Multivariate Frequency Analysis of

Hydro-Meteorological Variables: A
Copula-Based Approach - eBook PDF
Visit to download the full and correct content document:
https://ebooksecure.com/download/multivariate-frequency-analysis-of-hydro-meteorol
ogical-variables-a-copula-based-approach-ebook-pdf/
MULTIVARIATE FREQUENCY ANALYSIS OF
HYDRO-METEOROLOGICAL VARIABLES
This page intentionally left blank
 MULTIVARIATE
FREQUENCY ANALYSIS OF
HYDRO-METEOROLOGICAL
VARIABLES
A COPULA-BASED APPROACH

FATEH CHEBANA
Water, Earth and Environment Centre, National Institute of Scientific Research,
University of Quebec, Quebec, Canada
Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
Copyright © 2023 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage and
retrieval system, without permission in writing from the publisher. Details on how to seek
permission, further information about the Publisher’s permissions policies and our arrangements
with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency,
can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the
Publisher (other than as may be noted herein).

Notices
Knowledge and best practice in this field are constantly changing. As new research and experience
broaden our understanding, changes in research methods, professional practices, or medical
treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described herein. In
using such information or methods they should be mindful of their own safety and the safety of
others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability for any injury and/or damage to persons or property as a matter of products
liability, negligence or otherwise, or from any use or operation of any methods, products,
instructions, or ideas contained in the material herein.

ISBN: 978-0-323-95908-7

For Information on all Elsevier publications

visit our website at https://www.elsevier.com/books-and-journals

Publisher: Candice G. Janco

Acquisitions Editor: Maria Elekidou
Editorial Project Manager: Andrae Akeh
Production Project Manager: Rashmi Manoharan
Cover Designer: Harris Greg
Typeset by MPS Limited, Chennai, India
 Contents

Acknowledgments ix

1. Introduction 1
1.1 Context 1
1.2 Purpose and aims 2
1.3 Readership 4
1.4 Structure and content 5
1.5 How to read this book? 7
1.6 Final points 7
References 8

2. Multivariate hydrological frequency analysis, overview 11

2.1 General aims of hydrological frequency analysis 11
2.2 From univariate to multivariate hydrological frequency analysis 14
2.2.1 Main steps of a complete multivariate hydrological frequency analysis 17
2.2.1.1 Exploratory analysis 18
2.2.1.2 Testing basic assumptions 18
2.2.1.3 Modeling and parameter estimation 19
2.2.1.4 Multivariate quantile and return period 20
2.3 Hydrological events and their main features 21
2.3.1 Flood features 21
2.3.2 Illustrative example 23
2.3.3 Drought features 23
2.3.4 Rainfall storm features 25
2.3.5 Sediment features 26
References 26

3. Multivariate preliminary analysis 31

3.1 Context and motivation 31
3.2 Visualization 33
3.3 Cross-dependence measures 36
3.4 Outliers 39
3.4.1 Outlyingness 40
3.4.2 Threshold 40
3.5 Location measures 42
3.5.1 Sample mean 43
3.5.2 Component-wise median 43

v
vi Contents

3.5.3 Depth-based median 43

3.5.4 Spatial median 44
3.5.5 α depth-trimmed mean 44
3.6 Scale measures 45
3.6.1 α-trimmed sample dispersion matrix 47
3.6.2 Scalar form of scale 47
3.7 Asymmetry 49
3.7.1 Spherical symmetry 49
3.7.2 Elliptical symmetry 50
3.7.3 Antipodal symmetry 50
3.7.4 Angular symmetry 50
3.8 Kurtosis 51
3.8.1 Lorenz curve of Mahalanobis distance 52
3.8.2 Shrinkage plot 53
3.8.3 Fan plot 53
3.8.4 Quantile-based measure 54
References 56

4. Checking basic assumptions for multivariate hydrological

frequency analysis 57
4.1 Introduction and general considerations 57
4.2 Stationarity 60
4.2.1 Multivariate trend tests 61
4.2.1.1 Mann
Kendall type tests 62
4.2.1.2 The covariance inversion test 63
4.2.1.3 The covariance sum test 63
4.2.1.4 The covariance eigenvalue test 63
4.2.1.5 Spearman’s rho type tests 64
4.2.2 Performance evaluation 65
4.2.3 Further discussion 67
4.2.3.1 Example 67
4.3 Homogeneity 69
4.3.1 Multivariate shift detection tests 71
4.3.1.1 The Cramér test 71
4.3.1.2 The M-test 73
4.3.1.3 The T-test 73
4.3.1.4 The Wilcoxon test 74
4.3.1.5 The quality index test 74
4.3.1.6 The Zhang test 75
4.3.2 Comparisons and other approaches 75
4.3.3 Example 76
4.4 Serial independence 77
4.4.1 Serial empirical copula test 79
4.4.2 Illustrative example 82
4.5 Complete illustrative example 83
References 86
 Contents vii

5. Modeling in multivariate hydrological frequency analysis with

copula 89
5.1 Introduction 89
5.2 Description of copula models 93
5.3 Classes of copula 96
5.3.1 Archimedean copulas 97
5.3.2 Extreme-value copulas 100
5.3.3 Meta-elliptical copulas 102
5.3.4 Other classes of copulas 103
5.4 Dependence measures 104
5.4.1 Overall dependence measures 105
5.4.2 Tail dependence measures 109
5.5 Copula parameter estimation 111
5.5.1 Inference functions for margins method 112
5.5.2 Maximum Pseudo-likelihood method 112
5.5.3 Moment-based method 113
5.5.4 Multi-parameter copula estimation 113
5.6 Copula selection 114
5.6.1 Preliminary step 115
5.6.2 Copula goodness-of-fit testing 119
5.6.3 Selection criteria for copula 123
5.6.4 Margin modeling 125
References 129

