~ www.iahs.infoINTERNATIONAL ASSOCIATION OF HYDROLOGICAL SCIENCES

IAHS, established in 1922, is a nonprofit-making nongovernmental scientific organizationcommitted to serving the science of hydrology and the worldwide community of hydrologists

IAHS is concerned with all components of the hydrological cycle and related processes, as well asthe application of hydrology and its implications for society and the environment. Currently 10International Commissions deal with the hydrological cycle, water for society and the environment,and specific techniques:

- Surface Water - Water Resources Systems- Groundwater - Coupled Land-Atmosphere Systems- Continental Erosion - Remote Sensing- Snow and Ice Hydrology - Tracers- Water Quality - Statistical Hydrology

The Panta Rhei - Everything Flows: Change in Hydrology and Society decade (2013-2022), andthe earlier Prediction in Ungauged Basins initiative (PUB, 2003-2012), involve all Commissions.

Information about IAHS, the Commissions, PUB, and their activities is available at www.iahs.info.or from the IAHS Secretary General:

Professor Christophe CudennecIAHS, UMR SAS, Agrocampus Ouest,CS 84215, 35042 Rennes Cedex, Francee-mail: Gudennec@agrocampus-ouest.fr

Membership of IAHS is free of charge and open to anyone wishing to participate in IAHSactivities, such as IAHS symposia and workshops. The worldwide membership is now more than5500. Register online at www.iahs.info. For information about IAHS and publications, contact:

Membership Secretary,IAHS Press, Centre for Ecology and Hydrology, Wallingford, Oxfordshire OX10 8BB, UKe-mail: jilly@iahs.demon.co.uk

Members receive the Newsletter and information about IAHS activities via email, and can obtaindiscounts on Hydrological Sciences Journal and most other IAHS publications of 25%.

IAHS Publications IAHS has a long tradition of publishing titles on topics in hydrology andrelated themes. This series, the IAHS Red Books (ISSN 0144-7815), provides a means forinternational dissemination of research results and discussion from authors worldwide. Many ofthe publications are collections of peer-reviewed papers selected from the proceedings of IAHS-convened and other international meetings. Others are reports on specific topics. Publications1-250 can be downloaded from www.iahs.info.click on Publications. IAHS also publishesHydrological Sciences Journal (ISSN 0262-6667; available online) and produces a Series ofSpecial Publications (ISSN 1024-4891), and the Benchmark Papers in Hydrology Series (ISSN1993-4572). For more information about these go to www.iahs.info. or ask jilly@iahs.demon.co.uk(address as above)for catalogues and flyers.

Task Force for Developing Countries (TFDC) Distribution of IAHS Publications to "countries inneed" is organised by TFDC. Hydrological Sciences Joumal and other IAHS publications aredistributed free to some 50 organizations worldwide, to help scientists there to acquire pioneeringresults of hydrological sciences and their application. The publications are provided on theunderstanding that they are made available to specialists both inside and outside the recipientorganizations. All correspondence concerning the distribution of IAHS publications to "countries inneed" should be sent to:

Professor Denis HughesInstitute for Water Research, Rhodes University, Grahamstown 6140, South Africarei: +27 (0)466224014 ; Fax: +27 (0)466229427email: d.hughes@ru.ac.za

Hydrology in a Changing World:Environmental and HumanDimensions

Recent IAHS Publications

Evolving Water Resources Systems:Understanding, Predicting andManaging Water-Society InteractionsEditors: Aftilio Castel/ann, Serena Ceole, Elena Toth &Alberto Montanan

A selection of 89 peer-reviewed papers by research groupsactive in five continents addressing topics associated withwater resources assessment and management in achanging environment, and particularly the two-wayinteraction between water and society, the focus of thePanta Rhei IAHS research initiative. Water resourcessystems, catchment hydrology, eco-hydrology, groundwaterhydrology, water security and socio-hydrology are drivingscientific areas, which are covered.

Publ.364 2014 978-1-907161-42-1 548+ xii pp. £118

Considering Hydrological Change inReservoir Planning and ManagementEditor: Andreas SchumannCo-editors: Vladimir Belyaev, Emna Gargouti, George Kuczera,GiI Mahe & Stephen Mal/ory

An excellent overview of contemporary problems inreservoir management, from planning aspects of largemulti-objective reservoirs and of small farm dams in Africa,to governmental matters, to sedimentation issues, toclimate change impacts. Given the stochastic nature ofhydrological conditions, the limited information available tocharacterize it and the multi-faceted targets of reservoirmanailement, reservoir planning and operation areambitious challenges for hydrologists and water managers.

Publ.362 2013 978-1-907161-40-7 214+ x pp. £61

Understanding Freshwater QualityProblems in a Changing WorldEdited by Bent ArlleimerC04Jdited by Adtian CoIfins, Valentina Krysanova,Elango Lakshmanan, Michel Meybeck & Mike StoneContributions from each continent provide a review of waterquality problems worldwide, with articles describing presentregional/local freshwater quality status and highligfltlngresearch needs. How the situation may develop into tilefuture, given onijoing changes in environment and SOCiety,is discussed. Questions addressed include: How tounderstand the behaviours of changing hydrologicalsystems and impacts on freshwater quality? How toeffectively bring together theoretical and experimentalhydrology, ana new measurement techniques to advanceknowledge of water quality processes for the future?

Pub!. 361 2013 978-1-907161-39-1 372+ xii pp. £87

Deltas: Landforms, Ecosystems andHuman ActivitiesEditors Gordon Young &Gerardo M. E. Penlk:Associate Edrlors H. Aksoy, J. Bogen, AI. Gelfan, G. MaM,P. Marsh &H. SavenijeDeltas pose great challenges, whether marine or lacustrine,as re\lions ofpurely natural conditions or of intense humanactivil\' set in the context of complex and often rapidlychanging natural environments. Physically they are complexsystems, the end-products of catchment processesinvolving water supply, sediment delivery and water quality- elements that are fast changing over time as a result bothof human influences and change in climatic drivers. Thecontributions result from a joint symposium of IAHS) andlAPSO.

Publ.358 2013 978-1-907161-36-0 246+ x pp. £65

Floods: From Risk to OpportunityEdited by A. Chavoshian & K. TakeuchiCo-edited by X. Cheng, E. Plate, S. Simonovic, S. Uhlenbrook & N.Wtight

A paradigm shift from focusing on emergency response andrecovery to flood risk management is required to build thecapacity necessary to cope with floods. The process shouldbe supported by vulnerability monitoring and developmentof tools such as standard measures of risk and prepared-ness in an integrated approach to improve capacity to dealwith floods, taking advantage of their benefits whileminimizing their social, economic and environmental risks.The papers, selected from ICFM5, cover flood riskmanagement, disaster management, forecasting and earlywarning, and management in different regions.

Pub'.357 2013 978-1-907161-35-3 470+ x pp. £96

BENCHMARK PAPERS IN HYDROLOGYThe IAHS Seties that collects together, by theme in stand-alonevolumes, the papers that provided the scientific foundations forhydrology.

HYDRo-GEOMORPHOLOGY, EROSION ANDSEDIMENTATIONMichael J. KirkbyProvides a systematic analysis of the relationships betweenhydrology arid geomorpholOllY with commentanes on thepapers that have been most Influential in developments atthe hydrology-geomorphology interface.

BM6 2011 978-1-907161-14-8 Hardback, 640 pp. £70

FORESTHYDROLOGYDavid R. DeWal/eThe papers selected include the early review of forest andwater by Zen (1927) and the Wagon Wheel Gap pairedwatershed study (Bates & Henry, 1928) report, but coversall aspects from evapotranspiration to water yields andquality, to soil erosion.

