Land Use Policy 57 (2016) 526–539

Contents lists available at ScienceDirect

Land Use Policy

journal homepage: www.elsevier.com/locate/landusepol

Joint problem framing in sustainable land use research

Experience with Constellation Analysis as a method for inter- and
transdisciplinary knowledge integration
Martina Schäfer ∗ , Melanie Kröger
Center for Technology and Society, Technische Universität Berlin, Hardenbergstr. 16-18, Sekr. HS 1, D-10623 Berlin, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Land use is an important field of interest regarding sustainability transformations. Research projects
Received 16 October 2015 which deal with the multiple dimensions of sustainable land use usually apply an inter- and transdisci-
Received in revised form 21 April 2016 plinary design and are confronted with challenges of integrating heterogeneous knowledge.
Accepted 13 June 2016
In this paper we refer to experience we had during the ELaN project, which followed a systemic
Available online 23 June 2016
approach by linking research on water and land management. Due to this relatively uncommon approach
it was necessary to bring together scientists and practical actors as well as distributed knowledge from
Keywords:
different areas of expertise. Considering the heterogeneity of the actors it was of great importance to
Knowledge integration
Problem framing
establish a shared understanding of the research problem the project was to deal with during the initial
Mutual understanding phase. For this step the method of Constellation Analysis (CA) was applied: a visualisation and analysis
Transdisciplinarity tool which aims at joint problem framing by focusing on the dominant elements of a social-ecological
Interdisciplinarity problem and their relations in a discursive process. Due to the size of the project team and the necessity
Sustainable land use to involve a broad range of actors, a group of scientists led the iterative process and prepared CA drafts
Land management which were validated by practitioners. This design can be categorised as ‘consulting’ rather than ‘par-
Fenlands ticipatory’ transdisciplinarity. Proceeding this way can be seen as a compromise between more intense
Constellation Analysis
forms of transdisciplinary exchange and forms that are manageable when considering time and resource
constraints in third-party funded projects.
CA has proven to be a suitable tool for organising processes of mutual understanding between hetero-
geneous actors and fostering social integration in inter- and transdisciplinary research groups. In ELaN the
main benefit of the process was an adjustment and enrichment of problem framing which was formulated
in the project proposal thus contributing to integrated system knowledge as a basis for the interdisci-
plinary project consortium and involved practitioners. The insights gained during this process led to
changes in the design of some of the sub-projects as well as the targeted end products. This experience
confirms the importance of a structured process of joint problem framing in inter- and transdisciplinary
projects, especially for thematic fields of such high complexity as land use research.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction
Land use research exemplifies the challenges that have arisen
from the sustainability paradigm. While monitoring and modelling
of the ecological impacts of land cover change prevailed in the past,
currently, a more integrative understanding that moves beyond
the limits of disciplinary knowledge and sectoral perspectives is
being pursued (Zscheischler and Rogga, 2015). Land is central in
human-nature interactions. But its functions, such as land-based
∗ Corresponding author.
E-mail address: schaefer@ztg.tu-berlin.de (M. Schäfer).
production (food and non-food), space for recreation and the provi-
sion of ecosystem services, are being greatly threatened by current
trends such as climate change, globalisation, demographic changes
and energy politics (Pérez-Soba et al., 2008; Zscheischler and Rogga,
2015). To be able to deal with these complex and uncertain socio-
ecological problems and challenges usually demands an inter-
and transdisciplinary research design that enables integration of
knowledge from different disciplines as well as experience from
practice. This so called “mode 2” research supplements the tra-
ditional disciplinary “mode 1” research and implies that socially
robust knowledge for solving socially caused, complex problems
is increasingly being co-produced by scientific researchers, pol-
icy makers, stakeholders and private actors (Brand, 2000; Gibbons