6. Multivariate return period and quantile 133

6.1 Risk assessment in hydrology 133
6.2 Multivariate return periods and multivariate quantile: generalities 134
6.2.1 Definitions and presentation 135
6.2.2 Illustrative example 1 139
6.3 Methods to select combinations 140
6.3.1 Most likely design realization approach 143
6.3.2 Structure-based approach 144
6.3.3 Kendall return period approach 145
6.3.4 Conditional distribution approach 146
6.3.5 Regression-based approach 147
6.3.6 Multivariate quantile curve, proper part approach 147
6.3.7 Alpha-region approach 148
6.3.8 Ilustrative example 2 149
References 152

7. Multivariate nonstationary frequency analysis 155

7.1 Nonstationary hydrological frequency analysis 155
7.2 Multivariate nonstationary hydrological frequency analysis literature 156
7.3 Multivariate nonstationary models 158
7.3.1 Modeling description 158
viii Contents

7.3.2
Covariate-varying copulas 160
7.3.3
Covariate-varying margins 161
7.3.4
Nonstationary model selection 162
7.3.5
Bayesian multivariate nonstationary model 163
7.3.6
Modeling procedure steps 164
Step 1: descriptive study 164
Step 2: testing trends 164
Step 3: joint distribution selection 164
Step 4: nonstationary multivariate quantile and return period 166
7.3.7 Data series length and moving window series 168
7.4 Illustrative example 168
References 172

8. Multivariate regional frequency analysis 175

8.1 Regional hydrological frequency analysis 175
8.2 Multivariate regional frequency analysis 177
8.3 Delineation 179
8.3.1 Multivariate discordancy 179
8.3.2 Multivariate homogeneity 181
8.4 Multivariate estimation (index-flood model) 184
8.5 Discussion 187
References 188

Appendix A: Ties in the copula-based framework and

in hydrology 191
Appendix B: Statistical depth functions 193
Appendix C: Multivariate L-moments 197
Appendix D: p-Value computation 201
Index 203
 Acknowledgments

I would like to acknowledge all those who have contributed to the

preparation and completion of this book. I am grateful to my colleagues
and friends, from INRS and other institutions from Canada and abroad,
who gave comments directly or indirectly related to the topic of this
book, in particular, the STAHY (International Commission on Statistical
Hydrology) members. I have been greatly influenced by the discussions
and support of Abdelhameed El-Shaarawi, founder and former Editor-
in-Chief of Environmetrics.
I would like to express my gratitude to my students and team mem-
bers, especially Dorsaf Goutali, for all of her help during the last year.
All efforts, commitments, and patience from Elsevier team are greatly
appreciated.
I’m grateful to my family, especially my wife, Narimane, and our
kids, for their patience, support, and encouragement over the years.
I express my sincere thanks to each person concerned.
I dedicate this book to my parents, Achour and Laatra!!
Fateh Chebana
Quebec City, Canada

ix
This page intentionally left blank
 C H A P T E R

1
Introduction

1.1 Context

Extreme hydro-meteorological events, including floods, droughts,

and storms, may have serious economic and social consequences. Given
the importance of dealing with such adverse events, several studies
have focused specifically on each of these events. In water resource
management, one of the challenging topics is related to droughts,
whereas the design of hydraulic structures is based on floods. An accu-
rate estimation of the risk caused by these events is essential. In this
regard, hydrological frequency analysis (HFA), as a set of statistical
methods and techniques, is commonly considered. HFA is mainly com-
posed of the following steps: (1) performing exploratory analysis and
outlier detection, (2) testing the basic assumptions (stationarity, homo-
geneity and serial independence), (3) modeling and estimating model
parameters, and (4) making inference, including risk evaluation. In the
univariate HFA framework, all these steps are extensively studied and
usually considered in the analysis.
Environmental and hydro-meteorological processes, such as floods,
droughts, rainstorms, hurricanes, tornadoes, windstorms, weather
extremes, and tides, are generally complex. They are often described
by more than one correlated variable (e.g., flood volume, peak, and
duration), which involves simultaneous consideration of these vari-
ables using multivariate models and methods (see e.g., Barnett, 2012).
In particular, dealing with extreme hydro-meteorological events
requires multivariate HFA. Traditional multivariate HFA methods are
too restrictive and do not even apply in many cases. Consequently, over
the last two decades, copula functions have emerged as a preferred
method in a variety of applications, especially hydro-meteorology and
in multivariate HFA.

Multivariate Frequency Analysis of Hydro-Meteorological Variables

DOI: https://doi.org/10.1016/B978-0-323-95908-7.00009-8 1 © 2023 Elsevier Inc. All rights reserved.
2 1. Introduction

Adopting the multivariate HFA framework in hydro-meteorology in

preference to univariate HFA was extensively justified in the literature.
Indeed, univariate HFA can only provide limited representativity and
understanding of extreme events and their probability of occurrence. In
addition, the univariate framework of each event characteristic sepa-
rately does not take into account their dependence structure, leading to
potentially less accurate risk estimation. Nevertheless, univariate HFA
can be useful in some situations such as when only one variable is sig-
nificant for design purposes or when the dependence between these
variables is not significant. However, multivariate HFA is more reliable
for modeling hydro-meteorological variables, leads to better risk assess-
ment, and is a more flexible framework (see Chapter 2 for more details).