BM7 2011 978-1-907161-17-9 Hardback, 474 pp. £65

ISOTOPE HYDROLOGY

P. K. AggafWa~ K. O. Fr6h/ich, J. R. Gat& R. GonfiantiniThe potential of using stable isotopes of water wasrecognized in the 19305, but not fully explored until the195

Hydrology in a Changing World:Environmental and HumanDimensions

Editor in Chief:

TREVOR M. DANIELLSchool ofCivil. Environmental and Mining Engineering.University ofAdelaide, Australia

Co-Editors:

HENNY A. J. VAN LANEN, SIEGFRIED DEMUTH, GREGOR LAAHAERIC SERVAT, GIL MAHE, JEAN-FRANCOIS BOYER,JEAN-EMMANUEL PATUREL, ALAIN DEZETTER& DENIS RUELLAND

Proceedings of FRIEND-Water 2014, Montpellier, France, October 2014

IAHS Publication 363inthe IAHS Series ofProceedings and Reports

Published by the International Association of Hydrological Sciences 2014

IAHS Publication 363

ISBN 978-1-907161-41-4

British Library Cataloguing-in-Publication Data.A catalogue record for this book is available from the British Library.

© IAHS Press 2014This publication may be reproduced as hard copy, in whole or in part, for educational or nonprofit use,without special permission from the copyright holder, provided acknowledgement of the source is made. Nopart of this publication may be electronically reproduced, transmitted or stored in a retrieval system, and nouse of this publication may be made for electronic publishing, resale or other commercial purposes withoutspecific written permission from IAHS Press.

The papers included in this volume have been reviewed and some were extensively revised by the Editors,in collaboration with the authors, prior to publication.

IAHS is indebted to the employers of the Editors for the invaluable support and services provided thatenabled them to carry out their task effectively and efficiently.

The information, data and formulae provided in this volume are reproduced by IAHS Press in good faith and as finallychecked by the author(s); IAHS Press does not guarantee their accuracy, completeness, or fitness for a given purpose.The reader is responsible for taking appropriate professional advice on any hydrological project and IAHS Press doesnot accept responsibility for the reader's use of the content of this volume. To the fullest extent permitted by theapplicable law, IAHS Press shall not be liable for any damages arising out of the use of, or inability to use, the content.

The designations employed and the presentation of material throughout the publication do not imply the expression ofany opinion whatsoever on the part of lAHS concerning the legal status of any country, territory, city or area or of itsauthorities, or concerning the delimitation of its frontiers or boundaries.

The use of trade, firm, or corporate names in the publication is for the information and convenience of the reader. Suchuse does not constitute an official endorsement or approval by lAHS of any product or service to the exclusion ofothers that may be suitable.

Publications in the series of Proceedings and Reportsare availablefrom:IAH5 Press, Centre for Ecology and Hydrology, WaJllngford, Oxfordshire OX10 8BB, UKtel.: +44 1491 692442; tax: +441491 692448; e-ma~: jllly@iahs.demon.co.uk

Printedby BerfortsInformation Press

Cover image: Moroccan landscape (photograph by EricServat).

Preface

The FRIEND-Water (Flow Regimes from International and Experimental Network Data)programme is an international collaborative research initiative under the umbrella ofUNESCO's International Hydrological Programme (IHP) and has disseminated researchresults by convening international scientific conferences every four years since 1989. Thispublication is the proceedings of reviewed papers of the 7th World FRIEND Conference heldin Montpellier, France, 7-10 October 2014 on the theme of Hydrology in a Changing World:Environmental and Human Dimensions. FRIEND-Water aims to facilitate understanding abouthow climate, river basin and human factors influence the spatial and temporal distribution ofwater.

A large number of universities, research institutes and water-related agencies in over 150countries around the world have actively participated in FRIEND-Water. It is a worldwideprogramme, not only with respect to its geographical coverage, but also with respect to theissues that are investigated.

The response to the "call for papers" for this conference generated 230 abstracts, which maywell provide an indication of the increasing significance of water issues in the world and theFRIEND-Water and IHP scientific communities. The papers share knowledge on changes inhydrological processes and their impacts, including ecological flows, erosion andsedimentation, and the development of adaptable water management and water policies toaccount for these impacts, as well as changes in the frequency and variability of floods anddroughts due to a variety of factors. Six main topics were addressed: hydrological databases;trends in hydrological regimes and extremes; changes in ecological flows and coastalecohydrology; erosion and sediment transport processes and trends; regional observationalhydrological modelling frameworks; and water resources and prospective scenarios.

Editor-in-ChiefTrevor M. DanielI

School ofCivil, Environmental and Mining EngineeringFaculty 0/Engineering. Computer and Mathematical Sciences

University 0/Adelaide, Adelaide, 5005. Australia

Acknowledgements The conference organizers would like to thank the Japanese-Funds-In-Trust, the Institute of Research for Development (IRD, France), the German IHP Committee,the Laboratory HydroSciences Montpellier and the Montpellier Institute for Water andEnvironment for their financial support to the Conference. We would also like to thank theUniversity of Wageningen (the Netherlands) for financial support, which was provided todisseminate the EU project DROUGHT-R&SPI (Fostering European Drought Research andScience-Policy Interfacing). A special thank goes to Nicole Couesnon, Sandra Ardoin-Bardin,Jean-Francois Boyer and Alain Dezetter from the Local organisation Committee without theireffort the conference would not be possible. Finally the organizers want to thank Penny Perrinsand Cate Gardner from IAHS who deserve special mention in controlling the final editing ofthe publication of the Red Book.

Co-editorsHennyA.J. van LanenHydrology and Quantitative Water Management Group, Wageningen University, Droevendaalsesteeg 3, 6708 PBWageningen, the Netherlands

SiegfriedDemuthHydrological Systems & Water Scarcity Section. Division of Water SciencesNatural Sciences Sector. UNESCO, I rue Miollis, 75732 Paris cedex 15, SP France

Gregor LaahaInstitute ofApplied Statistics and ComputingUniversity of Natural Resources and Life Sciences. BOKU ViennaPeter Jordan-Strafie 82, A-1190 Vienna. Austria

Eric ServatIRD, UMR HydniSciences Montpellier (HSM, CNRS. IRD, UMI, UM2). Universite Montpellier 2CC MSE. 34095 Montpellier Cedex 5, France

Oil MaheIRD, UMR Hydro Sciences Montpellier (HSM, CNRS, IRD, UMI, UM2).15 rue Abou Derr BP 8967. 10000 Rabat-Agdal, Maroc

Jean-Francois BoyerIRD, UMR HydrcSciences Montpellier (HSM, CNRS. IRD, UMI, UM2), Universite Montpellier 2CC MSE. 34095 Montpellier Cedex 5, France

Jean-Emmanue1 Pature1IRD, UMR HydroSciences Montpellier (HSM, CNRS, IRD, UMI. UM2).01 BP 182, Ouagadougou 01, Burkina Faso

Alain DezetterIRD, UMR Hydro Sciences Montpellier (HSM, CNRS.IRD, UMI. UM2), Universite Montpellier 2CC MSE, 34095 Montpellier Cedex 5, France

Denis RuellandCNRS, UMR HydrtiSciences Montpellier (HSM, CNRS. IRD, UMI, UM2), Universite Montpellier 2CC MSE. 34095 Montpellier Cedex 5, France

Contents

Preface by Trevor M Daniell

Over 25 years of FRIEND-Water: an overview Henny A. J. Van Lanen,Siegfried Demuth, Trevor Daniell, David M Hannah, Gregor Laaha, Gil Mahe& Lena M Tallaksen

Trends in hydrological regimes and extremes

Regional assessment of low flow processes and prediction methods acrossEuropean regimes Gregor Laaha, Anne F. van Loon, Cla ire Lang Delus &Daniel Koffler

Progressive aridity impact on the hydrological regime in the Volta River basin inBenin (West Africa) Henri S. V. Totin, Arnaud Zannou, Ernest Amoussou,Abel Afouda& Michel Boko