http://dx.doi.org/10.1016/j.landusepol.2016.06.013
0264-8377/© 2016 Elsevier Ltd. All rights reserved.
 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
et al., 1994; Hinkel, 2008; Hirsch Hadorn et al., 2008; Tress et al.,
2006; Tress et al., 2007).
In general, transdisciplinarity is a particular type of co-
production of knowledge which is able to contextualize scientific
and real world knowledge (Scholz et al., 2000: 478) by ideally
producing three types of knowledge: system, target and transfor-
mation knowledge (ProClim, 1997: 15). While a commonly shared
definition of transdisciplinarity is still lacking, some aspects of
research practice can be seen as common ground: transdisciplinary
research deals with complex real-world or societal problems and it
is conducted in cooperation between scientific research and soci-
ety, which are ideally envisioned as being enabled to mutually learn
from each other (Jahn et al., 2012; Scholz et al., 2000; Zierhofer and
Burger, 2007; Zscheischler and Rogga, 2015). Integration can be
seen as one of the most important challenges of transdisciplinary
endeavours, as due to their very nature, such projects have to cope
with thematic as well as problem- or product-oriented integration
of knowledge and social integration of scientists and practitioners
(Bergmann et al., 2012; Hinkel, 2008; Scholz et al., 2000; Zierhofer
and Burger, 2007). The involved scientific and practical experts usu-
ally are driven by different rationalities and logics of action and
locate problems in their particular “world of relevance” (Limoges
1993). Scavarda et al. (2006) point out that every expert has his or
her own mental causal map, representing his or her beliefs about
the causes of a problem, its main drivers and possible solutions. This
diversity in perspectives must be taken into account while identi-
fying and structuring a problem and while developing and testing
means to deal with it. Several authors agree that the first step in
mutual learning and integration is to acknowledge the diversity of
perspectives and to explore and clarify their differences (Giri, 2002;
Loibl, 2006; Tress et al., 2006; Schäfer et al., 2010).
Both knowledge integration and mutual learning often occur
more or less implicitly in the process of inter- and transdisciplinary
research projects. Usually it is taken for granted by funding agen-
cies as well as researchers involved in such projects. Zierhofer
and Burger (2007) conclude from an evaluation of 16 inter- and
transdisciplinary projects that problem oriented integration of dif-
ferent knowledge is rarely approached systematically and often is
not even recognized as a methodological challenge. Truffer (2007)
and Hunecke (2011) point out that knowledge integration is more
than simply a technical or organisational problem but is also an
active social process of negotiation and construction which has
to be planned, designed and organized in an active manner and
with appropriate methods. Several authors agree that a broad and
coherent debate on the methodology of integrated knowledge pro-
duction is still lacking (Bergmann et al., 2012; Defila and Di Giulio,
2015), although a number of compilations of methods of integra-
tion have been produced during the last years (Niederberger and
Wassermann, 2014; Bammer, 2013; Bergmann et al., 2012).
Various authors have pointed out that inter- and transdisci-
plinary research projects should aim at knowledge integration
from the very beginning and throughout the entire research pro-
cess (Bergmann et al., 2012; Lang et al., 2012). Ideally, within
the initial phase of a project, the leading questions and common
research goals should be defined and a shared view about the sys-
tem and its dynamics generated so that the various participants
can focus on “one and the same problem and [. . .] synthesize the
information sampled in an effective manner” (Scholz, 2000: 16).
Joint problem framing,1 which integrates complementary aspects
1
The terms problem framing (Rossini 2009), problem structuring (Scholz et al.,
2009; Pohl and Hirsch Hadorn, 2008a,b) and problem formulation (Bergmann et al.,
2005) are used with a rather similar meaning in the literature on transdisciplinary
research processes. ‘Joint’ and ‘shared’ problem framing are also used as synonymous
terms.
527
of system knowledge and gives an impression of the range of dif-
ferent perspectives involved, seems to be an ideal starting point
for a transdisciplinary research process (Pohl and Hirsch Hadorn,
2008b). Tress et al. (2007) refer to the main barriers for knowledge
integration in land use research and identify joint problem formu-
lation as a crucial step for integrative research. Jahn et al. (2012:
5) point out that arriving at a shared understanding of a problem
in a reflexive, methodically-guided process is the only way that
(diverging) expectations among participants (both from science
and society) regarding desired research outcomes can be managed
successfully.
This paper seeks to take up the gap in the literature regard-
ing how processes of knowledge and social integration can be
facilitated by applying specific methods and to reflect upon the
potentials and difficulties of such endeavours. It focuses on the first
step of structuring real-world problems within interdisciplinary
and transdisciplinary cooperation. We refer here to experiences
we had during the ELaN2 project, which dealt with questions
of integrated water and land management. To obtain a problem
formulation which could be shared by the scientific and prac-
tical partners involved, Constellation Analysis (CA) was applied.
CA is a visualisation and analysis tool which facilitates dialogue
between actors with different backgrounds. Feedback regarding
the difficulties and benefits of this process was obtained by work-
shop evaluation via questionnaires and qualitative interviews with
participants, thus allowing reflection on the added value of this
methodological step.
The paper is structured as follows: After a brief introduction of
the ELaN project (Section 2) and the CA method (Section 3), the
results of carrying out CA during the project are presented (Section
4). Section 5 discusses the suitability of the method for knowl-
edge integration and the quality of the transdisciplinary process.
The paper closes with conclusions for inter- and transdisciplinary
research processes as well as for sustainable land use research (Sec-
tion 6).
2. The ELaN project: sustainable land management in
Northeastern Germany
ELaN combined two thematic areas which have thus far rarely
been dealt with together: water and nutrient management, on one
side, and land use on the other. In doing so it sought to take up the
challenge of adopting a systemic view on sustainable land use in
the face of climate change. One of the main strands of the project
was examining the current German practice of discharging treated
waste water into rivers and surface water, which has negative con-
sequences in terms of regional loss of water and nutrients as well
as eutrophication of rivers and oceans. As part of ELaN, scientists
from a variety of disciplines (e.g. hydrology, soil science, limnol-
ogy, agriculture, regional planning, sociology, economy, political
science) were investigating whether the use of treated waste water
has the potential to serve as one element of sustainable water and
land use management. A central question was to estimate the risks
of using treated waste water for irrigation of specific sites with
regard to the subsequent quality of the soil and ground water. Par-
allel to exploring different options of water management, land use
options adapted to different ground water levels were also ana-
lysed. These investigations were embedded within consideration of
2
The ELaN project – Entwicklung eines integrierten Landmanagements durch
nachhaltige Wasser- und Stoffnutzung in Nord-Ostdeutschland (Developing an
integrated land management scheme for sustainable water and nutrients use in
Northeastern Germany) – was funded from 2011 to 2015 by the German Federal
Ministry for Education and Research. For further information see: <www.elan-bb.
de>.
528 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
current legal, institutional and political framework conditions, with
the objective of formulating recommendations for adapted gover-
nance measures. The potential of using treated waste water was
investigated at two sites in Northeastern Germany: areas which
were formerly used as sewage farms in the outskirts of Berlin and
(degenerated) fenlands in rural areas of Brandenburg. These sites
were chosen since they enabled dealing with different facets of
integrated water and land management as well as connected ques-
tions of economic valorisation and legal restrictions.
ELaN comprised an interdisciplinary team of 12 institutional
partners and approximately 45 scientists from different disci-
plinary backgrounds.3 It was structured in four thematic sub areas
(water and nutrient flows, land use, socio-economic governance
and knowledge integration) and 14 subprojects and was coordi-
nated by a team of five scientists who represented each of the four
subareas. One of the 14 subprojects was responsible for support-
ing and reflecting upon the inter- and transdisciplinary exchanges
that occurred throughout the whole research process. The authors
of the present paper were responsible for this subproject.
The choice of the overall research topic and formulation of the
proposal for the project were based on prior interaction between
researchers and practitioners. After the phase of joint problem
framing, the 14 subprojects worked on different research ques-
tions in the four subareas. The results were integrated into several
synthesis products which addressed different target groups: (a)
an administrative manual with recommendations regarding risk
assessment for the use of treated waste water, (b) a decision sup-
port system for farmers containing a variety of land use options
depending on different groundwater levels, (c) scenarios for land
use options depending on different framework conditions and (d)
governance recommendations for sustainable water and land use
options. Additionally, the main results of the project were sum-
marised in 11 core statements. Most of the synthesis products, such
as the manual, the scenarios and the governance recommendations,
mainly aimed at providing different target groups with integrated
system knowledge as a basis for decision making. These three
products also contributed towards generating target knowledge by
reflecting upon, for example, questions of risk management (e.g.
weighting the risk of polluting surface waters versus pollution of
soil and groundwater) or principles of land management (adapting
soils to intensive agriculture with the aim of economic optimisa-
tion versus adapting land use to natural conditions). The decision
support system aimed at providing target and transformational
knowledge by supporting farmers in their decisions regarding land
use options that could be adapted to the high groundwater levels
of fenland areas.4
This paper focuses on the first project phase of joint problem
framing. This methodological step aimed at integrating distributed
knowledge about the challenges of water and land management
at the two different sites and structuring the various perspectives
among the group of scientists as well as of different groups of prac-
titioners. The process of joint problem framing was meant to be
the starting point for mutual understanding between these dif-
ferent perspectives and to establish a common basis for further
cooperation.
Practitioners were involved in the project in different ways