1.2 Purpose and aims

Multivariate HFA is a very active research topic in statistical hydro-

meteorology. A relatively large body of literature, dealing with multi-
variate HFA, is available mostly as journal papers (theoretical develop-
ments, case studies, etc.). In general, these papers treat specific aspects
such as a hydrological event (e.g., flood, drought), a particular step of
the analysis (e.g., modeling, testing, exploratory analysis), or a given
statistical approach or technique (e.g., copulas, L-moments). In addition,
most of the literature focuses on the modeling step mainly based on the
copula function. Fig. 1.1 illustrates some of the aspects of multivariate
HFA highlighted in the literature and their importance and the volume
of studies focused on this. It shows that the modeling step, especially
based on copulas, is most prevalent the literature. Although copula and
modeling are important and essential, many other aspects and steps
also need to be considered to perform a complete and appropriate anal-
ysis. Multivariate HFA associated literature as papers and reports is not
easily accessible to practitioners and students. Hence, this has led to an
increasing gap between research and practice in this field. Therefore,
the desperate need of a reference book where the reader can find all the
relevant material covering the different steps and situations of a multi-
variate HFA in a simplified and accessible presentation, the connections
between them as well as a complete overview of all steps of the analysis
has been felt.
This book attempts to reduce or eliminate some of the challenges and
difficulties faced by practitioners in multivariate HFA. This book com-
piles all the relevant background material and new developments in
one place and also presents this material in a homogeneous and peda-
gogical way in order to allow students, engineers, practitioners, and
researchers to access and use efficiently all the information about this

Multivariate Frequency Analysis of Hydro-Meteorological Variables

 1.2 Purpose and aims 3

FIGURE 1.1 Illustration of the importance/volume of studies in the literature on each

step/topic in multivariate hydrological frequency analysis.

topic. In addition, given the advanced nature of the approaches in mul-

tivariate HFA and the ongoing developments, even though useful and
necessary, they are complex for a majority of practitioners and students,
especially readers without statistical background. Therefore, this book
tries to simplify the presentation of these concepts and hence aims to fill
the gap between theory and practice. Also, a major part of the literature
neglects some of steps of the analysis (Fig. 1.1), potentially leading to
incomplete analysis or even wrong conclusions. Consequently, this
book highlights the importance of those steps and provides the recent
and advanced approaches to deal with them as along with examples
from real-life situations.
To the best of the author’s knowledge, there is no such existing book
that deals specifically and directly with the topic of multivariate HFA as
a whole and in an integrated manner. Indeed, the existing books mainly
cover copula functions either in hydrology or statistics, such as Salvadori
et al. (2007), Zhang and Singh (2019), and Chen and Guo (2019) in water
sciences and Joe (2014) and Hofert et al. (2018) in statistics. This book pro-
vides a solid platform bringing together multivariate HFA tools in hydro-
meteorological practice and contributes to filling the gap between theory
and practice and the advancement of the field of statistical hydro-
meteorology. This book enables the reader to perform a well-justified
multivariate HFA covering all relevant steps and aspects of the analysis,

Multivariate Frequency Analysis of Hydro-Meteorological Variables

4 1. Introduction

including the preliminary important steps (e.g., testing the assumptions)

and useful extensions (nonstationary, regional). This book provides
detailed and comprehensive descriptions of the techniques and all steps
involved in performing a complete multivariate HFA.
In this book, the copula-based approach is given due importance and
a large chapter (Chapter 5) is dedicated to this topic, along with cover-
ing other important topics, including hypothesis testing of the basic
assumptions, the return period and quantile, and preliminary analysis
such as outliers and descriptive statistics. Where appropriate, some
examples based on the same datasets are presented across several chap-
ters to show how to perform the analysis and the steps involved.

1.3 Readership
This book is aimed to be a reference for researchers, practitioners, and
graduate students in the field of multivariate HFA, with a clear and com-
prehensive presentation of all relevant approaches and steps involved in
performing a complete analysis. It also serves as an ideal multidisciplin-
ary introductory book for hydrologists, climatologists, and engineers to
make themselves familiar with the most up-to-date and advanced multi-
variate methodologies in hydrological design, planning, and manage-
ment, to mention some, and their practical applications. This book also
serves as a guide for the readers in applying the most recent approaches
available toward evaluating hydro-meteorological risks, designing
hydraulic structures, and teaching (faculty members), and as state-of-the
art methodologies to move rapidly to the next level in their research pro-
jects (graduate students and postdocs).
A variety of readers from industry, government agencies, or academia
(for research and graduate teaching) as well as statisticians and non-
statistician readers can benefit from this book. Advanced approaches are
presented in an easy-to-understand manner and with an appropriate level
of detail. Even though the primary target readers are hydrologists, clima-
tologists, engineers and statisticians, given that some material is interdisci-
plinary, it can be used for reference by practitioners from other application
fields such as financial institutions, insurance companies (damages caused
by floods and droughts), earth sciences, environmental modeling, and gov-
ernment agencies (e.g., public safety, environment and transportation).
Readers interested in understanding theoretical concepts and practical
aspects related to copula-based modern multivariate HFA can find this
book with in-depth technical details extremely helpful, where advanced
and complex mathematics/statistics have been avoided to the extent possi-
ble. Nevertheless, basic knowledge of probability and statistics, such as
random vectors, estimation methods, and statistical tests, is expected.