Multi-annual droughts in the water-stressed English Lowlands: long-termvariability and climate drivers Jamie Hannaford, Chris K. Folland,Mike C. Kendon, & John P. Bloomfield

Synthesizing changes in low flows from observations and models across scalesKerstin Stahl, Jean-Philippe Vidal, Jamie Hannaford, Christel Prudhomme,Gregor Laaha & Lena M Tallaksen

Understanding and adapting to flood risk in a variable and changing climateAnthony S. Kiem

Trends in floods in small Norwegian catchments - instantaneous vs daily peaksDonna Wilson, Hege Hisdal & Deborah Lawrence

Climate change impact on flood generation process Biljana Radojevic,Pascal Breil & Dejan Dimkic

Flood duration frequency analysis in a changing climate: the methodologyapplied to Fengle River (Yangtze basin, China) C. Salles, Y. Chu, J. L. Perrin,M G. Tournoud, L. Boudet, F. N. Cres, C. Rodier, S. Zheng, L. Huang & Y. Ma

Future low flows and hydrological drought: how certain are these for Europe?Marcel A. A. Alderlieste, Henny A. J. Van Lanen & Niko Wanders

Do large-scale models capture reported drought events? Marjolein H. J. VanHuijgevoort, Henny A. J. Van Lanen, Adriaan J. Teuling & Remko Uijlenhoet

Analyse des periodes seches dans le bassin d 'lchkeul (Tunisie) apartir dedonnees journalieres de pluviometrie / Analysis of dry spells in the Ichkeul basin(Tunisia) using daily rainfall data Majid Mathlouthi & Fethi Lebdi

v

11

17

23

30

36

42

48

54

60

66

72

viii

Changes of drought characteristics in small Czech and Slovakian catchmentsprojected by the CMIPS GCM ensemble Martin Hanel, Adam Vizina,Marta Martinkova, Stanislav Horacek, Diana Porubska, Marian Fendek &Miriam Fendekova

Impact of land-use change on the hydrology of North Lao PDR watershedsEmmanuel Mouche, Francois Moussu, Claude Miigler, Olivier Ribolzi,Christian Valentin, Oloth Sengtahevanghoung & Guillaume Lacombe

Assessment of wildfire impact on hydrological extremes in eastern SiberiaLyudmila Lebedeva & Olga Semenova & Nina Volkova

The correlation between variations of climatic factors and zonal runoff ofmountain rivers Mikhail Bolgov & Marina Trubetskova

Linking variations in large-scale climatic circulation and high groundwater levelsin southern England David A. Lavers, David M Hannah & Chris Bradley

Rainfall regime evolution and drought forecasting in eastern AlgeriaMohamed Meddi & Abdelwahab Rahmouni

Up-to-date climate forced seasonal flood changes (the case study for theEuropean part of Russia) Natalia Frolova, Maria Kireeva, Dmitry Nesterenko,Svetlana Agafonova & Pavel Tersky

Improvement of reservoir operation by hybrid optimization algorithm: case studyof HuongDien Reservoir, Vietnam Van Hoa Ho & Joong Hoon Kim

Optimal reservoir operation strategy for balancing ecosystem and human needsFi-John Chang, Wen-Ping Tsai, Yu-Chung Wang, Pin-An Chen, Li-Chiu Chang,Alexandra Coynel & Georges Vachaud

Recent trends in monthly temperature and precipitation patterns in Europelrene B. Nilsen, Anne K. Fleig, Lena M Tallaksen & Hege Hisdal

Large-scale climate control on lake inflow in the Waitaki basin, New ZealandDaniel G. Kingston, Clare S. Webster & Pascal Sirguey

Sensitivity analysis of hydrological modelling to climate forcing in a semi-aridmountainous catchment D. Ruelland, A. Dezetter & P. Hublart

Prediction of water resources in the Chao Phraya River Basin, ThailandSupattana Wichakul, Yasuto Tachikawa, Michiharu Shiiba & Kazuaki Yorozu

Modelling runoff and its components in Himalayan basins Hong Li, SteinBeldring, Chong-Yu Xu & Sharad K. Jain

Statistical analysis of river discharge projected using the MRI-AGCM3.2Sdataset in Indochina Peninsula D. T. Duong, Y. Tachikawa, M Shiiba &K. Yorozu

78

84

90

96

102

107

113

119

126

132

138

145

151

158

165

Extreme precipitation events and related impacts in western IberiaMargarida L. R. Liberato & Ricardo M Trigo

Unidirectional trends in rainfall and temperature of BangladeshShamsuddin Shahid, Xiao-Jun Wang & Sobri B. Harun

2 Water resources and prospective scenarios

ix

171

177

Assessing past and future water demands under climate change and 185anthropogenic pressures on two Mediterranean basins B. Grouillet, J. Fabre,D. Ruelland, L. Collet & J-F. Boyer

A multi-tracer approach to understand the hydrogeochemical functioning of a 191coastal aquifer located in NE Tunisia Fethi Lachaal, Anis Chekirbane,Ammar Mlayah, Bilel Hjiri & Jamila Tarhouni

L'amenagement des barrages de retenue d'eau, une strategic d'adaptation it la 197secheresse dans le bassin du fleuve Niger? / Is the building ofdams an adaptationstrategy to fight against drought in the basin of the River Niger? G. Lienou,G. Mahe, S. L. Piih, D. Sighomnou, J. E. Paturel & F. Bamba

Assessing the long-term evolution of water supply capacity: comparison of two 203Mediterranean catchments J. Fabre, L. Collet, M Milano, D. Ruelland,A. Dezetter, S. Ardoin-Bardin & E. Servat

Assessing the capacity to meet irrigation water needs for viticulture under climate 209variability in the Chilean Andes P. Hublart, D. Ruelland, A. Dezetter &H. Jourde

Present-day surface and subsurface water resources of European Russia: 2 I5conditions, use and forecast R. Dzhamalov, N. Frolova, M Kireeva & E. Rets

La prospective territoriale, un outil de reflex ion sur la gestion de I'eau du bassin 22 Ide la Durance en 2050/ The territorial foresight, a reflection framework for watermanagement planning in the Durance River basin in 2050 R. Samie, C. Monteil,Y. Arama, H. Bouscasse & E. Sauquet

Modifications des temperatures et des precipitations sur le Bani, un sous-bassin 228du fleuve Niger / Changes in temperature and precipitation in the Bani watershed,a sub-basin of the Niger River Jean-Emmanuel Paturel & Malicki Zorom

Climate change and debris flow: hazards maps in Matucana village Peru under 234IPCC scenarios Juan W. Cabrera Cabrera & Leonardo F. Castillo Navarro

A reconnaissance study of water and carbon fluxes in a tropical watershed of 239Peninsular Malaysia: stable isotope constraints Muhammad I. Syakir,Kern Y. Lee, fan D. Clark & Jan Veizer

Characterizing infiltration of a tropical watershed from ALOSP ALSAR data 245using the Green-Ampt infiltration model Nor Liyana Mohammad Khan,Muhamad Askari & Ab. Latif fbrahim

x

Projected climate change impacts on water resources in northern Morocco with 250an ensemble of regional climate models Y. Tramblay, D. Ruelland, R. Bouaicha& E. Servat

Modele conceptuel de l'aquifere du Continental Terminal d'Abidjan / Conceptual 256model of Abidjan Continental Terminal aquifer Auguste K. Kouassi,Francis W Kouassi, Jules 0. M Mangoua & Issiaka Savane

Characterization and evaluation of the confined limestone aquifer in Kuwait 263Fawzia M Al-Ruwaih

Etude geochimique et isotopique d'un aquifere phreatique cotier anthropise: 269Nappe de Oussja-Ghar El Melah (Tunisie) / Geochemical and isotope study ofcoastal anthropogenic phreatic groundwater: Oussja-Ghar El Melah aquifer(Tunisia) Safouan Ben Ammar, Jean-Denis Taupin, Kamel Zouari,Mohamed Khouatmia & Mohamed Ben Assi