and at varying levels of intensity. An advisory board with eleven
practitioners from different backgrounds (e.g. representatives
of agricultural and nature protection associations, experts from
3
ELaN was financed by the German Federal Ministry of Education and Research
(BMBF) between 2011 and 2015 in the program “Sustainable Land Management”.
More information www.elan-bb.de.
4
More information on the results of the different sub areas and synthesis products
can be found at www.elan-bb.de.
administration in water management and regional planning, an
entrepreneur in energy consulting, representatives of waste water
treatment plants) was consulted regularly to discuss intermedi-
ate results and integrate the specific knowledge and experiences
of these actors. At certain points in the project, a wider range of
practitioners was integrated via stakeholder workshops in order to
present elements of regional strategies for sustainable water and
land management and obtain feedback regarding the suggestions.
These workshops were complemented by bilateral consultations
with experts in certain fields. The participation of scientists and
practitioners in the methodological step of applying CA for joint
problem framing will be described in detail in Section 3.2.
3. Methods for facilitating and reflecting upon inter- and
transdisciplinary exchange
This section introduces the Constellation Analysis method,
which was applied in ELaN to achieve joint problem framing Sec-
tion (3.1), describes the involvement of scientists and practitioners
in carrying out CA Section (3.2) and explicates the methods which
were used to reflect upon the process of inter- and transdisciplinary
exchange in this phase of the project Section (3.3).
3.1. Constellation analysis
3.1.1. Methodological background of Constellation Analysis
Constellation analysis (CA) is a methodological approach that
has been designed to cope with complex situations requiring inter-
disciplinary cooperation and transdisciplinary solutions. It is a
‘middle range’ concept for bridging different areas of expertise
inside or outside of a scientific field and supporting mutual under-
standing (Schön et al., 2007). The development of CA was inspired
by Science and Technology Studies (STS) and Actor-Network Theory
(ANT), both of which stress the importance of dealing with hetero-
geneous elements like human actors, technology, signs and natural
elements equally, implying that none of the elements per se can
claim priority (Latour, 1987; Callon et al., 1986). The central the-
sis of Actor-Network Theory is that science and technology evolve
as a result of linking such heterogeneous elements in networks.
Technical artefacts as well as social actors, norms and institutions
are understood as being the subjects and results of interdepen-
dent relations in networks and are dealt with as having (potential)
agency (Ohlhorst, 2009). CA takes up the idea of “distributed action”
(Rammert, 2003) and attempts to map hetereogenous or hybrid
constellations. In the mapping process, differences between social
actors, technical elements, natural elements and conceptual and
communicative elements (or ‘systems of signs’) are highlighted,
but there are no a priori decisions made about their relevance for a
particular issue (Kruse, 2008: 49).
Constellation Analysis combines the following aspects: (1)
structured sampling of all elements that are relevant for a given
issue, (2) visualisation of how these elements relate and, based
on the latter, (3) a thorough analysis in written form. This map-
ping process results from a discursive process between actors with
different backgrounds regarding what they see as the most appro-
priate way to represent reality (Ohlhorst, 2009).
Discursive elements and visualisation are also part of other
methods which are used for knowledge integration. Discussion
among experts is seen as a basic means to “arrive at a shared
problem framing” (Defila and Di Giulio, 2015: 125). Integrating
knowledge implies more than simply sampling facts but also an
appreciation of different epistemologies and their underlying val-
ues, interests and worldviews (McDonald et al., 2009). McDonald
et al. (2009) describe several methods and techniques which aim
at supporting dialogue between participants engaged in transdis-
 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
ciplinary research, such as consensus conferences, the Delphi and
nominal group techniques, but most of these lack some form of
visualisation. Experience with CA shows that visualisation facil-
itates the discussion process and allows actors with different
backgrounds to better relate to certain aspects of a problem (Kruse,
2008: 48). Misunderstandings can be identified and – in the pro-
cess of attaining a shared understanding – complex interrelations
can be illustrated using different variations (Ohlhorst, 2009). On
the other hand, the trap of simplification can also be better avoided
by combining visualisation with a written analysis.
3.1.2. Basic elements and methodological steps for carrying out
CA
The nucleus of each CA is the visualisation of a specific con-
stellation constituted by various interrelated elements and having
a certain degree of regularity and structure. The tool used for the
ELaN project was based on four types of elements (see Fig. 1, Schön
et al., 2007:18–20):
• natural elements (e.g. landscape, livestock, soil, ground water),
• technical elements (e.g. energy, factories, infrastructure),
• systems of signs (e.g. rules, laws, concepts and principles)5 and,
• social actors (e.g. persons, groups, organisations).
Some of the elements were related. The following symbols are
provided in the CA handbook to characterise such relations Schön
et al. (2007):
• Single relation: Elements are related.
• Directional relation: One element has an impact on one other or
several other elements.
• Conflicting relation: One element is in conflict with one or several
other elements.
• Hampered relation: One element passively resists one or several
other elements.
• Feedback relation: Two elements are in a relation with mutual
self-reinforcement.
Constellations usually are built from the center putting the most
important elements on a ring. Elements which are grouped besides
each other in ‘clusters’ are considered to be closely related. If it
is helpful to understand a constellation, relational symbols can be
used to further characterise certain relations (e.g. conflicts or self-
enforcing relations). The symbols can also be used to point out
relations between elements of different clusters. The basic idea
of CA is to bring together different approaches, data sources (lit-
erature searches, interviews, discussion) and forms of knowledge
(scientific and practical) to create an image consisting of a constel-
lation that can be accepted or even shared by all practitioners and
researchers involved (Bruns and Ohlhorst, 2011). In cases where
it is not possible to reach agreement on such a constellation, the
method can also be helpful for identifying conflicting perspectives
at an early stage of a project. Here, dissent should be documented
and the project team should agree on the existence of conflicting
perspectives (Schön et al., 2007: 23).
There are several possible ways in which CA can be developed,
depending on the size of the group and available time resources: (a)
the whole team passes through all the steps of CA as a group, (b) part
of the team prepares maps to discuss them with the whole team,
(c) single members of the team prepare maps and discuss them
with other team members (Schön et al., 2007: 21). CA can also be
used for various purposes: (a) structuring multiple perspectives,
5
Some authors who have been applying CA by using the term ‘conceptual and
communicative elements’ for these types of elements (Kruse 2008:49).
529
(b) integration of results, (c) analysis of governance of processes,
(d) developing strategies (Schön et al., 2007: 15). Depending on
the main objective CA is applied for, there might be different logics
for selecting the most important elements of a certain constella-
tion and, subsequently, visualisation of relations between them.
In ELaN, CA was used to integrate distributed knowledge on chal-
lenges of water and land management for the two chosen sites
(fenlands and former sewage fields) and, therefore, also served
the purpose of explicating different perspectives and enhancing
mutual understanding between them. The process of carrying out
CA during the ELaN project is described in the following section.
3.2. Applying Constellation Analysis in ELaN
The joint problem framing process undertaken for the ELaN
project, achieved via Constellation Analysis, complemented and
more precisely defined assumptions initially formulated in the
project proposal. But here, all participating researchers were
systematically involved in generating the updated problem formu-
lation, not only those who took part in writing the project proposal.
This seemed especially important, since the project relied on coop-
eration between scientists from various disciplines – including
hydrologists, regional planners, biologists and economists – who
had thus far rarely worked together. Thus, the process of updat-
ing problem framing also served the purpose of social integration
within the group of approximately 45 scientists during the first
year.
With regard to relevant practitioners, the project drew on exper-
tise from very different fields, such as ground water protection,
waste water treatment, water management, agriculture and pro-
duction of bioenergy, whose actors had previously had little or no
contact with each other in their professional routines. In contrast
to other projects, where the antagonists and their positions regard-
ing certain topics are well known, this was less the case in ELaN,
since the aspect of using treated waste water as element of sus-
tainable land use had rarely been dealt with beforehand. Each of
the participating practitioners had previously been able to describe
in detail part of the problem – eutrophication of surface waters,
degeneration of fenlands, or rising costs of waste water treatment
– but would not necessarily have linked this part to other relevant
aspects. To foster especially exchange between actors from water
and land management was, therefore, an important objective of the
project as a whole, starting with joint problem framing.
Joint problem framing was done separately for the two sites
ELaN was dealing with: former sewage-irrigated fields and (degen-
erated) fenlands. In this article we only refer to one of these, as the
iterative methodological steps were the same for both analyses. To
get an idea about the content changes occurring during the process
of joint problem framing, this article describes the carrying out of
CA for water and land management of fenlands. Within the scope
of this article, however, it is not possible to elaborate a complete
analysis of the constellation (see Kröger et al., 2012). Insights and
conclusions regarding the challenges and the added value of this
process are identical for both analyses, since the methodological
design was the same.
Joint problem framing was accomplished on the basis of a vari-
ety of different methods, employed via a number of recursive steps,
in the period between April 2011 and May 2012. As a first step, a
review of the relevant literature and interviews with 17 experts
from research and practice were carried out. The group of prac-
titioners comprised three representatives of administrative units
(water and land authorities), three representatives of water or agri-
cultural associations, two representatives of environmental NGOs,