Multivariate Frequency Analysis of Hydro-Meteorological Variables

 1.4 Structure and content 5

1.4 Structure and content

Over the last two decades, multivariate HFA has gained icreasing atten-
tion in both applications and theoretical developments. This book is com-
posed of seven other chapters and an Appendix. Chapter 2 introduces HFA
and briefly makes the connection between the subsequent chapters.
Chapters 3
6 discuss the main HFA steps involved in a standard multivar-
iate HFA. Chapters 7 and 8 are dedicated to advanced analysis as well as
extensions of the standard analysis, dealing with multivariate nonstationary
modeling and multivariate regional analysis. To maintain the fluency of the
content, some technical concepts are presented in the Appendix.
Chapter 2 provides an overview and the basics of HFA. It starts with
describing the general aims and goals of HFA as well as the essential
concepts of return period and quantile for hydrological risk assessment.
Then, it briefly introduces the main steps involved in performing the
whole HFA. This chapter also discusses the advantages and challenges
faced while transitioning from univariate to multivariate HFA frame-
works, especially using copula functions. The multivariate character of
a number of hydrological phenomena, such as floods, droughts, rainfall
storms, and sediments, is described as along with their main features to
be treated in the multivariate HFA framework.
Chapter 3 treats the preliminary analysis within the framework of the
multivariate HFA. Several statistical properties of the multivariate sam-
ple are discussed, such as location, scale, skewness, kurtosis as well as
outlier detection. This step can be useful in summarizing, describing,
and understanding the information contained in the data series. On the
other hand, this preliminary analysis is useful for modeling hydrologi-
cal variables and hence for risk evaluation. The presented methods are
general and can be adapted and applied to a variety of hydro-
meteorological events such as floods, droughts, storms, and sediment
transport along with other fields.
Chapter 4 addresses the testing step within the framework of multi-
variate HFA. In this chapter, the corresponding techniques and methods
are presented in more detail with a few examples. Nonstationarity, het-
erogeneity, and serial dependence need to be tested before the modeling
step in a multivariate HFA. The testing step is important to ensure the
basic assumptions are met and thus the selected models are appropriate.
These statistical tests are generic and can be adapted and applied to a
variety of hydro-meteorological variables as well as to other fields.
Chapter 5 introduces the modeling step of multivariate HFA based
on copula functions. Even though copula modeling is the heart of the
multivariate HFA, the preliminary analysis and testing of the basic
assumptions should be performed first (presented, respectively, in

Multivariate Frequency Analysis of Hydro-Meteorological Variables

6 1. Introduction

Chapters 3 and 4). Here, the basic assumptions are assumed to be ful-
filled (see Chapter 4). This chapter also presents an overview of the sta-
tistical approaches and methods regarding copula modeling, including
parameter estimation and goodness-of-fit testing as well as model selec-
tion criteria with illustrations.
Chapter 6 examines the last step of multivariate HFA, which deals
with the inference, in particular risk assessment in terms of return periods
or quantiles. This step is performed based on the analysis and decisions
made in all previous steps (described in previous chapters), especially the
modeling step (Chapter 5). Here, risk assessment in hydrology is briefly
presented, followed by the basics regarding multivariate return period
and quantile, and, finally, an overview of the statistical approaches and
methods regarding the selection of the multivariate combinations for a
given return period with illustrative examples.
Chapter 7 treats nonstationarity in the multivariate setting. Combining
those two aspects (nonstationarity and multivariate) leads to the multivari-
ate nonstationary HFA, which aims at estimating hydro-meteorological
quantiles (risks) in the presence of nonstationarity (caused, for instance, by
climate change). Prior to performing nonstationary analysis, appropriate
tests should be accomplished (Chapter 4). Hydro-climatology phenomena
are naturally multivariate with stationarity assumption either fulfilled or
not. Therefore, it is more realistic and representative to consider the joint
multivariate and nonstationary HFA setting. This chapter briefly intro-
duces the basics of nonstationarity in HFA followed by presenting the
multivariate nonstationary context. Then, the modeling methodology of
the latter is described followed by an illustrative example.
The last chapter briefly introduces the basics of regional HFA fol-
lowed by presenting the multivariate context of regional frequency anal-
ysis (RFA). Then, the delineation and the regional estimation, as the two
main components of RFA, are presented. This chapter also deals with
RFA in the multivariate setting. Combining regional and multivariate
aspects leads to the multivariate RFA, which aims at estimating hydro-
meteorological quantiles (risks) at ungauged sites. Usually, in the latter,
no hydrological data are available unlike the at-site (local) HFA analysis
seen in the previous chapters. RFA in the univariate setting is widely
used by hydrologists. The multivariate nature of hydro-meteorological
phenomena is present at sites either gauged or ungauged. Therefore, it
is more realistic/useful to consider multivariate RFA.
In order to facilitate readability, some of the technical statistical
concepts and tools needed in the previous chapters are outlined in
the Appendix, which includes ties in the multivariate framework and
in hydrology, statistical depth functions, multivariate L-moments, and
p-value computation. These tools are also generic and can be useful
in other fields and disciplines as well.

Multivariate Frequency Analysis of Hydro-Meteorological Variables

 1.6 Final points 7

1.5 How to read this book?

On one hand, chapters can be followed independently based on the
requirements of the reader. For instance, if the objective is to study
trends, the reader can directly go to Chapter 4 or if the reader is inter-
ested in multivariate modeling, including model selection and parame-
ter estimation, then the reader can find the appropriate material in
Chapter 5. On the other hand, if the reader aims to perform a complete
multivariate HFA, it is recommended to read the chapters in a sequen-
tial order starting from Chapter 2 to Chapter 6 for a standard HFA.
Since Chapters 7 and 8 present advanced material, it is recommended to
start with the chapters dealing with the standard analysis (Chapters
2
6). This is illustrated in Fig. 1.2, which provides links between all
chapters as a diagram showing how they are connected and reading
path options.

1.6 Final points

Even though the book attempts to address most of the practical

issues and methodological facets related to multivariate HFA, it does
not pretend to cover all works in the field. In particular, the applications
and case studies are focused on floods even though the presented mate-
rial and approaches are valid for other hydro-meteorological applica-
tions or in other fields as well. Indeed, floods are the most common
natural hazards that account for close to half of the total worldwide dis-
asters, affecting hundreds of millions of persons (NDRR, 2020). In addi-
tion, multivariate HFA is largely used to study floods. As regards
theory, focus was made on the bivariate setting while the high-
dimension situations are discussed briefly. The bivariate setting is opted
for since it is the most studied case and simple as well. Some other
topics like the uncertainty in the estimation as well as vine copulas are
only mentionned mainly because this book is a first of its kind in this
field, and most of the these concepts are still in development in hydro-
meteorology.
Even though reasonable efforts have been made to publish reliable
data, illustrations, and information, the author and publisher are not
responsible for the validity of all the presented materials or the conse-
quences of their use.
The book is entirely based on the subject knowledge of the author.
Yet, as in any other publication, it may contain some errors. Comments,
corrections, and suggestions from the readers about the material pre-
sented in the book and related matters are welcome.