Choosing metrics that matter - quantifying performance to help address reservoir 276operation challenges in Kenya's Tana basin Anthony P. Hurford &Julien J Harou

Environmental and geomatics assessment of the Ha Thanh basin to hydrological 282hazards in central Vietnam Stephane Grivel, Anh Tu Ngo, Jean-Marc Zaninetti &Thi Van Luong

Forecasting and mitigation of flooding in a Mediterranean karstic watershed 288V. Borrell-Estupina, B. Ladouche, r.o. Malaterre, S. Ricci, Y. Caballero,M Coustau, N. Dorfliger, P. Fleury, M Jay-Allemand, .L-C. Marechal &0. Thual

3 Erosion and sediment transfer processes and trends

Using IJ7Cs and 210Pbex measurements to document erosion rates for different 297time windows in a small catchment in southern Italy Paolo Porto &Desmond E. Walling

Lessons from long-term monitoring of soil erosion in three southeast Asian 303agricultural catchments undergoing rapid land-use changes C. Valentin,A. Boonsaner, J L. Janeau, P. Jouquet, T. Henry des Tureaux, S. Huon,K. Latsachack, Y. Le Troquer, J L. Maeght, D. Orange. Pham Dinh Rinh,A. Pierret, P. Podwojewski, 0. Ribolzi, A. de Rouw, 0. Sengtaheuanghoung,N. Si/vera, H. Robain, B. Soulileuth, W Thothong, Tran Due Toan &Tran SyHai

Study of the relationship between sediment transport and rainfall extremes in the 309watershed of the wadi Mina (northwest Algeria) F. Hallouz, M Meddi &G. Mahe

"Blue-green" corridors as a tool for erosion and stream control in highlyurbanized areas - case study of Belgrade city Ratko Ristic, Boris Radic,Goran Trivan & Ivan Malusevic

4 Hydrologicaldatabases: how to cope with future questions

xi

315

Ensuring hydrometric data are fit-for-purpose through a national Service Level 323Agreement Katie Muchan & Harry Dixon

Future Flows: a dataset of climate, river flow and groundwater levels for climate 330change impact studies in Great Britain C. Prudhomme, C. Jackson, T. Haxton,S. Crooks, S. Dadson, D. Morris, 1. Williamson, A. Barkwith, 1. Kelvin,1. Mackay, L. Wang, G. Goodsell, L. Boelee, H. Davies, G. Buys, T. Lafon,A. Young & G. Watts

Architecture of environmental information systems applied to scientific 336observatories: examples of Carnoules and MEDYCYSS observatories1. F. Boyer, E. Servat, C. Casiot & H. Jourde

Cooperative WebGIS interactive information systems for water resources data 342management Charalampos Skoulikaris, Jacques Ganoulis, Nikolaos Karapetsas,Fotis Katsogiannos & Georgios Zalidis

Towards a pan-European assessment oflow flow indices Claire Lang Delus, 348Gregor Laaha, Daniel Koffler, Kerstin Stahl, Hege Hisdal, Christel Prudhomme,Miriam Fendekova & Wojciech Jakubowski

Selecting an optimal climatic dataset for integrated modelling of the Ebro 355hydrosystem A. Dezetter, J. Fabre, D. Ruelland & E. Servat

Growing season length and rainfall extremes analysis in Malawi 361Cosmo Ngongondo, Lena M Tallaksen & Chong-Yu Xu

Standardized precipitation-evapotranspiration index (SPEI): Sensitivity to 367potential evapotranspiration model and parameters James H. Stagge,Lena M Tallaksen, Chong- Yu Xu & Henny A. 1. Van Lanen

Minimizing "geopolitically ungauged" catchment area oftransboundary river 374basins to support disaster risk reduction Kelly M Kibler

Probability distribution of rainfall in the Bia watershed: contribution ofMarkov 379chains Meledje N'diaye Hermann, Kouassi Kouakou Lazare, Ngo YaoAlexis &Savane Issiaka

Assessment of rainfall observed by weather radar and its effect on hydrological 386simulation performance Nghi Vu Van, Nhung Tran Thi & Quyet Le Dinh

Predicting soil water retention characteristics for Vietnam Mekong Delta soils 392Nguyen M Phuong, Le V. Khoa & Wim Cornelis

xii

5 Regional observational-hydrological modelling frameworks

Connecting streamflow and atmospheric conditions in Europe: state-of-the-art 401review and future directions David M Hannah, Anne K. Fleig,Daniel G. Kingston, James H. Stagge & Donna Wilson

Water budget in the Amazon basin and impacts on flood modelling 407Augusto Getirana, Sujay Kumar, Christa Peters-Lidard & Kristi Arsenault

Measurement, modelling and analysis of hydrological and hydrogeological 413processes and trends in a marsh area Branka Bracic Zeleznik & Lidija Globevnik

From drought to floods in 2012: operations and early warning services in the UK 419Charlie Pilling, David Price, Adrian Wynn, Andrew Lane, Steven J. Cole,Robert J. Moore & Timothy Aldridge

Les lacs collinaires en Tunisie: technique de conservation des eaux et du soli 425Small lakes in Tunisia: water and soil conservation technique Hamadi Habaieb,Taoujik Hermassi, Slaheddine Ghedaoui & Issam Anatar

Application of a hydrological model to evaluate the potential hydro energy in a 431mountainous small river basin of Japan Kazumasa Fujimura,Masahiro Murakami, Yoshihiko Iseri & Shinjiro Kanae

Influence des activites anthropiques sur le regime hydrologique du fleuve Logone 438de 1960 it 2000 I Human activities influence on the hydrological regime of theLogone River basin from 1960 to 2000 N. Tel/ro Wai; B. Ngounou Ngatcha,G. Mahe, J. C. Doumnang, F. Delclaux, N. Goundoul & P. Genthon

Estimation of runoff curve numbers using a physically-based approach of 443preferential flow modelling Pori Das, Aminul Islam, Subashisa Dutta,Amit K. Dubey & Rupak Sarkar

6 Changes in ecological flows and coastalecohydrology

Hydrological regime of a tidal system in the Red River Delta, northern Vietnam 451Luu Thi Nguyet Minh, Didier Orange, Tran Hong Thai, Josette Gamier,Le Lan Anh & Trinh Anh Due

Measuring impact of water management on ecological health of a river: Poudre 457River, Colorado, USA Robert T. Milhous

Impact of climate change on aquatic ecosystems along the Asse River network 463Thomas Cipriani, Francois Tilmant, Flora Branger, Eric Sauquet &Thibault Datry

Ecological flow for integrated planning of small hydropower plants: a case study 469from Greece Thomas Patsialis, Charalampos Skoulikaris & Jacques Ganoulis

Key word index 475

Hydrology in a Changing World: Environmental and Human DimensionsProceedings of FRIEND-Water 2014,Montpellier, France, October2014 (IAHSPub!.363,2014).