one local politician, one entrepreneur in the field of bioenergy
production and one landscape planner. The aim of the interviews
was to deepen knowledge about the related topics and to identify
530 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
• l
natural elements
technical elements
single relation directional relation conflictual relation
Fig. 1. Types of elements and relations of Constellation Analysis (source: Schäfer et al., 2010: 120).
Table 1
Inter- and transdisciplinary exchange for joint problem framing in ELaN.
Step Period
1 04–10/2011
2 04–09/2011
3 10–11/2011
4 12/2011
5 12/2011
6 01–02/2012
7 03/2012
9 03/2012
10 05/2012
and correlate the relevant elements of the specific ELaN constel-
lations. The interviews lasted between 45 and 90 min. They were
transcribed and analysed according to the four types of elements
mentioned above and their interrelations. Based on the literature
review and interviews, a first draft of the graphic representation of
the constellation was created by the group of scientists who were
responsible for the subproject of knowledge integration. Table 1
gives an overview of the subsequent steps of developing CA for
joint problem framing.
The first drafts of the graphic representations of the constella-
tions for each site were discussed within the ELaN coordination
team and with a number of ELaN researchers from different disci-
plinary backgrounds. After this internal feedback, the figures were
revised and written analyses about characteristics of the constel-
lations and challenges with regard to sustainable water and land
management at the two sites were formulated by the responsi-
ble subproject. The two constellation analyses then were presented
l
Systems of social actors
signs
hampered relation feedback relation
Procedure
Review of the relevant literature about regional
water and land use management and with regard
to the different sites (fenlands and former sewage
irrigation fields)
Conduction and analysis of 17 structured
interviews: Interviews with six members of the
advisory board and five additional external experts
as well as with the practitioner who was part of
the project team. Additionally, five interviews with
researchers of the project team were carried out.
Preparation of a first draft of two figures
visualising the constellation on each site by the
responsible sup project
Discussion of the graphic representation of the
constellations for the two sites in the ELaN
coordination team
Feedback interviews with eight ELaN researchers
with different disciplinary backgrounds regarding
the graphic representation of the two
constellations
Revision of figures and draft of written CA
Presentation and discussion of revised CA at the
annual meeting of the consortium with 47
researchers (discussion with working groups for
each site); documentation of discussion,
assessment of comments and revision of joint
problem framing (figures and text)
Forwarding of revised CAs (figures and text) to
consortium, requesting written comments
Analysis and consideration of written comments
Finalisation of interdisciplinary agreement on
problem framing
Presentation and discussion of problem framing
for the two sites at a stakeholder workshop with
43 participants (discussion in two working groups)
Discussion of the problem framing in a meeting
with the advisory board
Revision of joint problem framing according to
comments from the discussions
Finalisation of problem framing on the basis of CA
and discussed at the annual meeting of the consortium, consisting
altogether of 47 persons. After presentation to the whole consor-
tium, they were discussed in two sub groups of participants, sorted
according to their being involved with one the two sites. These
discussions were recorded. Suggested changes regarding the most
important elements and their relations were visualised directly in
the meeting by adding additional cards and symbols. The comments
were analysed thoroughly after the workshop by members of the
responsible subproject. After revision, the whole consortium was
asked to comment on the revised problem framing for the two sites
in case there were further suggestions. The figures and the written
analyses were distributed via mail with a deadline of two weeks
for commentary. Ten researchers of the consortium sent back com-
ments which were then analysed and taken into consideration
during the final revision.
In the next step, CA for joint problem framing was presented and
discussed at a transdisciplinary stakeholder workshop, the aim of
 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
which was to discuss the findings with a greater number of regional
actors who had up to then only been loosely or not yet associated
with ELaN. 43 practitioners participated at the workshop, including
representatives from local, regional or national government admin-
istrations, employees of the Berlin forestry and natural preserves
services, representatives of water and agricultural associations and
sewerage boards as well as members of environmental NGOs. Farm-
ers, consulting engineers and landscape planners were also present
but to a lesser extent.
The workshop was structured as follows: After presenting the
constellation analyses for the two sites in a plenary session, the
participants were designated to two working groups according to
their expertise in terms of the two land use types the project dealt
with (fenlands and former sewage fields). The discussion in each
working group (duration one hour) was stimulated by two 10-min
inputs from practitioners who had been selected in the course of
preparing the workshop.6 These discussions were moderated and
documented by the project team. The participants were asked to
comment on problem framing on the basis of CA (the figure was
distributed to all participants at the workshop) and, more generally,
to discuss the following questions:
- What could be potential benefits for the region from using treated
waste water?
- What questions need to be answered before considering the use
of treated waste water?
The workshop ended with a plenary session discussing the
potentials and risks of using treated waste water on the two sites
and practitioners’ expectations regarding the project.
Eight of eleven members of the advisory board took part in the
workshop. After the stakeholder workshop, there was a supple-
mentary 2.5-h meeting with the advisory board to discuss some
of the issues in more detail. In preparation for this meeting, the
members were provided with the figures and the accompanying
text of the CA, giving them a chance to comment on them. The
comments of the stakeholder workshop and the advisory board
meeting were documented and taken into account for a last revision
of the CA, which was conducted after the workshop by the respon-
sible team of scientists. Two months later, the formulation of joint
problem framing on the basis of CA was finalised and subsequently
published (Kröger et al., 2012).
3.3. Methods for reflecting upon inter- and transdisciplinary
exchange
The subproject which was responsible for facilitating inter-
and transdisciplinary exchange reflected upon the quality of the
exchange processes on the basis of participatory observation, brief
evaluative questionnaires as well as qualitative interviews with sci-
entists and practitioners in which aspects as the perceived benefit
and problems of these integrative processes were addressed.
The meetings of the project consortium as well as the stake-
holder workshop were analysed using participatory observation
regarding differences in the use of central terms as well as
understanding of central concepts such as risk perception and
assessment. It was also observed whether stereotypes were used
either to characterise certain disciplines or the differences between
researchers and practitioners. If members of the group reflected
6
CA of former sewage irrigated fields was commented by a representative of the
water administration on the level of the Federal State Berlin and the director of a
nature park which is linked to a former sewage irrigated field. CA of degenerated
fenlands in rural areas of Brandenburg was commented by the director of a water
and soil association and the chairman of the regional agricultural association.
531
the position of other discussion partners this was viewed as an
indicator for ‘mutual understanding’. Mutual learning was opera-
tionalised as a self-reported change of one’s own mind set, resulting
in, for example, adjustment of research questions or adapted use of
certain terms.
After the consortium meeting, a questionnaire was distributed
to all participants, of which 38 out of 47 responded (81%).
The questionnaire contained several questions addressing the
appropriateness of the method and whether joint problem fram-
ing succeeded in motivating processes of mutual exchange and
understanding (see Section 4.3). Participants were also asked for
additional remarks with two open questions:
• What was the main benefit of the consortium meeting for your
subproject?
• Do you have additional remarks or comments, even regarding the
organisation of the event?
In the days after the consortium meeting, the responsible sub-
project carried out six telephone interviews to obtain further
feedback on the process of joint problem framing using the CA
method. Researchers from every sub area were represented in the
sample. The interviews lasted 15–30 min and were recorded and
transcribed afterwards.
After the stakeholder workshop, a brief questionnaire was also
distributed, which contained several questions regarding the rel-
evance of CA for practitioners and the quality of the exchange
between researchers and practical actors (see Section 4.3). Since
several participants had left during the last 30 min of the event,
feedback was only obtained from 17 of 43 participants (40%).
4. CA as the basis for joint problem framing for
management of fenlands
In this section, the results of carrying out CA for challenges
regarding water and land management on fenlands will be very
briefly introduced Section (4.1) before the contributions of the prac-
titioners to the problem description are presented in more detail
in Section 4.2.
4.1. Problem framing for management of fenlands
We developed the CA for this topic based on the following
central question: What problems and risks does current water man-
agement in rural areas of Brandenburg face, particularly regarding
fenlands?
Fig. 2 shows that the status quo constellation consisted of three
clusters: (1) land use and nature protection, (2) surface water and
(3) waste water.
Cluster 1, “land use and nature protection”, highlights the impli-
cations of the current agricultural use of fenlands. Due to conditions
shaped by agricultural markets, energy policy and the Common
Agricultural Policy (CAP), farmers continue to intensify agrarian
production, especially by improving conditions for cultivating their
land via drainage. This dynamic is symbolized by a self-reinforcing
feedback loop. Protagonists of nature protection blame this practice
for degeneration of wetlands and consequent loss of biodiversity.
They therefore promote demolition of drainage systems or adapted
water management which can support the preservation or regener-
ation of fenlands. Since fenlands have an important function for the
storage of carbon dioxide, this debate has intensified with regard
to measures for climate protection.
Cluster 2, “surface water”, illustrates the current practice of dis-
posing of treated waste water. Usually this water is discharged into
surface waters, mostly rivers. Since this treated waste water still
532 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
Fig. 2. Constellation Analysis for water and land management of rural fenlands.