Multivariate Frequency Analysis of Hydro-Meteorological Variables

8 1. Introduction

FIGURE 1.2 Illustration of the different path options that can be taken by the reader.

References
Barnett, V. (2012). Outliers, multivariate. Encyclopedia of Environmetrics. https://doi.org/
10.1002/9780470057339.vao020.
Chen, L., & Guo, S. (2019). Copulas and its application in hydrology and water resources
(p. 290) Singapore: Springer.
Hofert, M., Kojadinovic, I., Mächler, M., & Yan, J. (2018). Elements of copula modeling with R
(pp. 9
79). Springer.

Multivariate Frequency Analysis of Hydro-Meteorological Variables

Another random document with
no related content on Scribd:
 between May 1, 1898, and June 30, 1899.

[Off. = Officer; Enl. = Enlisted Men.]

COUNTRY KILLED. DIED OF

DISEASE. ACCIDENT.
WOUNDS
Off. Enl. Off. Enl. Off.
Enl. Off. Enl.
REGULARS.
United States 1 5 10 32
874 1 51
Cuba 19 184 5 60 8
381 7
Porto Rico 3
73 3
Hawaiian Islands
10 1
Philippine Islands 4 81 1 33 4
109 10
At sea 1 11 4
77

Total 24 270 7 114 51

1,524 1 72

VOLUNTEERS.
United States 1 87
2,836 3 111
Cuba 3 39 10 16
457 2 12
Porto Rico 3 1
157 5
Hawaiian Islands
33 1
Philippine Islands 14 146 3 67 5
215 6
At sea 5
122 2

Total 17 188 3 78 114

3,820 5 137

Aggregate 38 458 10 192 165

5,344 6 209

COUNTRY. DROWNED. SUICIDE.

MURDER TOTAL.

HOMICIDE
Off. Enl. Off. Enl. Off.
Enl. Off. Enl.
REGULARS.
United States 1 16 19
18 35 993
Cuba 7 5
6 32 650
Porto Rico 1 3
1 3 81
Hawaiian Islands 1
12
Philippine Islands 19 1 3
1 10 256
At sea 1 4 2
6 94
Total 2 48 1 32
26 86 2,086

VOLUNTEERS.
United States 23 1 15
22 91 3,008
Cuba 4
3 21 525
Porto Rico 2 1
1 1 169
Hawaiian Islands
34
Philippine Islands 1 9 3
23 446
At sea 2 1
5 127

Total 1 40 1 20
26 141 4,309

Aggregate 3 88 2 52
52 224 6,395

{630}

Recapitulation of casualties in action in the armies of the

United States between May 1, 1898, and June 30, 1899.

[Off. = Officer; Enl. = Enlisted Men.]

COUNTRY. KILLED. WOUNDED.

TOTAL. AGGREGATE
Off. Enl. Off. Enl. Off.
Enl.
REGULARS.

Cuba 18 183 86 1,126 104

1,309 1,413
Porto Rico 1 2 15 2
16 18
United States 1 5 10 1
15 16
Philippines, to
August 13, 1898 7 1 25 1
32 33
Philippines since
February 4, 1899 2 74 20 410 22
484 506

Total 21 270 109 1,586 130

1,856 1,986

VOLUNTEERS.

Cuba 3 39 15 218 18
257 275
Porto Rico 3 2 21 2
24 26
Philippines,
to Aug. 13, 1898 11 9 74 9
85 94
Philippines, since
February 4, 1899 14 135 62 865 76
1,000 1,076

Total 17 188 88 1,178 105

1,366 1,471

Grand total 38 458 197 2,764 235

3,222 3,457

HOSPITALS.

From the declaration of war with Spain to September 20, 1899,

there have been established:

Beds.
20 field division hospitals, averaging 250 beds
each 5,000
31 general hospitals with a total capacity of
about 13,800
Railroad ambulance train
270
4 hospital ships
1,000

Total
20,070

In addition to these over 5,000 cases were treated in civil

hospitals. It is difficult even to approximate the number of
men treated in these hospitals. During that period somewhat
over 100,000 cases were admitted on sick report, a number
equal to 2,147 per 1,000 of strength during the year, or to
179 per 1,000 per month—the ratio of admissions to hospital
cases being 13 to 8. Using these data as a basis, and assuming
the mean strength of the Army (Regulars and Volunteers) to have
been 154,000, it would appear that from May 1, 1898, to
September 20, 1899, about 275,000 cases have been treated in
 these hospitals.

TRANSPORTATION OF SPANISH PRISONERS OF WAR TO SPAIN.

The following is a statement showing the dates of embarkation,

names of vessels, and number of officers, enlisted men, and
others who took passage:

[Date = Date of Embarkation,

Off. = Officers
Men = Enlisted men,
Women = Women and children over 5 years of
age,
Priests = Priests and Sisters of Charity.]

Date Name of Vessel. Off. Men Women

Priests Total

August 9 Alicante 38 1,069 6

11 1,124
August 14 Isla de Luzon 137 2,056 40
4 2,237
August 16 Covadonga 109 2,148 79
2,336
August 19 Villaverde 52 565 34
651
August 19 Isla de Panay 99 1,599 26
5 1,729
August 22 P. de Satrustegui 128 2,359 68
2,555
August 25 Montevideo 136 2,108 122
2 2,368
August 27 Cherihon 18 905 37
960
August 28 Colón 100 1,316 59
1,475
August 30 do 23 726 5
754
September 1 Leon XIII 113 2,209 108
2,430
September 3 San Ignacio 59 1,408 20
12 1,499
September 6 Leonora 15 1,118
1,333
September 12 Cindad de Cadiz 53 19
14 86
September 17 San Augustin 65 800 45
910
September 17 San Francisco 18 588 11
617

Total 1,163 20,974 679

48 22,864

ARMS AND AMMUNITION CAPTURED AT SANTIAGO.