Over 25 years of FRIEND-Water: an overview

HENNY A. J. VAN LANENt, SIEGFRIED DEMUTH2, TREVOR DANIELL\DAVID M. HANNAH4, GREGOR LAAHA5, GIL MAHE6 &LENA M. TALLAKSEN7I Hydrology and Quantitative Water Management, Wageningen University, Droevendaalsesteeg 3,6708 PB Wageningen, the Netherlandshenny. vanlanen@wur.nl2 Hydrolorical Systems & Global Change Section, Division of Water Sciences, Natural Sciences Sector, UNESCO, I rueMiollis, 7 732 Paris cedex 15, SP France3 School ofCivil, Environmental and Mining Engineering, Faculty ofEngineering,Computer and Mathematical Sciences, University ofAdelaide, Adelaide 5005 Australia4 School ofGeography, Earth & Environmental Sciences, University ofBirmingham, Birmingham BI5 2TT. UK5 Institute ofApplied Statistics and Computing, University ofNatural Resources and Life Sciences, BOKU Vienna,Peter Jordan-Strafle 82, A-1190 Vienna, Austria6IRD/HSM, 15 rue Abou Derr BP 8967, 10000 Rabat-Agdal, Maroc7 Department ofGeosciences, University ofOslo, P'OiBox 1047, Blindern, NO-0316 Oslo, Norway

Abstract FRIEND-Water has been a flagship programme of the UNESCO-International HydrologicalProgramme (IHP) for more than 25 years. This network has delivered outstanding scientific value (researchpapers, conferences) and made significant contributions to international cooperation in the water sciencesand associated water resources management, an objective at the core of the IHP and UNESCO mandate. Anoverview of major activities and outcomes from this cross-cutting programme is provided, includingdatabase progress, enhanced scientific understanding of hydrological processes across scales, developmentof analytical tools, education, capacity building, dissemination and cooperation with other networks. Theprogramme is not only a network of scientists; it is also aims to be a network of several databases andinstitutions. Ongoing and future activitiesof the FRIEND-Waternetwork are also addressed.Key words regional hydrology; flow regimes; hydro-extremes; water resources; databases; education;capacity building; dissemination; international cooperation

BACKGROUND

In 1985 a research programme was initiated as a follow-up of the UNESCO InternationalHydrological Decade (IHD) 1965-1974. During the rHO, a number of research basins wereestablished around the world and a variety of scientific studies were conducted. However, only a fewstudies worked across national boundaries (i.e. at the regional scale). This apparent lack ofinternational comparative analysis was the impetus for establishing the FRIEND-Water (FlowRegimes from International and Experimental Network Data) programme, as an internationalcollaborative research initiative under the umbrella of UNESCO's International HydrologicalProgramme (IHP). FRIEND-Water aims to develop a better understanding of hydrological variabilityand similarity across scales through exchange of data, knowledge and techniques. This has been acontinuous effort over more than 25 years. Sharing of data, models and research tools betweencountries, organizations and researchers has always been a high priority for the FRIEND-Waterprogramme since it helps to advance science and to bridge the gap between research and operationalwater management. A large number of universities, research institutes and water-related agencies inover 150 countries around the world have actively participated. It became a worldwide programme,not only with respect to its geographical coverage, but also due to the vast number of well-recognized publications in international journals, its attractive training courses for students and watermanagers around the world, and by offering an international network for young researchers.

The current FRIEND-Water programme aims at improving water science and sustainable useof current and future water resources. The programme achieves these goals through:(a) collecting and exchanging environmental (mainly river flow) data, particularly in an

international context;(b) enhancing scientific understanding of hydrological processes across scales from local up to

global;

Copyright © 2014 IAHS Press

2 Henny A. J.Van Lanen et al.

(c) developing innovative analytical tools for further improvement of water resourcemanagement and reduction of hydrohazards' risk (i.e. floods and droughts);

(d) educating and developing capacity building pathways through PhD and MSc courses andtechnical training courses; and

(e) disseminating knowledge through journal publications, books and manuals, conferences,technical workshops and web-based platforms, to cooperate with other international networksand professional organizations, and recently to be active in science-policy-making.

With eight regional groups (Fig. I), FRIEND-Water reflects the best in internationalcooperation between scientists, water managers, stakeholders and policy makers.

The following section highlights some major achievements of the FRIEND-Waterprogramme, before the last section concludes with some on-going and future activities, which onlycan be achieved through a well-linked international science network.

MAJOR ACHIEVEMENTS

Databases

The FRIEND-Water programme has collated river flow data for different regions of the world. Inaddition to flows, an inventory of river basin properties was compiled to provide thematic andmetadata to aid interpretation of hydrological response and spatial variability. The regionaldatabases have grown over the years and are updated regularly to meet modem research needs.Hannah et at. (20 II *) I highlighted the importance of large-scale river flow archives for evidence-based assessment of past hydrological variability, and for supporting hydrological modelling offuture changes. They highlighted the FRIEND-European Water Archive (EWA) as an example ofa valuable transnational data resource for basic and applied hydrological research.

Increased scientific understanding

This section provides examples of increased scientific understanding, without attempting to beinclusive. Van Lanen & Demuth (2014) give a comprehensive overview of FRIEND-Waterpublications from the different regional groups which have been published in the period 2010 to2013, and it provides reference to FRIEND-Water publications from previous reporting periods.

Hydrohazards A key research topic of the FRIEND-Water programme is hydrohazards: bothfloods and drought. The impressive progress in the flood field has recently been synthesised bySimonovic (2013*) and others in a collection of four books on flood disaster management theoryand practice within the context of anthropogenic climate change. The books include: (a) floodforecasting and hydrological processes linked to heavy rainfall, (b) stochastic modelling of heavyprecipitation for runoff forecasting, (c) tools for warning and forecasting in ungauged basins,including hydrological and inundation modelling, (d) enhanced understanding, forecasting andmanagement of flash floods in urban areas, and (e) development of methodologies for reliableinundation mapping, and development of flood risk and vulnerability maps.

AP-FRIEND has conducted a number of workshops bringing together countries across theregion to address Rainfall Intensity Duration Frequency (IDF) analysis for the Asia Pacific Region(Daniell & Tabios, 2008*) and conversion of these values using flood estimation and floodhydrograph procedures (Daniell, 20 II *) as this was seen to be important for lessening damagefrom disasters by relevant design of infrastructure for floods.

The FRIEND-Water program has been performing pioneering work on low flows and droughthydrology. The core focus on data networks of FRIEND-Water laid the foundation for the firstregional low flow studies yielding enhanced understanding of low flow generating processes,

I References marked with an asterisk (*) have been included in the FRIEND-Water overview report (VanLanen & Demuth, 2014), and are not listed under References in this paper. Full description of all referencesin this paper is also available at: http://www.wageningenur.nllenJShow/References-IAHS2014-Paper.htm.

•

o(§..,N01

~i:iQ-.

I~

Q

(§

~ .:;;:

NO

[UIlO FRIEND I "'LOI RItNO

• £lJAOP{AN Flll(l'!l lf URO FRlENO)

• ",rDlTTRRANU~ FRlEf'OlM{O FRlENO)

• ~Sl"" rACJFK:FI;1FND(APFRlfNOJ

• LAlINAMERIC~N & CAAIBBEAN FlllEIlOWNGO FRIIJIOJ

• TtlE"'t.'OU· ~VSH HI.... LAy. S FRlf'NO(HKHrRl£NO\

• FRl£UONIU If NI

)•

t..I ~

~!;'~\. ~

.:

~

I - -:

~ ~... ->~ ~~ '.

;

Fig. 1 FRIEND-Water Regional Grou ps and their geograph ica l co ver age.

w

4 Henny A. J.Van Lanen et al.

catchment similarity, and estimation procedures on a regional scale, long before being on theagenda of any other international activity. Through this core expertise, the FRIEND-Waternetwork significantly contributed to the success of: (a) the WMO Manual on Low-flow Estimationand Prediction (Gustard & Demuth, 2009), and (b) the recent IAHS decade to assess current state-of-the-art of low flow prediction in ungauged basins (PUB, Laaha et al., 2013*). Later, pioneeringwork has been done on drought hydrology. These recent achievements have extended the earlieroverview work of Smakthin (2001 *) and show that low flow modelling still needs improvement(e.g. Staudinger et al., 2011 *). The FRIEND-Water network initiated research on: (a) definitionand development of drought identification methods (e.g. Fleig et al., 2006*; Van Huijgevoort etal., 2012*), (b) streamflow deficiencies and circulation patterns (e.g. Stahl, 2001 *; Fleig et al.,2011 *), (c) propagation of meteorological drought into hydrological drought, and the associateddevelopment of a process-based drought typology (e.g. Peters et al., 2003*; Van Loon & VanLanen, 2012*), (d) spatio-temporal drought development (e.g. Peters et al., 2005*; Tallaksen et al.,2009*, Hannaford et al., 2010*, Corzo Perez et al., 2011 *), and (e) trends in low flows anddrought (e.g. Hisdal et al., 2001 *; Hannaford & Marsh, 2008; Stahl et al., 2010*; Wilson et al.,2010*; Stahl et al., 2012*; Hannaford et al., 2013*). This vast scientific achievement also led to abetter assessment as to what extent global models used for climate change assessment capturedrought (Prudhomme et al., 2011 *; Stahl et al., 2011; Gudmundsson et al., 2012a* ,b*; Van Loonet al., 2012*; Van Huijgevoortet al., 2013*).