contains nutrients and pollutants, the surface water as well as the
North and the Baltic seas have become polluted over time, result-
ing in problems of eutrophication. To avoid further contamination,
the EU Water Framework Directive demands that all EU countries
improve surface water quality towards attaining a higher standard.
It is, consequently, expected that standards for treated waste water
quality will also rise during the coming years.
Cluster 3, “waste water”, consists of the relevant elements
related to current treatment of waste water in Northeastern
Germany. Due to migration into cities and other demographic
changes, associations for sewage treatment in rural areas are
faced with increasing costs for an oversized waste water treat-
ment infrastructure. They are also irregularly confronted with huge
amounts of waste water caused by extreme weather events, which
cannot be dealt with adequately by current waste water treatment
plants.
After the mapping process, the constellation was analysed,
revealing competing interests or problems that have to be dealt
with in the future. The quality of treated waste water has to improve
to be able to fulfill the standards of the EU Water Framework
Directive. However, in rural areas waste water treatment plants
are already confronted with the problem of maintaining existing
infrastructure and lack the financial leeway for new investments.
Regarding water management, there is a clear conflict between
the interests of agriculture and those of environmental protec-
tion. A high percentage of fenland is already severely degenerated
due to water management practices which allow intensive agricul-
tural production. During the last decade, there has been growing
recognition that the existence of fenlands is important for climate
protection and for maintaining sound regional water balance as
well as biodiversity. On the other hand, pressure on the agricul-
tural sector to use land as efficiently as possible has risen due to
economic and political trends. In the long run, the melioration of
soils and intensive forms of production will result in gradual dete-
rioration of soils, which is also economically unattractive. The main
challenge is to provide economically viable options for farmers to
cultivate land with higher ground water levels. The establishment
of regional value-added chains for crops produced on fenlands as
well as public compensation schemes for less intensive production
will be key elements for finding acceptable solutions for all actors
involved.
The special focus of the ELaN project regarding this constellation
was to analyse whether the use of treated waste water for irrigating
fenlands could be a feasible component of a possible solution for
the sketched problems of water and land use management. To date,
the use of treated waste water has been forbidden, due to precau-
tionary ground water protection, and can only be tested within the
context of a research project. By using treated waste water for irri-
gating land, however, it could be possible to solve the problem of
polluting surface water. It is assumed that remaining nutrients and
pollutants will be removed by microorganisms and plants when
the water passes through fenland soil. However, the effects of long-
term irrigation of sensitive soils with treated waste water have not
yet been analysed sufficiently.
On the micro level, CA made clear that heterogeneous elements
are interacting and forming the constellation of water and land
management on fenlands. The current practices for dealing with
treated waste water are strongly influenced by a regulation which
 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
– following a precautionary logic – puts limitations on gaining
experience with innovative forms of water management. Research
projects which contribute towards the assessment of risks and ben-
efits of different water and land management options (symbolised
by the element ‘knowledge’) play an important role as ‘experi-
mental sites’. Besides regulation, material elements like size and
design of waste water treatment plants as well as the current shape
of the existing melioration infrastructure also play a very impor-
tant role for the constellation. Infrastructure is designed for long
time periods (path dependency) and can only be modified within
certain windows of opportunity when facilities have to be refur-
bished or renewed. Meanwhile, forms of land use can be altered
within shorter time cycles, as they depend to a large extent on the
global agricultural market and corresponding national regulations.
The existence of self-enhancing mechanisms stresses the stabil-
ity of parts of the current constellation. This analysis confirms the
complex socio-ecological interdependencies which are typical for
sustainability problems, as it is not possible to identify one major
actor, one central regulation or one single technology which needs
to be addressed in order to substantially change the constellation
(Fig. 2).
On the macro level two competing logics of action were iden-
tified which influence the constellation. While administrators and
actors from nature protection are guided by a logic of “guarantee-
ing public goods and services”, farmers are acting according to an
economic logic with the components of efficiency and profit ori-
entation. Organisations such as drinking water and waste water
treatment companies also primarily follow an economic logic.
However, since it is their responsibility to assure high water quality,
they need to act within a narrow framework defined by the logic
of caring for a public good. For development of strategies towards
sustainable water and land use options, it is important to be able
to bridge these different logics of action (see in more detail Kröger
et al., 2012).
4.2. Integration of practitioner knowledge in problem framing
In the following section we characterize the input and com-
ments of the practitioners involved in order to build an impression
regarding their contributions towards structuring, enriching and
differentiating the problem framing.
4.2.1. Structure of the constellation
One of the central aspects of mapping a constellation is the deci-
sion regarding the most important elements which are put in the
center of the constellation as a ‘core’. The original ELaN project
proposal stressed the potential of using treated waste water for
innovative water management on both sites. In the expert inter-
views with representatives of the water and soil association and
actors from natural protection, however, it became clear that the
use of treated waste water is not a very important issue in remote
rural areas of Brandenburg but, rather, that the project should take
up the question of water management as a whole. To be able to
understand this point it is important to know that water man-
agement and melioration was organised by the state under the
Germand Democratic Republic (GDR). Since German unification in
1990, responsibility for maintaining the melioration infrastructure
as well as for water management has never been clearly assigned
between the local communities, the water and soil associations and
landowners. The issue is difficult to handle, since substantial costs
are related to these tasks. That is why there is discontent with the
current water management system from both sides: farmers as well
as proponents of nature protection. The possible use of treated
waste water to confront water scarcity in remote rural areas is
only a minor point when considering the whole range of problems
related to water management. One major problem which reduces
533
the potential of using treated waste water is that many degener-
ated fenland areas are not located near sewage treatment plants
and that the transport of waste water via pipes is very expensive.
These comments were taken up in structuring the two different
constellations: while “treated waste water” was put in the centre of
the CA of former sewage farms, “water management” was chosen
as the most important element at the centre of the constellation
analysis for fenland areas (Kröger et al., 2012). Through the itera-
tive process problem framing, the element “treated waste water”
lost importance in the fenland constellation, as shown in Fig. 3.
This point was discussed intensely at the annual meeting of the
project team. Especially the practical project partner – the waste
water treatment company – was at first arguing in favor of keeping
‘treated waste water’ at the center of the constellation, since the
possible use of treated waste water reflected its main interest in
the project. Exchange of perspectives on the matter resulted in an
agreement regarding the general differences of the constellations
for the two sites. This change of priorities resulted in a shift in the
scientific subprojects which were dealing with new land manage-
ment options on fenlands. They then focused on land use options
for areas with higher ground water levels, independent of whether
these higher ground water levels were based on different water
management approaches or due to irrigation with treated waste
water.
The interviews with practitioners also made clear that water
and soil associations are important players in the field of interest,
as they operate at the interface of water and land management.
This element was, thus, added to the core of the constellation.
4.2.2. Enrichment of the problem description
The practitioners contributed many comments which led to a
more complete problem description. These were mostly integrated
into the written version of the CA. For example, some practitioners
commented that, under the current regulations, there are diverse
contradictions in dealing with risks to ground water quality. For
instance, although the use of treated waste water is controlled by
very strict regulations, the same standards are not applied for the
use of sewage sludge or liquid manure on agricultural fields.
One example of how the comments from practitioners led to
an enrichment of the constellation is shown in Fig. 4. Practitioners,
who were familiar with the agricultural situation in East Germany,
stressed that individual farmers often have limited leeway to decide
between land use options if they have leased the land being used.
‘Land owners’ were therefore integrated as a new element, symbol-
ising their relevance for changes in land management. It was also
pointed out that, besides energy politics, the agricultural market
as a whole and the Common Agriculture Policy (CAP) have major
influences on the economic viability of land management options.
After discussing their importance, these elements were thus also
integrated into the constellation.
4.2.3. Differentiation of perspectives regarding the constellation
The discussions with practitioners also resulted in the identifica-
tion of different perspectives on central issues of the project, which
provided important clues regarding starting points for strategies
towards sustainable water and land management and were inte-
grated into the written analyses of CA (Kröger et al., 2012). For this
paper, we have chosen to sketch out three of these important issues.
• Differentiation of risk perception: One of the project goals, as stated
in the project proposal, was to gain more knowledge on the risks
of using treated waste water for land use. Especially the waste
water treatment company and the responsible administrators
were interested in a thorough monitoring of nutrients and syn-
thetic substances which could endanger the quality of ground
water. During the process of attaining a shared understanding of
534 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539