Mauser carbines, Spanish, 7 mm 16,902

Mauser rifles, Argentine, 7½ mm 872
Remington rifles, 7 mm 6,118
Total rifles 23,892

Mauser carbines, Spanish 833

Mauser carbines, Argentine 7½ mm 84
Remington carbines, 7½ mm 330
 Total carbines 1,247

Revolvers 75

{631}

Mauser-Spanish—cartridges, 7 mm. 1,500,000

Mauser-Argentine—cartridges, 7½ mm. 1,471,200
Remington cartridges, 7½ mm 1,680,000

Total. 4,651,200

Nine hundred and seventy-three thousand

Remington
cartridges, 7½ mm., worthless.

STRENGTH OF THE NAVY, REGULAR AND AUXILIARY.

The number of enlisted men allowed by law prior to the

outbreak of hostilities was 12,500. On August 15, when the
enlisted force reached its maximum, there were 24,123 men in
the service. This great increase was made necessary by the
addition of 128 ships to the Navy. The maximum fighting force
of the Navy, separated into classes, was as follows:

Battle ships (first class). 4

Battle ships (second class). 1
Armored cruisers. 2
Coast defense monitors. 6
Armored ram. 1
Protected cruisers. 12
Unprotected cruisers. 3
Gunboats. 18
 Dynamite cruiser. 1
Torpedo boats. 11
Vessels of old Navy,
Including monitors. 14
Auxiliary Navy:
Auxiliary cruisers. 11
Converted yachts. 28
Revenue cutters. 15
Light-house tenders. 4
Converted tugs. 27
Converted colliers. 19
Miscellaneous. 19

NAVAL, PRISONERS OF WAR CAPTURED OFF SANTIAGO,

JULY 3, 1898.

Officers. 99
Enlisted men. 1,675

CASUALTIES IN ACTION.

ENGAGEMENT. Casualties Killed

Wounded Died later

from wounds
Action at Manila Bay,
May 1 9 9
Action off Cienfuegos,
May 11 12 1 11
1
Action off Cardenas,
May 11. 8 5 3
Action off San Juan,
 Porto Rico, May 12 8 1 7
Engagements at Guantanamo,
Cuba, June 11 to 20 22 *6 16
Engagement off Santiago:
June 22 10 1 9
July 3 11 1 10
Miscellaneous:
Yankee, June 13. 1 1
Eagle, July 12 1 1
Bancroft, August 2 1 1
Amphitrite, August 7 1 1 †l

Total 84 16 68
2

* One accidentally killed.

† Accidentally shot.

Congressional Record,
February 1, 1901, pages 1941-1962.

UNITED STATES OF AMERICA: A. D. 1898-1899.

Investigation of the conduct of the War Department
in the war with Spain.

Severe criticism of the conduct of the War Department during

the war with Spain, including many charges of inefficiency in
its service, produced by improper appointments made for
political reasons, and other charges of misdoing in the
purchase of supplies, under influences either political or
otherwise corrupt, led to the appointment by the President, in
September, 1898, of an investigating commission, composed of
nine soldier and civilian members, as follows:
General Grenville M. Dodge, President.
Colonel James A. Sexton.
Colonel Charles Denby,
Captain Evan P. Howell,
Honorable Urban A. Woodbury,
Brigadier-General John M. Wilson, U. S. A.,
General James A. Beaver,
Major-General Alexander McD. Cook, U. S. A.,
Dr. Phineas S. Conner.

The report of the Commission, made in the following February,

cannot be said to have been a convincing and satisfactory one
to the country at large. It was indignantly described as a
"whitewashing report," even by many journals and writers of
the party in power. Its inquiries did not appear to have been
keenly and impartially searching; its conclusions were not
thought to be drawn with a rigorous and fearless hand.

The charges against the War Department which excited most

feeling and drew most public attention related to the quality
of the fresh beef supplied to the army, which was in two
forms, refrigerated and canned. Major-General Miles,
commanding the Army, had declared that much of the
refrigerated beef furnished to the soldiers should be called
"embalmed beef," maintaining that it had been "apparently
preserved with secret chemicals, which destroy its natural
flavor" and which were believed to be "detrimental to the
health of the troops." He intimated that hundreds of tons of
such beef had been contracted for by the Commissary-General
"under pretense of experiment." In repelling this serious
accusation, Commissary-General Charles P. Eagan read a
statement before the Commission, so violent and unmeasured in
its vituperation of the commanding general that it was
returned to him for correction; many newspapers declined to
publish it, and he was subsequently tried by court-martial in
consequence—as related below. The conclusion of the
Commission on the subject of the charges relating to
refrigerated beef was stated in its report as follows:
 "The Commission is of the opinion that no refrigerated beef
furnished by contractors and issued to the troops during the
war with Spain was subjected to or treated with any chemicals
by the contractors or those in their employ."

Concerning the canned beef, which had caused much disgust in

the army, the Commission reported:

"The result of our own testing and of all the analyses made at
our instance … is that the canned meat which has been brought
to our attention is pure, sound, and nutritive. It has not
been found to contain any acids or any deleterious substance,
but to be unadulterated meat. The testimony before us is that
the canned meat is not, in general, intended to be issued to
troops except as an emergency ration. The preponderance of the
proof is that meat on the hoof and the refrigerated beef are
more acceptable. A number of officers and others have
testified that the meat is unpalatable. Its palatability
greatly depends upon the mode in which it is cooked. In a
tropical climate, carried on the march, exposed to heat, the
meat so changes in appearance as to become repulsive. In the
Navy, where the meat is properly cared for, there has been no
complaint, so far as has appeared in evidence before us. After
careful consideration we find that canned meat, as issued to
the troops, was generally of good quality, was properly
prepared, and contained no deleterious substance.
{632}
At times probably material of poor quality is issued; in one
of the cans sent to us and examined by the chemist a large
amount of gristle was found. That it was not issued 'under
pretense of an experiment' is indicated by the fact that it
has been in use in the Army for more than 20 years."