Large-scale hydrology Recent FRIEND-Water research has made significant contributions toadvancing knowledge of: (a) flow regimes and climate-hydrology interactions across Europe (e.g.Lavers et al., 2013*), the North Atlantic (e.g. Kingston et al., 2011 *), UK (Lavers et al., 2010*),and other parts of the world such as the Mediterranean (e.g. Saris et al., 2010*; Milano et al.,2012*) and West and Central Africa (e.g. Ardoin-Bardin et al., 2009*), (b) the role of river basinproperties in modifying hydrological response (e.g. Laize & Hannah, 2010*), and (c) linksbetween large-scale climate patterns and hydrological extremes (e.g. Kingston et al., 2013*). Webelieve the breadth of these contributions demonstrate a move towards a more interdisciplinaryand united approach (including combining empirically-based and modelling studies) tounderstanding hydrology at large-scales that will underpin major future research advances (Cloke& Hannah, 2011 *).

Semi-arid and coastal hydrology Semi-arid areas in Africa faced both a rainfall decrease andan increase of population and associated agricultural activities, which led to a rapid degradation ofland cover. The regional FRIEND-Water groups in these dry regions (AOC FRIEND and MEDFRIEND, Fig. I) study erosion and sediment transport as they are major issues for development.Most sediments are trapped into dams and lakes and thus do not reach the sea, moreover manycontrolled rivers have reduced outflows. This has great impact on coastal ecohydrology, and inmany places, including RAMSAR sites, changes in water quality and in coastal geomorphologyare observed, endangering the sustainability of environmental resources and ecosystems, andincreasing their vulnerability to sea level rise. Research topics in the dry regions concentrate onerosion and sediment flux modelling, with an emphasis on multidisciplinary approaches, butimpact of extreme events, and coastal biogeochemical changes linked to human activities was alsostudied (Mahe et al., 2010).

Research has clearly benefited from scientists being a member of an international networksuch as FRIEND-Water. The participation of many FRIEND-Water programme members ininternationally-funded research projects (e.g. ARIDE, GRAPES, ASTHyDA, WATCH, SPLASH,XEROCHORE and DROUGHT-R&SPI) demonstrates the advantage of holding suchmembership.

Innovative analytical tools

Various analytical tools have been developed by the FRIEND-Water programme. For example, aCD with analysis tools (http://www.geo.uio.no/edc). including data, worked examples and guidedtours accompanied a textbook on Hydrological Drought (Tallaksen & Van Lanen, 2004*).

Over 25 years of FRIEND-Water: an overview 5

Additional tools were made available by the Manual on Low-flow Estimation and through a web-based, user-friendly software package, Ifstat (http://cran.r-project.org/web/packages/lfstatJ). Themanual has a practical approach targeted to meet the needs of national hydrological andmeteorological services (Gustard & Demuth, 2009).

Education and capacity building

The FRIEND-Water programme developed comprehensive course material and has conductedover 30 courses on various topics to about 530 participants from> 100 countries (Table I).

Table 1 FRIENDtraining courses by topic and region, 1995-20I0120 12.

Region Number Numberof Numberof Topicsof courses countries participants

involvedEUROFRIEND 5MED FRIEND 2SA FRIEND 6HKH FRIEND 11

NILE FRIEND 3AP FRIEND 3

AMIGOFRIENDAOe FRIEND 2Total 33

35 106 Low flows and droughts11 39 Low flows, GIS8 90 Data, GIS, extremes, modelling10 177 Low flows, sediments,water quality, mass

balancemonitoringof glaciers, database10 30 Floods,droughts, modelling20 45 Floods,water resources management, low

flows12 20 Low flows

15 24 Dataanalysis, GIS121 531

FRIEND-Water activities include capacity-building through: (a) sponsored studentships (e.g.the Norwegian NUFU Water-Sciences project), and supervision of local postgraduate researchers.Capacity building is also facilitated through joint research projects between partners fromdeveloped and less developed nations; here the FRIEND data and research and training networksplay a key role. The Hindu Kush Himalayan (HKH) region, as an example, closely cooperatedwith CEH Wallingford on regional hydrology and options for hydropower generation. AGCFRIEND and MED FRIEND helped participants in Africa and the Mediterranean throughinternational courses and workshops, the use of analytical tools for data series analyses, such asMVR (regional vector method) and Khronostat (trends and ruptures analysis), developed by IRD.

Dissemination

FRIEND-Water has generated several hundred publications in peer-reviewed scientific journals(Van Lanen & Demuth, 2014). The network programme helped to disseminate the research resultsby convening well-known international scientific conferences every 4 years. It has organizedinternational workshops jointly with professional organisations, such as IAHS and held numeroussessions at the European Geophysical Union (EGU) assemblies. The meetings help to give theFRIEND-Water community the opportunity to exchange experience and knowledge beyond thescope of their own programme.

Through the virtual European Drought Centre (EDC, http://www.geo.uio.no/edc/). which aimsto promote collaboration and capacity building between the scientists and the user communities,FRIEND-Water participants contribute knowledge and skills on low flows and droughts.

There are growing efforts of the various regional FRIEND Water groups to feed scientificinformation into the design, implementation and review of water policies of national governments,

6 Henny A. J.Van Lanen et al.

and international policy frameworks (e.g. EU Water Framework Directive, WFD). A commonoutput from this science-policy interfacing is Science Policy Briefs".

Cooperation with other networks

The FRIEND-Water programme is linked to many international programs and professionalorganisations within the intemational hydrological community, and interacts with other IHPtechnical programmes (e.g. Intemational Flood Initiatives (IF!), International Drought Initiative(IDI), International Sediment Initiative (lSI». For instance, the programme helped IFI to promotethe paradigm shift from flood management as a strategy of defence, towards the transsectoral andtransdisciplinary approach of integrated flood management to maximize the long-term benefits offloods and to minimize the hardship, loss of life and damage to goods and assets that result fromfloods. FRIEND-Water has productive links with other UN organizations (e.g. WMO, UN-ISDR,WWAP) and professional scientific organizations (e.g. IAHS, EGU, ICSU). Close co-operationdeveloped, in particular with the World Meteorological Organization (WMO), through theCommission for Hydrology Programme on disaster mitigation on floods and droughts, theHydrology and Water Resources Programme (HRWP) and the World Climate ResearchProgramme (WCRP). In West and Central Africa the AOC FRIEND research community wasassociated with the international AMMA experiment on monsoon flow (2001-2010). In theMediterranean the new network of the lagoons of Maghreb (MAGHLAG), which is associatedwith the EUROMEDLAG network, has begun to collaborate with MED FRIEND.