surface
waters climate
fens
change
treated waste water soil
water availability drainage
intensive
agriculture

Initial draft of the constellation core for fenlands based on the proposal

surface
waters
Water and Soil
Associations

treated water soil

waste water management drainage

fens
precipitation

Constellation core for fenlands after the stakeholder workshop

Fig. 3. Changing the constellation core of the fenland CA in the process of problem framing.

CAP

agricultural
energy markets
policy
energy land owners
degradation
farmers degradation policy
farmers
intensive
agriculture agriculture
soil
drainage soil
drainage

First draft of part of the constellation based on Final draft of this part of the constellation
interviews (November 2011) (May 2012)

Fig. 4. Enrichment of CA for fenlands via practitioner comments.

the problem, it became obvious that different understandings of
risk assessment and dealing with risks were present within the
group of scientific researchers as well as in the group of involved
practitioners. Some of the participants stressed the precaution-
ary principle of protecting ground water from human impacts,
which is the basis of current legal regulations. According to this
principle, the use of waste water can only be permitted if risks for
ground water quality can be completely excluded. Others pointed
out that, considering the number of new synthetic substances
which are constantly being released on the market, it would not
be possible to do risk assessments on the basis of single-substance
monitoring. If it is additionally taken into account that surface
water and ground water are interconnected systems, the cur-
rent practice of disposing treated waste water into rivers also
presents a risk for ground water quality in the long run. Represen-
tatives of the second group stressed the necessity for integrated
risk assessment, comparing benefits and risks for different water
management options, and dealing with the complete system of
surface and ground water instead of focusing on only one of them.
• Differentiation of positions on land use: As described above, there is
a general conflict between farmers and actors promoting nature
protection regarding the question of to what extent the ground
water level of fenland areas has to be adapted to the needs of
agricultural use. The antagonistic positions “human land use has
to be adapted to natural conditions” versus “natural conditions
have to be adapted to the requirements of human land use” can
be identified. However, in the interviews and during the stake-
holder workshop it became clear that there are protagonists of
both actor groups who are open to more nuanced views. On the
one hand, there are actors in the field of nature protection who
recognise that protection of biotopes and natural reserves can
only be ensured in cooperation with those actors who cultivate
the land. The relatively new slogan “Protection by Cultivation” of
some environmental organisations indicates that there is a shift
away from the previous paradigm that nature protection is best
accomplished by no human use at all. On the other hand, there are
farmers who recognise that soil quality in the area has been dete-
riorating severely with the currently implemented melioration
regime, which already limits forms of intensive cultivation and
results – or will result – in economic losses. The interviews and
discussions made these nuanced views visible and showed that
 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
there is a leeway for compromise between certain protagonists
of the two actor groups.
• Differentiation of positions on water management: At the begin-
ning, it was assumed by the project team that NGOs in the field of
nature or environmental protection would be in favor of demoli-
tion of the melioration infrastructure while the farmers voted for
maintaining – and partly reconstructing – those facilities. In the
course of the interviews and the workshop, it became clearer that
some of the environmentalists also did not think that a complete
demolition of the water management infrastructure would serve
the interests of nature protection best. Even if they wanted to alter
priorities in water management – taking the arguments of nature
and environmental protection more into account – they acknowl-
edged that it would not be possible to go “back to nature” by
resigning completely from human water and land management.
An active form of water management depending on functioning
infrastructure would, therefore, appear to be a common point of
interest for a broad range of actors.
These examples are intended here to show how practitioner
knowledge was integrated into joint problem framing facilitated
by use of CA in different ways.
4.3. Evaluation of the inter- and transdisciplinary exchange
process
Evaluation of the consortium meeting with a brief questionnaire
indicates that the event served the purpose of motivating exchange
quite well, resulting in a better understanding for each other’s per-
spectives, as well as regarding the overall context of the project (see
Table 2). The idea of embedding one’s own research questions into
a broader context was also stressed in the interviews, as the follow-
ing statements make clear: “It was helpful to see all the aspects which
one doesn’t always have in mind together in one picture.” “For me, as a
hydrogologist who works within a quite narrow range, it was interest-
ing to see all the other perspectives. Up to now I hadn’t approached the
project with an overall picture but rather only with my own detailed
research question.” Discussion in the group was seen as being valu-
able, suggesting processes of social integration: “It was good to look
at it as a group and discuss it.” “For me it was important to get to know
the research team in this way.”‘I remember that after the meeting I
went home with a very positive feeling.’
Approximately one third of the participants stated that they
ended up re-thinking the design of their own subproject follow-
ing the discussion during the consortium meeting, indicating here
a learning process. This is also illustrated by the following remark:
‘It had a certain influence on my research question. Not that I changed
it a whole lot, but there are aspects that I have in my mind now as a kind
of background information.’ As mentioned above, this is especially
true for the subprojects dealing with fenlands. After recognizing the
low relevance of irrigating fenlands with treated waste water, they
enlarged their focus towards examining the more general question
of water management on these sites (Fig. 4).
Feedback on the method was, however, ambivalent. While a
majority of the participants agreed that visualisation of the con-
stellation was very helpful for understanding the overall context
of ELaN, approximately one third found the presentation confusing
and had difficulties understanding the method (see Table 2).
The results of the evaluation of the stakeholder workshop show
that most of the participants were very interested in the project
and felt that the workshop motivated exchange and provided valu-
able insights into the topic (see Table 3). Most participants felt
that exchange between researchers and practitioners had worked
out well and were optimistic that their comments would be taken
into consideration by the research team. However, a majority of
the participants also agreed with the statement that there was
535
already enough research on this topic while practical actions were
lacking. The graphic representation of the topic was evaluated as
being helpful for understanding problem framing. From the feed-
back regarding the workshop it can be concluded that it served well
the purpose of introducing the project to a broader range of practi-
tioners and succeeded in creating an atmosphere of mutual respect
and openness for further exchange.
The event was also important for researchers within the project,
as articulated by the following comment: “Interesting was the reac-
tion of the different people who are dealing with this topic. How they
react to the question of using treated waste water for land use and how
they position themselves.”
5. Reflecting on the process of knowledge integration
during problem framing using Constellation Analysis
In this section, the process of inter- and transdisciplinary knowl-
edge integration in joint problem framing using Constellation
Analysis will be discussed. Section 5.1 reflects upon the use of Con-
stellation Analysis for the process of generating system knowledge
in land use research, while Section 5.2 discusses the quality of the
transdisciplinary design in this phase.
5.1. Suitability of Constellation Analysis for problem framing and
mutual understanding
As described above, the activities in the first phase of the project
had the objective of bringing together knowledge and perspectives
from different disciplines as well as from practitioners from differ-
ent fields. This seemed especially necessary since the two strands
– water and land management – had thus far not often been dealt
with in an integrated way. Besides the generation of system knowl-
edge as a basis for disciplinary work within the ELaN research team,
the process of coming to a shared understanding of the problem also
had the objective of contributing towards social integration of the
different partners.
The results suggest that these objectives were broadly met.
The different options for bringing in disciplinary perspectives (via
interview, during the consortium meeting and via commenting on
drafts) succeeded in broad participation of the research team and
initiated processes of mutual understanding. Combination of fig-
ures with accompanying written analysis seemed to be appropriate
in this setting. While visualisation of the most important elements
and their relations encouraged the participants to agree on the
project’s focus, the written analysis allowed for inclusion of greater
detail and the documenting of conflicting positions. These results
confirm the importance of visualisation as a key element accom-
panying discursive methods (Kruse, 2008). Refering specifically
to CA as a method, Ohlhorst (2009: 64) concludes that visualisa-
tion and language support each other in the process of mutual
understanding, since they compensate each other’s deficits and dif-
ficulties. However, as the feedback from some scientists indicates,
the method is not self-explanatory and could have been introduced
better by the responsible subproject.
Compared to the problem framing of the initial proposal, the
discursive process led to a more refined perception of impor-
tant elements and their interrelations. Especially the recognition
of more nuanced positions regarding risk assessment, the tension
between nature protection and intensive agriculture and the design
of future water management strategies was an important starting
point for the development of strategies able to bridge different
interests and logics of action. Further on in the project, the con-
sideration of these nuanced positions partially led to adjustments
regarding the character of the striven-for end products. To carry out
these discussions during an early stage of the project also helped
536 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539

Table 2
Results of the evaluation of the consortium meeting.

Item Construct Agree-ment* undeci-ded Disagree-ment**

The form of presenting the constellation is Appropriateness of the method 11 11 13

confusing for me.
I had difficulties in understanding the method Appropriateness of the method 10 12 10
of constellation analysis.
Visualisation with constellation analysis is Appropriateness of the method 25 11 2
helpful for my understanding of the overall
context of ELaN.
The discussions we had today on the basis of Quality of the exchange process 9 12 15
CA were too superficial from my point of
view.
Discussions have helped me in Mutual understanding 26 10 2
understand-ding the perspectives of the
other sub areas and subprojects better.
The consortium meeting has motivated me to Mutual understandingSocial integration 33 5 0
exchange ideas with colleagues.
Discussions today have motivated me to Mutual learning 12 12 12
re-think the design of my subproject at some
points.

N = 38, representing 81% of the participants.

*
The numbers summarise votes of complete and partial agreement.
**
The numbers summarise votes of complete and partial disagreement.

Table 3
Results of the evaluation of the stakeholder workshop.

Item Construct Agree-ment* undeci-ded Disagree-ment**

I am very interested in the project and its Practical relevance of the topic 16 0 0
results.
The topic is very relevant for my professional Practical relevance of the topic 8 4 5
life.
The scientific assumptions of the project are Practical relevance of the topic 5 6 5
not really linked to practical problems.
These topics have already been analysed and Practical relevance of the topic 12 2 1
discussed for decades – what is missing is
putting them into practice.
Graphic presentation of the topic was helpful Appropriateness of the method 10 4 3
for my understanding of the problem.
The event has provided valuable insights into Design of the workshop 13 4 0
the project and the topic.
The workshop has motivated exchange with Design of the workshop 15 2 0
colleagues and other participants.
Discussion today remained too superficial. Quality of the exchange 5 2 9
I have the impression that my comments were Quality of the exchange 2 4 9
not taken seriously and will not be
considered by the project team.
The important conflicts in this field were not Quality of the exchange 3 8 5
sufficiently addressed.
Exchange between scientific and practical Quality of the exchange 12 2 3
actors worked out well during the workshop.

N = 17, representing 40% of the participants.

*
The numbers summarise votes of complete and partial agreement.
**
The numbers summarise votes of complete and partial disagreement.

