Shelter Projects 2008
Madagascar, 2012, Tropical Storm ....................... Case study .............................. Shelter Projects 2011–2012
Malawi, 2009, Earthquake ................................... Case study .............................. Shelter Projects 2010
Mozambique, 2007, Cyclone ............................... Case study .............................. Shelter Projects 2008
Mozambique, 2007, Cyclone ............................... Case study .............................. Shelter Projects 2010
Myanmar, 2008, Cyclone ..................................... Case study .............................. Shelter Projects 2009
Myanmar, 2008, Cyclone ..................................... Two Case studies .................... Shelter Projects 2010
Nicaragua, 2007, Hurricane ................................. Case study .............................. Shelter Projects 2011–2012
Nicaragua, 1972, Earthquake ............................... Case study .............................. Shelter Projects 2009
Nicaragua, 1972, Earthquake ............................... Case study .............................. Shelter Projects 2008
Pakistan, 2010, Floods ......................................... 2 Case studies ......................... Shelter Projects 2011–2012
Pakistan, 2011, Floods ......................................... 2 Case studies ......................... Shelter Projects 2011–2012
Pakistan, 2010, Floods ......................................... Overview & 2 Case studies ...... Shelter Projects 2010
Pakistan, 2010, Floods ......................................... Technical ................................. Transitional shelters: 8 designs
Pakistan, 2009, Conflict ....................................... Case study .............................. Shelter Projects 2010
Pakistan, 2005, Earthquake.................................. Overview & 2 Case studies ...... Shelter Projects 2008
Peru, 2012, Flooding and Landslides .................... Case study .............................. Shelter Projects 2011–2012
Peru, 2007, Earthquake ....................................... Overview & 3 Case studies ...... Shelter Projects 2008
Peru, 2007, Earthquake ....................................... Technical ................................. Transitional shelters: 8 designs
Philippines, 2011, Cyclone ................................... 2 Case studies ......................... Shelter Projects 2011–2012
Philippines, 2010, Cyclone ................................... Case study .............................. Shelter Projects 2010
Gaza, Palestine, 2009, Conflict ............................ Case study .............................. Shelter Projects 2009
Romania, 2010, Floods ........................................ Case study .............................. Shelter Projects 2010
Rwanda, 2008, Conflict ....................................... Case study .............................. Shelter Projects 2008
Rwanda, 2008, Conflict ....................................... Update .................................... Shelter Projects 2009
Somalia, 2011, Famine / Conflict.......................... Case study .............................. Shelter Projects 2011–2012
Somalia, 2009, Conflict........................................ 2 Case studies ......................... Shelter Projects 2009
Somalia, 2008, Conflict........................................ Case study .............................. Shelter Projects 2009
Somalia, 2007, Conflict........................................ Case study .............................. Shelter Projects 2008
Republic of South Sudan, 2011, Conflict .............. Case study .............................. Shelter Projects 2011–2012
Sudan, 1985, Conflict .......................................... Case study .............................. Shelter Projects 2008
124
Sudan, Darfur, 2004, Conflict .............................. Case study .............................. Shelter Projects 2008
Sudan, Darfur, 2004, Conflict .............................. Case study .............................. Shelter Projects 2009
Thailand, 2011, Bangkok Floods .......................... Case study .............................. Shelter Projects 2011–2012
Thailand, 1979-1980, Conflict ............................. Case study .............................. Shelter Projects 2008
Tonga, 2010, Tsunami ......................................... Case study .............................. Shelter Projects 2010
Tonga, 1982, Cyclone .......................................... Case study .............................. Shelter Projects 2008
Tajikistan, 2010, Earthquake ................................ Case study .............................. Shelter Projects 2010
Tunisia, 2011, Conflict in Libya ............................ Case study .............................. Shelter Projects 2011–2012
Turkey, 1976, Earthquake .................................... Case study .............................. Shelter Projects 2009
Uganda, 2007, Floods .......................................... Case study .............................. Shelter Projects 2009
UK, 1945, Conflict ............................................... Case study .............................. Shelter Projects 2009
USA, 1906, Earthquake........................................ Case study .............................. Shelter Projects 2010
USA, (Chicago), 1871, Fire ................................... Case study .............................. Shelter Projects 2011–2012
Vietnam, 2009, Typhoon ..................................... Case study .............................. Shelter Projects 2010
Vietnam, 2004, Typhoon ..................................... Technical ................................. Transitional shelters: 8 designs
Yugoslavia (ex republic of:), 1963, Earthquake ..... Case study .............................. Shelter Projects 2009
Country Index Annex 2 - Conversion Tables Annexes
Annex 2 - Conversion Tables
These tables are included to help readers convert the measurements in the Bills of Quantities.
The data on this page is all rounded to 4 significant figures. Penny sizes are rounded to the nearest mm.
Length
Imperial 1 inch (in.) 1 feet (ft.) = 12in. 1 yard (yd.) = 3ft. = 36in. 1 mile = 1760yd.
Metric 25.4mm = 2.54cm 304.8mm 0.9144m 1.609km