ONGOING AND FUTURE ACTIVITIES

Databases Current activities concentrate on the harmonization of the databases andestablishing interfaces to provide easy access to the scientific community. The various regionalFRIEND-Water databases are planned to become an integral part of the Global Runoff DataCentre (GRDC) of WMO hosted by the Federal Institute of Hydrology in Koblenz, Germany.Different databases are being aligned (e.g. the FRIEND-Latin America and the CaribbeanDatabase system with the AOC FRIEND and the MED FRIEND database architecture and theEuropean Water Archive, EWA). An initiative has begun to yield a joint FRIEND Data Portal toprovide easy access for researchers to the different regional FRIEND-Water Archives.

Research The different groups, which may have a different research focus due to the specificchallenges of the regions (Table 2), will continue to enhance knowledge. Researcg will addressfuture challenges, in particular making use of the unique databases of streamflow observations andtheir potential for large-scale analysis of spatial and temporal variability, as well as for use inmodel validation studies. FRIEND-Water is identified in the UNESCO IHP-VIII (2014-2021)strategic plan "Water security: responses to local, regional and global challenges" as one of twocross-cutting programmes. It interacts with all IHP themes, such as water-related disasters andhydrological change, water scarcity and quality, ecohydrology, groundwater in a changingenvironment.

Education, capacity-building and dissemination These activities are intrinsic parts of theFRIEND-Water programme under the umbrella of UNESCO. It will contribute to: WaterEducation, key for water security (Theme 6, UNESCO IHP-VIII, 2014-2021).

FRIEND-Water with its extensive and huge network of researchers and its excellent scientificrecord and hydrological databases, is the most successful hydrological programme within theentire UN system. Despite all the constraints an international programme faces, FRIEND-Waterwill continue to create new grounds for international cooperation across political frontiers and be aplatform for young scientists to work across national boundaries. However, challenges still remain,

1 The XEROCHORE project had a major outcome a number of Science Policy Briefs in four languages;eachof themaddressed a certain WFDarticle (http://www.feem-project.net/xerochore/downloads.php).

Over 25 years of FRIEND-Water: an overview 7

for instance: (a) regular updating of databases, (b) secure research funding, particularly in lessdeveloped countries, and (c) revitalising FRIEND-Water in some regions because not all regionsare equally active.

Table 2 Main research themes of the regional FRIEND-Water groups (see Fig. I for acronyms)

Main research themes by regional EURO MED SA AOC NILE HKH APFRIEND groups

Database ./ ./ ./ ./ ./ ./ ./Low flows and droughts ./././././././Regime variability and large scale ././ ./ ./hydrological variationChange detection and attribution ./Techniques for extreme rainfall and ./ ./ ./ ./ ./ ./ ./flood runoff estimationRainfall-runoff modeling ./././././././Physical processes of streamflow ./generation in small basinsCatchment hydrological and ./biogeochemical process in a changingenvironmentKarstic hydrogeology ./Erosion and solid transport ././ ./Water quality ././Snow and glaciers ./Integrated catchment management./ ././Human influences./ ./Information management ./ ./Water resources assessment ././ ./ ./Coastal ecohydrology ./

AMI-GO

Acknowledgements The authors acknowledge the continuous support from UNESCO and theinput from FRIEND-Water Regional Coordinators: Yan Huang, Denis Hughes, Eduardo O. PianosGutierrez, Bamory Kamagate, Tarek Shawki and Luc Sigha Nkamdjou, the coordinators of theprojects in the regions and the many FRIEND-Water members across the world.

REFERENCESGustard, A. & Demuth, S. (ed.) (2009) Manual on Low-flow Estimation and Prediction. WMO - Operational Hydrology Report

No. 50/ WMO-No. 1029, Geneva, 136 p.Mahe, G., Baka1owicz, M., Boyer, J-F., Ferrari, E., Snoussi M. & Benina Touaibia (2010) MEDFRIEND: Global perspectives

for the UNESCO research network in hydrology for the Mediterranean. Secheresse 21(4), 285-293.Van Lanen, H.A.J. & Dernuth, S. (2014) FRIEND-Water: A Global Perspective 2010-2013. UNESCO Web report [will

become available at: http://www.unesco.org/new/en/natural-sciences/environment/water/ihp/ihp-programmes/frien

1 Trends in hydrologicalregimes and extremes

Hydrology in a Changing World: Environmental and Human DimensionsProceedings of FRIEND-Water 2014, Montpellier, France,October2014 (IAHSPub!.363,2014).

Regional assessment of low flow processes and predictionmethods across European regimes

GREGOR LAAHA t, ANNE F. VAN LOON2, CLAIRE LANG DELUS 3 &DANIEL KOFFLERt

11

1 University ofNatural Resources and Life Sciences, BOKU Vienna, Peter Jordan-StrafJe 82. A-1190 Vienna. Austriagregor.laaha@boku.ac.at2 Hydrology and Quantitative Water Management Group, Wageningen University. PO Box 47, 6700 AA, Wageningen,The Netherlands3 Departement ofGeography, University ofLorraine, France

Abstract In this contribution we present two pilot studies initiated from the EURO FRIEND-Water LowFlow group. The first study gives an example for bi-lateral assessment of regionalization methods forpredicting low flows at ungauged sites. The study covers the Meuse and Moselle basin in NE-France andinvestigates the performances of the geostatistical method top-kriging and a process-based method, thecatchment model GR4j, in different hydrological environments. The second study aims to explore low flowgenerating processes on a catchment scale by comparing runoff signatures on a regional scale. Based onhydrological drought types we use hot spots of close-by gauges to explore climate-eatchment interactionsbased on catchment similarity. The outcome shall build the foundation for a hydrological drought typologyacross European regimes. The examples illustrate that FRIEND-Water provides a very precious networkwhich facilitates international collaboration of water experts to tackle wickedwater problems.Key wordslow flows; droughts; regionalisation; top-kriging; drought typology

INTRODUCTION

Learning from similarity and differences of runoff regimes, to enhance our understanding ofdrought-generating processes and to develop more accurate regionalisation methods for predictinglow flows at ungauged sites, has always been a very active research field of the EURO FRIEND-Water Low Flow and Drought Group. Major achievements include the development ofregionalisation models and atlas products for several countries (Austria, Norway, SW-Germany,Switzerland, and the UK), but such models are still missing for other countries. Therefore, it isimportant to share and transfer the findings to other regions to fill in the picture on the spatialdimension of low flow and drought hazard across Europe.

One aim of the working group on regional assessment within the EURO FRIEND Low Flowand Drought Group is to continue the work carried out for the PUB Synthesis Report (Bloschl etal., 2013) on low flow estimation (Laaha et al., 2013b), and to focus on identified research gaps.While the PUB Synthesis Report assesses the performances of regionalization methods from ameta-analysis of existing studies, we will focus on bi- to multilateral studies. This allows us toaddress questions about the functioning of selected models under specific hydrological conditionsin more detail, which can then be applied to model improvements. A second aim is to extend theassessments of the PUB report towards a more detailed analysis of drought types and catchmentfunctioning, to shed light on the role of climate catchment interactions in drought generation.

Two pilot studies were initiated from the EURO FRIEND-Water Low Flow Group whichcontribute to these aims. The first study, on low flow regionalisation in the Meuse and Moselle basin,gives an example for bi-lateral assessment of regionalization methods for predicting low flows atungauged sites. The second study, on climate-catchment interactions for low flows in Austria, aimsto explore the link between low flow vulnerability and hydrological drought types (Van Loon & VanLanen, 2012) to shed light on low flow generating processes. In this paper we show both the processof the ongoing work and the way forward, and we present results ofthe analysis.

STUDY 1: ASSESSMENT OF REGIONALIZATION METHODS

Regionalisation of low flows is currently dominated by statistical methods. The PUB-Reportsummarized the state-of-the-art of low flow regionalisation and pointed out that process-based

Copyright © 2014 IAHS Press

12 Gregor Laaha et al.

methods are under-represented in the literature, so a more detailed analysis of these would be ofinterest (Laaha et al., 20l3b, Salinas et al., 2013). In this study we aim to test a process-basedmethod, i.e. catchment modelling, through detailed assessment of its predictive performancedepending on topological, geological and climatic conditions. We further aim to test thegeostatistical method top-kriging (Laaha et al., 2013a) which performed notably well in theAustrian study area in a different environment, and to learn from the relative skills of bothmethods to enhance regionalization. This is achieved in a bi-Iateral study, using the concept ofmodel and data sharing.