to identify actors in the different fields of water and land manage-
ment who were open to innovation and reflection upon established
practices. Identification of these actors made it possible to address
them as possible pioneers further on in the project. These experi-
ences therefore confirmed the importance of joint problem framing
as a first step in complex sustainability problems dealing with high
levels of uncertainty and diverging interests (Jahn et al., 2012; Pohl
and Hirsch Hadorn 2008b; Tress et al., 2007; Scholz et al., 2009).
This is especially true for issues in the field of land use management,
which are often characterised by competing interests, such as eco-
nomic considerations versus acknowledgement of environmental
concerns.
At some points, the process of joint problem framing led to
insights which indicated the need to broaden the focus of the
research project (e.g. concerning the question of water manage-
ment for fenlands). Since CA was carried out in the first phase
of a five-year project, this insight could still be taken up by the
responsible sub-projects. These adjustments suggest that learning
processes based on the exchange between scientists and practition-
ers did take place. The process therefore seems to have exhibited
characteristics of knowledge integration as “a process that leads to
a change in the structure and organization of a problem context by
extending and constraining both” (Jahn et al., 2012: 7).
The process of problem framing based on applying CA there-
fore appears to have initiated processes of mutual understanding
and supported the project in confronting knowledge fragmen-
tation, which is a general characteristic of modern societies.
Fragmentation of knowledge can not only be observed in research
(fragmentation in disciplinary knowledge) but also in governmen-
tal administration and politics (Kruse, 2008). In ELaN, for example,
it became very clear that, for example, dealing with issues related
to surface water, ground water and waste water in different admin-
istrative departments hampers perspectives for integrative risk
assessment of these interconnected systems. Systemic solutions
 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
will only be possible if regular exchange and joint problem solving
between such actors is institutionalised. ELaN offered an experi-
mental forum for such exchange.
5.2. Reflecting on the quality of the transdisciplinary exchange
Overall, a small team of researchers was responsible for inte-
grating the heterogeneous disciplinary and practical knowledge by
producing a draft of the Constellation Analysis first and then sub-
sequently completing and correcting this draft. Diverging from the
procedure chosen for ELaN, there is also the possibility of conduct-
ing a CA in close collaboration between scientific researchers and
practitioners from the very beginning (Schön et al., 2007). The iden-
tification of the main elements and their relations is then subject to
discussions and direct exchange, until the participants agree on a
joint picture. In this case, the responsible researchers mainly have a
moderating function. To generate CA in this way is only possible in
groups of 8–10 persons, however, and can be a very time consum-
ing process (Schön et al., 2007). On the other hand, it can provide
participants with in-depth insights into contrasting opinions. For
ELaN it was necessary to integrate a larger number of participants.
Since the research team already comprised over 45 persons, the
option of developing a CA in direct exchange did not seem feasible.
It also seemed to be reasonable to reach an interdisciplinary agree-
ment on the constellation first before opening up the discussion to
a broad range of practitioners. Although this procedure seems to be
adequate regarding the project context (size of the project team and
available time resources), there is no doubt that a relatively small
research team made the final decisions about which information
was given consideration and which supplements and corrections
were actually made. Going through several recursive loops, how-
ever, pretty much assured that a broad range of perspectives was
adequately included. The principle of recursiveness is stressed by
various authors dealing with transdisciplinary research (e.g. Pohl
and Hirsch Hadorn, 2008a: 430).
Practitioners participated in the process of problem framing at
different levels of intensity. Some members of the advisory board
and several other relevant practitioners were involved at an early
stage via interviews, integrating their expertise already in first
drafts of the graphic representations of the constellations. Mem-
bers of the advisory board also had the option to comment on the
written analysis of the CA and to provide detailed feedback in the
context of an advisory board meeting.
The workshop with a broader group of practitioners who had
not aligned with the project before had the main purpose of intro-
ducing the project to regional actors and getting feedback regarding
demands and reservations at an early stage. The method of present-
ing and discussing the CA for the two sites in the space of a four-hour
workshop can be seen as a compromise between the objective of
offering participative options and the limited time resources of the
invited practitioners. The evaluation via questionnaire indicates
that the workshop succeeded in raising interest about the project
and that the majority of the participants felt taken seriously. While
the feedback was mainly positive, there were critical remarks espe-
cially regarding the superficiality of the discussion and transfer of
research results into practical action. The research team concluded
from these experiences that more in-depth discussions should be
carried out with certain target groups, especially those which had
been underrepresented at the workshop, such as farmers.
Reflection on the process leads us to the conclusion that the
transdisciplinary design of ELaN during this early phase mainly had
the character of being ‘consultative’ rather than being “participa-
tory”, a differentiation made by Mobjörk (2010: 870). Practitioners
in this phase mainly had the role of responding and reacting to
problem framing already developed by the research team, who
had included information previously gained through interviews
537
with practitioners. This type of integration is characterized as being
“consultative” (Mobjörk, 2010). Only practitioners of the advi-
sory board were integrated into a more active way in the process
of knowledge production. But their integration does not appear
to fulfill the far-reaching demands formulated for “participatory
transdisciplinarity” by Mobjörk (2010), conceived as an approach
that achieves an open character on all levels: actors involved, meth-
ods chosen and developed, and problem addressed.
Throughout the research process, the advisory board was able
to act as an adequate forum for mutual discussion and learning on
an equal footing. It was consulted as a group twice per year and
in between on a bilateral basis with regard to certain aspects of
the research process. These practitioners, who also had a close link
to the research topic due to their professional backgrounds, had
a stronger influence on the research process than those who were
integrated on an irregular basis. They especially had a decisive influ-
ence on the design of some of the integrative end products, such
as the administrative manual and governance recommendations.
The research team concluded that these actors did achieve a shift
from consultative towards participatory transdisciplinarity via the
research process. The mutual relations involved, which included
valorisation of different types of knowledge as being equivalent,
can also be designated as “power balanced relationships”, which are
recommended by Scholz (2011: 376) for attaining an ideal trans-
disciplinary process.
The project team concluded that it made sense to involve dif-
ferent practitioner groups at different levels of intensity, which
has also been recommended by other authors (Tress et al., 2006).
A mixture of consultative and participatory transdisciplinary is
required in order to take the restrictions of time resources in
complex research teams as well as availability of practitioners
into account. Exploring different participatory modes can con-
tribute towards closing the diagnosed gap between theory and
practice within transdisciplinary research, acknowledging that the-
oretical approaches are often excessively complex for real-world
applications (Zscheischler and Rogga, 2015; Brandt et al., 2013;
Blättel-Mink and Kastenholz, 2005).
6. Conclusions
The ELaN project took up the challenge of integrating frag-
mented knowledge in the fields of water and land management –
both of which had previously been mostly approached separately
in scientific research as well as in practice. Knowledge from differ-
ent scientific and practical strands was brought together to explore
this new topic and to draw a more thorough picture of the chal-
lenges facing sustainable water and land management in specific
sites. A shared understanding of the problem was jointly devel-
oped in order to update, modify and refine the assumptions of the
preparatory phase of the project. For this purpose, the method of
Constellation Analysis (CA) was used.
Based on the experiences had during the ELaN project, we draw
the following conclusions regarding methods and processes for
inter- and transdisciplinary knowledge integration:
• The importance of an initial phase of joint problem framing
appears to have been confirmed for projects of this type. The main
benefits of this step are the integration of fragmented knowledge,
resulting in system knowledge as a basis for the subsequent work
in the subprojects and the social integration of heterogeneous
actors. Additionally, getting a first impression about conflicting
perspectives as well as possible bridging concepts can be benefi-
cial for subsequent phases of the project. Although the process
was rather time consuming due to several iterative feedback
538 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
loops, efforts of this kind seem to be justified for large consor-
tiums and projects of long duration.
• A method which links visualisation with written analysis seems
to be helpful for this step of joint problem framing. Visualisation
serves the need of clarifying a projects’ focus and can facilitate
obtaining a quick overview of the aspects involved, while the
written analysis enables inclusion of a range of differing per-
spectives and document conflicting positions. We feel that CA
has been proven to be suitable for this purpose. In comparison to
other methods, it also has the advantage of dealing with technical
and natural elements, actors and signs as being equally impor-
tant, which can encourage researchers from different disciplines
and practitioners from different fields to relate more easily to the
method.
• Integrating various practitioners at differing levels of intensity