On the general management of the Quarter-master's Department,

with which much fault had been found, the Commission reported:
"The conclusions drawn … are as follows:
"1. The Quartermaster's Department, a month before war was
declared, was neither physically nor financially prepared for
the tremendous labor of suddenly equipping and transporting an
army over ten times the size of the Regular Army of the United
States.

"2. That the department devoted the ability, zeal, and

industry of its officers to accomplish the herculean task
before it so soon as funds were made available and war was
declared.

"3. That it deserves credit for the great work accomplished,

for the immense quantity of materials obtained and issued
within so short a period, and for its earnest efforts in
reference to railroad transportation and in protecting the
great interests of the General Government committed to its
charge. Its officers, especially those at the head-quarters of
the department and at its depots, worked earnestly and
laboriously day and night, sparing themselves in no possible
way.

"4. There appears to have been a lack of system, whereby, even

as late as October, troops in camps and in the field were
lacking in some articles of clothing, camp and garrison
equipage; and hospitals, at least at two important localities
in the South—Fort Monroe, Virginia, and Huntsville,
Alabama—lacked stoves, while at Huntsville fuel was wanting.

"5. There appears to have been lack of executive or

administrative ability, either on the part of the
Quartermaster's Department or the railroad officials, in
preventing the great congestion of cars at Tampa and
Chickamauga when these camps were first established, which
congestion caused delay, annoyance, and discomfort to the
large bodies of troops concentrating at those places.

"6. There appears to have been a lack of foresight in

preparing and promptly having available at some central
locality on the seacoast the necessary fleet of transports
which it seemed evident would be required for the movement of
troops to a foreign shore, and, finally, when the call came
suddenly and the emergency was supreme, the department appears
not to have fully comprehended the capacity of the fleet under
its command; not to have supplied it with a complete outfit of
lighters for the immediate disembarkation of troops and
supplies; to have accepted without full investigation the
statement that the vessels were capable of transporting 25,000
men, while really they could not and did not transport more
than 17,000 with their artillery, equipments, ammunition, and
supplies, and lacked sufficient storage room for the necessary
amount of wagon transportation—that very important element
in the movement of an army in the face of an enemy.

"7. The Quartermaster's Department should maintain on hand at

all times a complete supply for at least four months for an
army of 100,000 men of all articles of clothing, camp and
garrison equipage, and other quartermaster's supplies which
will not deteriorate by storage or which cannot at once be
obtained in open market.

"Finally. In the opinion of this commission, there should be a

division of the labor now devolving upon the Quartermaster's
Department."

In another part of its report, dealing especially with the

Santiago campaign, the Commission makes a statement which
seems to reflect some additional light on the sixth paragraph
of the finding quoted above, relative to the unpreparedness of
the quartermaster's department for the landing of the Santiago
expedition. It says:

"The Navy Department, on the 31st of May, 1898, sent the

following communication to the honorable the Secretary of War:
'This Department begs leave to inquire what means are to be
employed by the War Department for landing the troops,
artillery, horses, siege guns, mortars, and other heavy
objects when the pending military expedition arrives on the
Cuban coast near Santiago. While the Navy will be prepared to
furnish all the assistance that may be in its power, it is
obvious that the crews of the armored ships and of such others
as will be called upon to remove the Spanish mines and to meet
the Spanish fleet in action can not be spared for other purposes,
and ought not to be fatigued by the work incident to landing
of the troops and stores, etc.' This information, so far as
can be ascertained, was never communicated to either General
Miles or General Shafter; the expedition therefore left Tampa
with no facilities for landing other than were afforded by the
boats of the several transports conveying the expedition, with
the exception of several lighters and steam tugs of light draft,
such as could be hastily secured."

On the conduct of the Medical Department, which was another

matter of investigation, the Commission reported: "To sum up,
in brief, the evidence submitted shows:

" 1. That at the outbreak of the war the Medical Department

was, in men and materials, altogether unprepared to meet the
necessities of the army called out.

"2. That as a result of the action through a generation of

contracted and contracting methods of administration, it was
impossible for the Department to operate largely, freely, and
without undue regard to cost.

"3. That in the absence of a special corps of inspectors, and

the apparent infrequency of inspections by chief surgeons, and
of official reports of the state of things in camps and
hospitals, there was not such investigation of the sanitary
conditions of the army as is the first duty imposed upon the
Department by the regulations.
 "4. That the nursing force during the months of My, June, and
July was neither ample nor efficient, reasons for which may be
found in the lack of a proper volunteer hospital corps, due to
the failure of Congress to authorize its establishment, and to
the nonrecognition in the beginning of the value of women
nurses and the extent to which their services could be
secured.

"5. That the demand made upon the resources of the Department
in the care of sick and wounded was very much greater than had
been anticipated, and consequently, in like proportion, these
demands were imperfectly met.

{633}

"6. That powerless as the Department was to have supplies

transferred from point to point, except through the
intermediation of the Quartermaster's Department, it was
seriously crippled in its efforts to fulfil the regulation
duty of 'furnishing all medical and hospital supplies.'

"7. That the shortcomings in administration and operation may

justly be attributed, in large measure, to the hurry and
confusion incident to the assembling of an army of untrained
officers and men, ten times larger than before, for which no
preparations in advance had been or could be made because of
existing rules and regulations.

"8. That notwithstanding all the manifest errors, of omission

rather than of commission, a vast deal of good work was done
by medical officers, high and low, regular and volunteer, and
there were unusually few deaths among the wounded and the
sick.

"What is needed by the medical department in the future is—

"1. A larger force of commissioned medical officers.

"2. Authority to establish in time of war a proper volunteer
hospital corps.