For equivalence tables in timber sizing see UNOCHA / IFRC / CARE International, Timber
Area
Imperial 1 square feet (sq. ft.) 1 square yard (yd2) = 9sq. ft. 1 acre
Metric 0.0929m2 0.8361m2 4046.9m2

1 perch = 30.25yd2 1 acre = 4,840yd2 1 hectare = 10 000m2
Volume
Imperial 1 cubic feet (ft3) 1 cubic yard (yd3)
Metric 28.32 litres = 0.02832m3 0.7646m3

1 US liquid gallon = 3.785 litres 1 US dry gallon = 4.405 litres 1 imperial (UK) gallon = 4.546 litres
Weight
Imperial 1pound (lb) Ton (UK, long ton) Ton (US, net ton, short ton)
Metric 0.4536kg 1.1016MT = 1016kg 0.9072MT = 907.2kg

Note that there are several different imperial systems of weights. We quote the British imperial ton as in the
Weights and Measures Act of 1824, and the United States customary system. Additional useful conversions are:
1lb = 16 Ounces (Oz.) 1 stone = 14 pounds (lb.)
1 hundredweight (cwt.) - UK = 112lb. 1 hundredweight (cwt.) - US = 100lb.
Nails - “penny sizes”
Penny size 2d 3d 4d 6d 8d 10d 16d 20d 40d 50d 60d 100d
Imperial
Inches 1 1¼ 1½ 2 2½ 3 3½ 4 5 5½ 6 10
Nearest 25 32 38 51 64 76 89 102 127 140 152 254

Metric length
(mm)
126
Annexes Shelter Projects 2009 Further reading
Annex 2 - Further Reading

In compiling this edition of Shelter Projects, we have drawn on key informant interviews and a variety of
sources. The published sources are listed below under General statistics and websites.
We also include a list of background documents - these are key shelter texts in which readers can find further
reading on many of the shelter related issues raised by these case studies. Some of them are directly referred to
in the text.
General statistics and case study related information

CRED, EM-DAT disaster database
www.emdat.be
Global database of world disasters.
IFRC, World Disasters Report, 2011 - Focus on hunger and malnutrition
IFRC, World Disasters Report, 2010 - Urban Risk
Available from www.ifrc.org
Annual report providing a global overview of disaster trends.
IFRC, reports
Available from reliefweb.int
These include Appeals, Operational updates, final, mid year and annual reports.
IDMC/ NRC, Internal Displacement Global Overview 2011: People internally displaced by conflict and violence,
April 2012
Available from www.internal-displacement.org
Overview of IDP issues.
OCHA, Sitreps,
Available from reliefweb.int
Situation reports on major responses. Where possible these have been used for response timeline data and figures
for the number of houses damaged.
OCHA appeal documents
www.unocha.org/cap
Financial appeals, action plans and reviews with narratives for OCHA coordinated responses.
UNHCR, Global Trends 2011
Available from www.unhcr.org
Annual summary of global statistics relating to refugees and persons of concern to UNHCR.
Websites
www.disasterassessment.org
A site where members of the disaster management community can meet to exchange tools and case studies
related to disaster risk assessment.
http://www.humanitarianresponse.info
Humanitarian Response aims to be the central website for Information Management tools and services, enabling
information exchange among operational responders during either a protracted or sudden onset emergency.
IFRC/ICRC Emergency relief items catalogue - website
procurement.ifrc.org/catalogue/
Detailed specifications of all items commonly used by IFRC and ICRC.
IFRC Shelter video channel
bit.ly/ifrcshelter
Red Cross Red Crescent videos related to emergency shelter.
www.reliefweb.int
Up to date information on complex emergencies and natural disasters as well as an archive of information, field
reports and situation reports from emergencies since 1996. OCHA situation reports (sitreps) and IFRC appeal
documents and operations updates have been of particular use in compiling these case studies.
Further reading Annex 2 - Conversion Tables Annexes
SDC - Cash transfer Projects