in such a process is recommendable, taking into account the rel-
evance of individual actors in relationship to a given problem
and their resource constraints. Participatory transdisciplinarity
probably can only be accomplished with a small group of actors
who have a strong interest in the issue at stake and are willing
to become involved in knowledge co-production. Consultative
transdisciplinarity, including a broader range of actors, can never-
theless be important for assuring basic acceptance and openness
towards sustainability projects in given contexts.
• Knowledge integration from the very beginning of a research
project is a demanding task. It has to be designed thoroughly, and
the additional efforts (in time and manpower) required have to
be taken into account. Without application of appropriate meth-
ods and a clear responsibility for carrying out and reflecting upon
them, this task can hardly be successfully accomplished.
Regarding the relevance of this method for land use research,
we conclude that discussions about problem framing in a broader
group of practitioners at an early project stage can be very helpful
for identifying actors who are open to reflection upon – and possibly
change of – established land management practices. Further on in
the project, these actors can play an important role as pioneers for
sustainability innovations in land use. Transdisciplinary research
projects can, in the process, play an important role in establishing
‘experimental sites’ which enable the testing and monitoring of
such innovative practices.
Actors in land management organisations are confronted with
uncertainties and inconsistencies in their daily practices regarding
issues such as regulation or dealing with risks. Transdisciplinary
projects can provide discursive platforms for exchange with other
actors in this field, even beyond the traditional boundaries of spe-
cialised expertise. These projects have the resources to organise
and moderate exchanges of this kind and document the results,
thus providing participants with a common point of reference for
their future work.
Summing up, this article has sought to contribute towards
making methodological advances for knowledge integration in
inter- and transdisciplinary research. In doing so, we would like
to encourage other authors to share their experiences with apply-
ing methods for problem framing or knowledge integration. As
Zscheischler and Rogga (2015) point out, a common understanding
of methods and success factors for knowledge integration are not
yet in sight, but transparent ways of proceeding and reflexive dis-
cussions about experiences had are preconditions for advancing in
this direction.
References
Bammer, G., 2013. Disciplining Interdisciplinarity: Integration and Implementation
Sciences for Researching Complex Real-world Problems. ANU E Press, The
Australian National University, Canberra, Australia.
Bergmann, M., Brohmann, B., Hoffmann, E., Loibl, C., Rehaag, R., Schramm, E., Voß,
J.-P., 2005. Quality Criteria of Transdisciplinary Research: A Guide for the
Formative Evaluation of Research Projects. ISOE-Studientexte, Nr. 13, Frankfurt
am Main.
Bergmann, M., Jahn, T., Knobloch, T., Krohn, W., Pohl, C., Schramm, E., 2012.
Methods for Transdisciplinary Research: A Primer for Practice. Campus,
Frankfurt/Main.
Blättel-Mink, B., Kastenholz, H., 2005. Transdisciplinarity in sustainability
research: diffusion conditions of an institutional innovation. Int. J. Sustainable
Dev. World Ecol. 12 (1), 1–12.
Brand, K.-W., 2000. Nachhaltigkeitsforschung. Besonderheiten, Probleme und
Erfordernisse eines neuen Forschungstyps. In: Brand, K.-W. (Ed.), Nachhaltige
Entwicklung und Transdisziplinarität. Besonderheiten, Probleme und
Erfordernisse der Nachhaltigkeitsforschung. Analytica, Berlin, pp. S9–28
(Angewandte Umweltforschung; 16).
Brandt, P., Ernst, A., Gralla, F., Luederitz, C., Lang, D.J., Newig, J., Reinert, F., Abson,
D.J., Wehrden, H., 2013. A review of transdisciplinary research in sustainability
science. Ecol. Econ. 92, 1–15.
Bruns, E., Ohlhorst, D., 2011. Wind Power Generation in Germany – a
transdisciplinary view on the innovation biography. J. Transdisciplinary
Environ. Stud. (TES) 10, 45–67 (no. 1).
Callon, M., Law, J., Rip, A. (Eds.), 1986. Mapping the Dynamics of Science and
Technology. MacMillan Press. Houndmills.
Defila, R., Di Giulio, A., 2015. Integrating knowledge: challenges raised by the
inventory of synthesis. Futures 65, 123–135.
Gibbons, M., Limoges, C., Nowotny, H., Schwartzman, S., Scott, P., Trow, M., 1994.
The New Production of Knowledge. In: The Dynamics of Science and Research
in Contemporary Societies. Sage, London.
Giri, A.K., 2002. The calling of a creative transdisciplinarity. Futures 34 (1),
103–115.
Hinkel, J., 2008. Transdiciplinary Knowledge Integration: Cases from Integrated
Assessment and Vulnerability Assessment. Ph. D. Thesis. Wageningen
University.
Hirsch Hadorn, G., Hoffmann-Riem, H., Biber-Klemm, S., Grossenbacher-Mansuy,
W., Joye, D., Pohl, C., Wiesmann, U., Zemp, E. (Eds.), 2008. Handbook of
Transdisciplinary Research. Springer, Heidelberg.
Hunecke, M., 2011. Wissensintegration in der transdisziplinären
Nachhaltigkeitsforschung. Eine Fallstudie zur Anpassung an zunehmende
Starkniederschläge in urbanen Räumen. GAIA 20/2, 104–111.
Jahn, T., Bergmann, M., Keil, F., 2012. Transdisciplinarity: Between mainstreaming
and marginalization. Ecol. Econ. 79, 1–10.
Kröger, M., Rückert-John, J., Schäfer, M., 2012. Wissensintegration im nachhaltigen
Landmanagement. Inter- und transdisziplinäre Problembeschreibung im
Projektverbund ELaN. ELaN Discussion Paper, Müncheberg (download: http://
www.elan-bb.de/media/pdf/Publikationen/EDP2 Kroeger 978-3-943679-05-2.
pdf).
Kruse, S., 2008. Structuring Multiple Perspectives in Environmental
Decision-Making: Flood Protection in the Middle Elbe River. In: Edmundson,
R., Rau, H. (Eds.), Environmental Argument and Cultural Differences: Locations,
Fractures and Deliberations. Peter Lang, Oxford, Great Britain, pp. 37–64.
Lang, D.J., Wiek, A., Bergmann, M., Stauffacher, M., Martens, P., Moll, P., Swilling,
M., Thomas, C.J., 2012. Transdisciplinary research in sustainability science:
practice, principles, challenges. Sustainability Sci. 7 (S1), 25–43.
Latour, B., 1987. Science in Action: How to Follow Scientists and Engineers through
Society. Harvard University Press, Cambrigde, MA.
Limoges, C., 1993. Expert knowledge and decision-making in controversy contexts.
Public Understanding Sci. 2 (4), 417–426.
Loibl, M.C., 2006. Integrating perspectives in the practice of transdisciplinary
research. In: Voß, J.-P., Bauknecht, D., Kemp, R. (Eds.), Reflexive Governance for
Sustainable Development. Edward Elgar, Cheltenham, Great Britain, pp.
294–309.
McDonald, D., Bammer, G., Dean, P., 2009. Dialogue tools for research integration
Canberra. ANU E Press. The Australian University, Australia.
Mobjörk, M., 2010. Consulting versus participatory transdisciplinarity: a refined
classification of transdisciplinary research. Futures 42, 866–873.
Niederberger, M., Wassermann, S. (Eds.), 2014. Methoden der Experten- und
Stakeholdereinbindung in der sozialwissenschaftlichen Forschung. VS Verlag
für Sozialwissenschaften.
Ohlhorst, D., 2009. Windenergie in Deutschland: Konstellationen, Dynamiken und
Regulierungspotenziale im Innovationsprozess. VS-Verlag, Wiesbaden,
Germany.
Pérez-Soba, M., Petit, S., Jones, L., Bertrand, N., Briquel, V., Omodei-Zorini, L.,
Contini, C., Helming, K., Farrington, J.H., Mossello, M.T., Wascher, D., Kienast, F.,
Groot, R. de, 2008. Land use functions – a multifunctionality approach to assess
the impact of land use changes on land use sustainability. In: Sustainability
Impact Assessment of Land Use Changes. Springer, Heidelberg, pp. 375–404.
Pohl, C., Hirsch Hadorn, G., et al., 2008a. Core terms in transdisciplinary research.
In: Hadorn, G. (Ed.), Handbook for Transdisciplinary Research. Springer,
Heidelberg, pp. 427–432.
Pohl, C., Hirsch Hadorn, G., 2008b. Methodological challenges of transdisciplinary
research. Nat. Sci. Soc. 16, 111–121 (download: http://www.nss-journal.org/
articles/nss/pdf/2008/02/nss8204.pdf).
ProClim Forum for Climate and Global Change, 1997. Research on Sustainability
and Global Change – Visions in Science Policy by Swiss Researchers
(download: http://proclimweb.scnat.ch/portal/ressources/1122.pdf).
 M. Schäfer, M. Kröger / Land Use Policy 57 (2016) 526–539
Rammert, W., 2003. Technik in Aktion: Verteiltes Handeln in soziotechnischen
Konstellationen. In: Christaller, T., Wehner, J. (Eds.), Autonome Maschinen.
Campus, Frankfurt am, pp. 289–315.
Rossini, M., 2009. Was ist das Problem? Problemstrukturierung in der inter- und
transdisziplinären Forschung. Technikfolgenabschätzung – Theor. Prax. 18 (1),
117–119.
Scavarda, A.J., Bouzdine-Chameeva, T., Meyer Goldstein, S., Hays, J.M., Hill, A.V.,
2006. A methodology for constructing collective causal maps. Decis. Sci. 37,
263–283 (number 2).
Schäfer, M., Ohlhorst, D., Schön, S., Kruse, S., 2010. Science for the future:
challenges and methods for transdisciplinary sustainability research. Afr. J. Sci.
Technol. Innovation Dev. (AJSTID) 2, 114–137 (No. 1).
Schön, S., Kruse, S., Meister, M., Nölting, B., Ohlhorst, D., 2007. Handbuch
Konstellationsanalyse. In: Ein interdisziplinäres Brückenkonzept für die
Nachhaltigkeits-, Innovations- und Technikforschung. Oekom Verlag,
München.
Scholz, R.W., Mieg, H.A., Oswald, J.E., 2000. Transdisciplinarity in groundwater
management – towards mutual learning of science and society. Water Air Soil
Pollut. 123, 477–487.
Scholz, R.W., Spoerri, A., Lang, D.J., 2009. Problem structuring for transitions: the
case of Swiss waste management. Futures 41 (3), 171–181.
539
Scholz, R.W. (2000). Mutual learning as a basic principle of transdisciplinarity. In:
R.W., Scholz, R., Häberli, A., Bill, M. Welti (Eds.): Transdisciplinarity: Joint
Problem-Solving among Science, Technology and Society. Haffmans, Zürich,
Vol. Workbook II, S. 13–17.
Scholz, R.W., 2011. Environmental Literacy in Science and Society. In: From
Knowledge to Decisions. Cambridge University Press, Cambridge, UK.
Tress, B., Tress, G., Fry, G. (2006). Defining concepts and the process of knowledge
production in integrative research. In: Tress, B., Tres, G., Fry, G., Opdam, P.
(Eds.) From Landscape Research to Landscape Planning: Aspects of Integration,
Education and Application: 13–26.
Tress, G., Tress, B., Fry, G., 2007. Analysis of the barriers to integration in landscape
research projects. Land Use Policy, 374–385.
Truffer, B., 2007. Wissensintegration in transdisziplinären Projekten: Flexibles
Rollenverständnis als Schlüsselkompetenz für das Schnittstellenmanagement.
GAIA 16/1, 41–45.
Zierhofer, W., Burger, P., 2007. Desentangling transdisciplinarity: an analysis of
knowledge integration in problem-oriented reserach. Sci. Stud. 20 (1), 51–74.
Zscheischler, J., Rogga, S., 2015. Transdisciplinarity in land use science – a review of
concepts, empirical findings and current practices. Futures 65, 28–44.