"3. A reserve corps of selected trained women nurses, ready to

serve when necessity shall arise, but under ordinary
circumstances, owing no duty to the War Department, except to
report residence at determined intervals.

"4. A year's supply for an army of at least four times the

actual strength, of all such medicines, hospital furniture,
and stores as are not materially damaged by keeping, to be
held constantly on hand in the medical supply depots.

"5. The charge of transportation to such extent as will secure

prompt shipment and ready delivery of all medical supplies.

"6. The simplification of administrative 'paper work,' so that

medical officers may be able to more thoroughly discharge
their sanitary and strictly medical duties.

"7. The securing of such legislation as will authorize all

surgeons in medical charge of troops, hospitals, transports,
trains, and independent commands to draw from the Subsistence
Department funds for the purchase of such articles of diet as
may be necessary to the proper treatment of soldiers too sick
to use the army ration. This to take the place of all
commutation of rations of the sick now authorized.

"Convalescent soldiers traveling on furlough should be

furnished transportation, sleeping berths or staterooms, and
$1.50 per diem for subsistence in lieu of rations, the soldier
not to be held accountable or chargeable for this amount."

Report of the Commission, volume 1.

Public opinion of the report, when divested of partisan

prejudice, was probably expressed very fairly in the following
comments of "The Nation," of New York:

"The two leading conclusions of the court of inquiry as to the

quality of the beef supplied to our troops during the war with
Spain, are in accordance with the evidence and will be
accepted as fairly just by the country. The court finds that
so far as the canned roast beef was concerned, the charges
which General Miles made against it as an unsuitable ration
are sustained, but that as regards the use of chemicals in the
treatment of refrigerated beef his charges were not
established. If instead of saying 'not established,' the court
had said 'not fully sustained,' its verdict would have been
above criticism on these two points. There was evidence of the
use of chemicals, but it was not conclusive and was flatly
contradicted. There is no doubt whatever that the use of the
refrigerated beef was a blunder, but there was very little
evidence to sustain a more serious charge than that against
it.

"But while the court has found justly on these points, it is

difficult to read its report without feeling that its members
did so reluctantly, and that, if left to follow their
inclinations, they would have censured General Miles and
allowed everybody else concerned to go free. General Miles is
the one person involved whom they allow no extenuating
circumstances to benefit in their report. At every opportunity
they take the worst possible view of his conduct, while almost
invariably taking the most lenient view possible of nearly
everybody else. … So far as the findings of the court apply to
Eagan's conduct, they are condemnatory in general terms, but
they do not seek to go behind him for the reasons for his
conduct. … No attention whatever is paid to the evidence of
several reputable witnesses that Eagan had told them that he
had to buy of certain contractors; none is paid, either, to
the evidence of Eagan's subordinates that he himself so
altered the refrigerated beef contracts that no one could say
whether they called for preservation for seventy-two hours or
twenty-four. Leniency of this kind is never shown toward
General Miles."

The Nation,
May 11, 1899.

Perhaps a weightier criticism is represented by the following,

which we quote from an article contributed to "The
Independent" by General Wingate, President of the National
Guard Association of the United States: "So far as the
refrigerated beef was concerned, the truth probably is that
there was little, if any, 'embalming' about it. Soldiers
generally agree that the beef itself was almost universally
good. … General Miles, on the other hand, was clearly right in
asking that the troops might be furnished with beef cattle on the
hoof, which could follow the army over any road and which
would keep in good condition on the luxuriant grasses of Cuba
and Porto Rico. This was the system pursued in our Civil War.
No one has yet explained why it was abandoned for the
experiment of furnishing this kind of beef to places in the
tropics where it had to be hauled in wagons for many hours
over muddy roads, and when most of the wagons required to move
it promptly had to be left behind for want of water
transportation.

"The matter of the refrigerated or so-called 'embalmed' beef

is, however, of very slight consequence compared with that of
the canned roast beef. The use of that beef as an army ration
in this country, at least, was new. Officer after officer has
testified before the court of inquiry that they never saw it
so issued before the Cuban campaign. It is true that the navy
uses it, but the facilities on shipboard for caring for and
cooking food are so different and so superior to those of an
army in the field that no comparison can justly be made
between them. Moreover, as was recently stated in the 'Army
and Navy Journal,' the belief is general in the navy that the
 canned beef it had rejected on inspection was afterward sold
to the army and accepted by it without inspection.
{634}
Be this as it may, the evidence is overwhelming that the
canned roast beef which was issued to the army was repulsive
in appearance and disagreeable in smell. … Governor Roosevelt
says in his testimony that 'from generals to privates he never
heard any one who did not condemn it as an army ration.' Its
defects appeared on the voyage to Santiago, if not before. It
was then so bad that the men would not touch it, and as
Governor Roosevelt says in his article in 'Scribner's,' his
Rough Riders, who certainly were not particular, could not eat
it, and as it constituted one-third of the rations, his men
had to go hungry. And yet, in spite of these facts, a million
pounds of that beef was purchased from Armour & Co. alone, and
its issue was continued not only in Cuba but in Porto Rico.
What is worse than all, after its defects were fully known it
was issued as a traveling ration to the fever-racked men on
their homeward voyage to this country; men who needed and were
entitled to receive the most nourishing food and to whom this
indigestible stuff was poison. This should never be forgotten
or forgiven by the plain people of the country. …

"No one in authority has been willing to admit that there was
the slightest thing wrong, or the least need for improvement
in his department. … This is another of the hundreds of
examples which have occurred in our past war, and which will
continue to take place in the future until the whole staff
system of the army has been rectified, of the reign of that
hide bound bureaucratic spirit which induces the head of a
department in Washington to decide in his office what should
be used by the troops in the field without practical
experience on the subject, and to stubbornly close his eyes
and ears to everything which will tend to show that it is
possible that his department has made a mistake. …

"It is noticeable that so far not an official in any of the