www.sdc-cashprojects.ch/en/Home/Experiences/SDC_Cash_Transfer_Projects
Compilation of cash projects by SDC. Includes shelter case studies.
www.shelterlibrary.org
A library of free documents relating to transitional settlement and reconstruction.
www.sheltercluster.org
Home page of the global shelter cluster - the coordination mechanism for shelter responses. Contains links to
individual responses, including strategy documents.
www.sphereproject.org
Download the sphere handbook, find information on trainings and other activities from the Sphere Project. The
Sphere Project aims to improve the quality of humanitarian assistance and the accountability of humanitarian
actors to their constituents, donors and affected populations.
UNHCR Data portals (data.unhcr.org/portfolio/):
data.unhcr.org/horn-of-africa
data.unhcr.org/MaliSituation
data.unhcr.org/syrianrefugees
Regional response portals for individual refugee related responses.
Background Documents
Mike Albu, The Emergency Market Mapping and Analysis Toolkit, 2010
Available from: www.emma-toolkit.org
A toolkit designed for generalists, as well as specialist staff on how to conduct an emergency market mapping
analysis.
Sultan Barakat, HPN Network paper 043, Housing reconstruction after conflict and disaster, ODI, 2003
Available from: www.odihpn.org/documents/networkpaper043.pdf
Review of housing reconstruction experiences and approaches.
Camp management project, Camp Management Toolkit, 2008
Available from: www.nrc.no/camp
A comprehensive field manual for camp management organisations and stakeholders involved in camp operations.
Corsellis and Vitale, Transitional Settlement: Displaced Populations, Oxfam publishing, 2005
Available from: www.shelterlibrary.org
Guidelines for the strategic planning and implemention of settlement responses for displaced populations.
CRS, Managing Post-disaster (Re)-Construction Projects, 2013
Available from : www.crsprogramquality.org
Step-by-step guide on management of owner-driven and contractor-built construction.
CRS, Learning From the Urban Transitional Shelter Response in Haiti - 2012
www.crsprogramquality.org
CRS’s experiences in planning and implementing its urban transitional shelter response in Port-au-Prince, Haiti.
CRS, Learning From Urban Transitional Settlement Response in the Philippines: Housing, Land and Property - 2012
Available from: www.crsprogramquality.org
Reflections and questions to ask to improve future transitional settlement and land programs.
IASC, Shelter Centre, Selecting NFIs for shelter, 2008
Provides information, case studies and guidance on how to choose the best items to distribute to those affected
by natural disaster or conflict.
ICRC/IFRC Guidelines for cash transfer programming, 2007
Available from: www.ifrc.org
Provides information on when and how to distribute cash in disaster response.
IFRC, Guidelines for assessment in emergencies, 2008
Practical information and guidance on how to conduct assessments in emergencies.
IFRC Owner Driven Housing Reconstruction Guidelines (ODHR), 2010
Guidance on the planning and implementation of assisted self help reconstruction projects.
128
Annexes Shelter Projects 2009 Further reading
IFRC, Oxfam GB, Plastic sheeting, 2007

Available from: www.plastic-sheeting.org
A guide to the use and specification of plastic sheeting in humanitarian relief, 2007. An illustrated booklet on
when and how to use plastic sheeting most effectively in emergencies.
IFRC, The IFRC shelter kit, 2010
A guide on the IFRC shelter kit and how to use it.
IFRC, Transitional Shelter: Eight Designs, 2011
Available from: www.ShelterCaseStudies.org
A review of risks in shelter construction and detailed structural analysis of eight different transitional shelters
designs that have been used in the field in large scale projects.
NRC, Shelter Centre, Urban Shelter Guidelines, 2010
General guidance for urban humanitarian response.
Shelter Centre, UN, DfID, Shelter after disaster - Strategies for transitional settlement and reconstruction, 2010
A book containing information and guidance on how to agree strategies for reconstruction after natural disasters.
contains description of the types of shelter programmes that organisations can implement.
Sphere Project, Sphere - Humanitarian charter and minimum standards in humanitarian response, 2011
Available from: www.sphereproject.org
Contains consensus standards agreed among major humanitarian organisations for key sectors, including shelter
and settlement. It also contains actions, indicators and guidance notes as to whether standards have been achieved.
UNDRO, (now UNOCHA), Davis, I., Shelter After Disaster, Guidelines for Assistance, 1982
Guidelines and description of shelter provision in all aspects of natural disasters (from preparedness to reconstruction).
UNHABITAT, UNHCR, IFRC, Shelter Projects 2010
Case studies of shelter projects implemented between 1906 and 2010. Includes many different types of response.
UNHABITAT, IFRC, Shelter Projects 2009
Case studies of shelter projects implemented between 1945 and 2009. Includes many different types of response.
UNHABITAT, IFRC, UNHCR, Shelter Projects 2008
More case studies. Stee also UNHABITAT, IFRC, Shelter Projects 2009.
UNHABITAT, Land and Natural Disasters - Guidance for Practitioners, 2010
Available from: www.disasterassessment.org or from www.unhabitat.org
A book containing guidance on land issues following natural disasters.
UNHCR, Handbook for Emergencies, Third Edition, 2007
Available from: www.unhcr.org
A book containing guidance on the management and all the key sectors in refugee emergencies.
UNICEF, Compendium of Temporary Learning Spaces (TLS): Design and practice in emergencies 2011
Designs and guidance on the construction of temporary learning spaces.
UN/OCHA, Tents - A guide to the use and logistics of tents in humanitarian relief, 2004
A booklet describing when and how to use tents as well how to support those living in them to best adapt them
to meet their needs.
UN/OCHA / IFRC / CARE International, Timber as a construction material in humanitarian operations, 2009
Available from: www.humanitariantimber.org
An illustrated booklet on how to source and use timber for the construction of basic structures.
Further reading Annex 2 - Conversion Tables Annexes
130

Shelter Projects2011/2012

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Shelter Projects2011/2012

Uploaded by

Copyright:

Available Formats

SHELTER PROJECTS

Shelter Projects 2011–2012

Available online from www.ShelterCaseStudies.org

Copyright for this book is retained by IFRC, UN-HABITAT and UNHCR.

Front Cover: A transitional shelter in the Philippines being relocated by 20 people.

Back cover left to right:

Section A – Case Studies

Including a case study in this

As a result of the projects being

Global shelter need Natural disasters Countries have differing

Where to find different types of response in the case studies

Short-term land, house

Guidelines / materials /mass

A.1 – Afghanistan – 2012 x x x x x x

* Although there are no examples in this edition of Shelter Explanation of columns:

Recurring themes Scale using the cluster approach to

Minimum materials and labour cost

5000 Beneficiary contribution

A.18 – Madagascar – 2012

A.2 – Burkina Faso – 2012

A.26 – Philippines – 2012

A.7 – DRC Goma – 2002

A.20 – Pakistan – 2010

A.22 – Pakistan – 2011

A.9 – Ethiopia – 2012

A.33 – USA – 1871

A.11 – Haiti – 2010

A.24 – Peru – 2012

Costing shelter projects Terminology Interpret and contribute

The general desire was to get out of buildings into tents or

The most remarkable concourse of these campers was that in

Shelter in response to the Lisbon earthquake of 1755.

1755 German copperplate image, The Ruins of Lisbon:

A.1 Afghanistan – 2012 – Conflict Returns A.17 Lebanon – 2011 – Conflict

A.1 Afghanistan – 2012 – Conflict Returns

Country: Project timeline

Strengths and weaknesses 8 It was challenging to identify the most vulnerable

Implementation had urgent sheltering needs and

After signing a memorandum of Instead of providing fixed Households were expected to

A.2 Burkina Faso – 2012 – Conflict

Country: Project timeline

Before the conflict Beneficiary selection Implementation

An overall view of one section of the camp.

Field monitors checked the Agencies working on water and

A.3 Colombia – 2010–2011 – Floods

Country: Project timeline

The main carpenter and his assis-

A.4 Côte d’Ivoire – 2010–2011 – Post-election Crisis

In Western Côte d’Ivoire, the

Government response The coordination team Closing of the

Goal for 2012 Result achieved

A.5 Côte d’Ivoire – 2010–2011 – Post–Electoral Crisis

Country: Project timeline

Strengths and weaknesses 8 Difficulties arose in validating beneficiary lists as

After the conflict

Roofing with corrugated irons. (Toa-Zéo) Drying mud bricks. (Niambly)

They were required to continue Coordination also allowed or- Staffing

A.6 Côte d’Ivoire – 2010–2011 – Post–Electoral Crisis

Country: Project timeline

Strengths and weaknesses electoral crisis, or as a result of a previous crisis.

Background For 40 families with land issues, Self help groups

Coordination Staffing and structure Shelter committees distributed

A.7 Democratic Republic of Congo – 2002 – Volcano

Country: Project timeline