Catchment modelling is based on the GR4j model (Perrin et al., 2003), a parsimonious dailyfour-parameter rainfall-runoff model. Its major components are a production store and a nonlinearrouting store. Parameters still need to be calibrated with gauged streamflow records, so someregionalisation method is needed to apply the method for ungauged sites. In this study, the modelis regionalised by model output averaging (Oudin et al., 2008; AERM-UL Report, 2012), whichwas shown in a preliminary analysis to clearly outperform regional or regression and parameteraveraging procedures for the study area. The method takes the model outputs (i.e. hydrographs) ofthe four nearest gauges (in terms of Euclidean distance in geographic space), and calculates theestimate for the subject site as an equally weighted average of these outputs. From this predictedhydrograph, the low flow characteristic Q95 of the subject site is calculated.

Top-kriging is a geostatistical method that accounts for the river network hierarchy (Laaha etal., 2013b). Low flow characteristics Q95 of an ungauged site are predicted by a weighted averageof the Q95 values of the gauges as in standard kriging methods, but in top-kriging the weights areestimated by a family of variogram models (regularisations) for different catchment areas (krigingsupport), which accounts for the different scales and the nested nature of the catchments. Thisassures that kriging weights are distributed to both hydrologically connected and unconnected sitesof the stream network according to the data situation: top-kriging gives most weight to close-bysites at the same river system, but when the next hydrologically connected site is far away, moreweight is given to a close-by site at an adjacent, unconnected river system. The distribution ofweights is in contrast to ordinary kriging and stream distance-based kriging which do not accountfor both spatial proximity and network connectivity.

The methods were tested on an extensive dataset of the Meuse and Moselle basin and a fewcatchments of the Rhine basin to cover the whole Vosges mountain area (area approx. 31000 krrr'),consisting of 103 gauging stations with only minor anthropogenic influences (density: 3.3 stations( 1000 krrr'). The study area is subject to maritime temperate climate, with considerable variabilityof precipitation (ranging from less than 800 mm for lowland to more than 2000 mm formountainous areas) and evapotranspiration (ranging from 550 mm to more than 650 mm).Geology varies as well, and the main formations are limestone, marl and clay in lowlands, andsandstone, schist and granite in the mountains. These physiographical differences give rise todifferent aquifer capacities and a high spatial variability in low flows.

Figure 1 presents scatterplots of predicted and observed values of both regionalisationmethods. Overall, both methods perform well. Top-kriging exhibits a coefficient of determinationof RCV2 = 0.61 that is somewhat lower than in other studies reported in the PUB report. This maybe attributed to the large spatial variability in geology and strong precipitation gradients. Theperformance of the process-based method is somewhat lower than top-kriging (RCV2 = 0.50), butsignificantly higher than other studies on process based models reported in the PUB report.

To shed light on the reasons for the different performances, we stratified the results accordingto stream topology and physiographic factors. The analysis yielded that both methods performbetter for non-headwater (NHW) catchments with an upstream gauge than headwater catchments(HW). The respective performances for NHW and HW are RCV2 = 0.75 and 0.51 for top-kriging,and Rcv2 = 0.59 and 0.45 for the rainfall-runoff model. For top-kriging, the results are in line withthe finding of Laaha et al. (2013a) that in headwaters top-kriging tends to extrapolate the values ofthe downstream gauge, unless close-by donors in neighbouring headwater catchments are available.This extrapolation might yield some bias due to precipitation gradients and different storage

Regional assessment of low flow processes and prediction methods across European regimes 13

Ralnfal-Runoff model Top-KnQInQ

6 6

•6 • 6

E ••• • ~~

5 •• •• • ~ 5"., • • • '" •B • •• • B • •-o fIlM -o.!l 3 • ~ 3 •0ij • u" • "D:: 2 D:: 2

2 4 6 6 2 4 6 6

Observed 095 (Vs/km') Observed 095 (Vs/km')

Fi~. 1 Scatterplots o~ predicted (cross-valid~tion) and observed values of specific low.flows Q95 (inI s km") for two regionalisation methods, rainfall-runoff model (left) and top-kriging (nght).

properties of the headwaters compared to the more downstream areas. The effect on the rainfall-runoff predictions is similar for headwater catchments with high specific low flows. This is likelydue to a bias in rainfall estimates in this mountainous area. Indeed, the model inputs are based onthe SAFRAN distributed data validated throughout the French territory (Quintana-Segui et al.,2008) which shows a fairly strong bias in this area (Vidal et al., 2010).

From all results, we conclude that top-kriging performs better than the process-based methodas we expected, but the results for the rainfall-runoff model used in this study seem veryencouraging. We believe that one reason for the good performance of the process-based method isits parsimonious structure which makes it much more suitable for regionalisation than other lessparsimonious models. We have also seen that the regionalisation method used to apply theprocess-based method at ungauged sites had a major impact on the performance. The model-outputaveraging method performed much better than parameter-averaging methods, but still has potentialfor improvement. First, we selected the donor gauges by a rather simple criterion from distance ingeographic space, and a more elaborate criterion will likely improve the estimates. Secondly, theoutput averaging was performed by an unweighted average of the model outputs of the donor sites,and it needs to be proved if and how weights can be determined to better represent the degree ofsimilarity between hydrographs during low flows, and to better consider untypical catchments likekarstic basins. Furthermore, the model was not specifically tailored for low flows and amodification of the model structure in the groundwater component may improve low flowpredictions in permeable basins of the studied area, as shown in Lang et af. (2008). However, thismodified version of the model is less parsimonious and may be less suitable for regionalisation.Finally, a correction of the rainfall bias for the mountainous area is anticipated to improvepredictions of headwater catchments located in the Vosges mountains, which were of majorconcern for the practical application of the model. We also aim to develop a combined methodwhich integrates the advantageous properties of the geostatistical method and the process-basedmethod to estimate the low flow part of the hydrograph at ungauged sites more accurately thanexisting state-of-the-art methods.

STUDY 2: LOW FLOW TYPOLOGY - CLIMATE AND CATCHMENT INTERACTIONS

The second study builds on the work of Van Loon & Van Lanen (2012) and Van Loon (2013) whoestablished a drought typology based on a classification of meteorological events. Theydistinguished six hydrological drought types, i.e. (i) classical rainfall deficit drought, (ii) rain-to-snow-season drought, (iii) wet-to-dry-season drought, (iv) cold snow season drought, (v) warmsnow season drought, and (vi) composite drought. The processes underlying these drought types

14 Gregor Laaha et al.

are the result of the interplay of temperature and precipitation at catchment scale in differentseasons. They also found differences in the drought characteristics between fast and slowresponding catchments and between hydrological drought types, but this catchment-specificcomponent was not fully explored.

(a)

I' I i II

I,I

I,I I

I,

"

,I

II

1

!

11I ! ,

~J \L,

J U~ G II U U U 'u u L \...J U'-' '-' \.J/ \....J "-..J U .

Regional assessment of low flow processes and prediction methods across European regimes 15

hydrograph is more represented by rainfall peaks in any season. This leads to a quite flat thresholdlevel in those hot spots.

In a second analysis we focused on the relationship of drought duration and deficit. Theduration-deficit plots of the drought events mostly had the expected linear shape (Van Loon et al.,2013), e.g. that of Weinv (Fig. 3(a)), where deficit volumes are linearly related to duration.However, some hot spots had a duration-deficit plot with a divergent pattern in which longduration droughts can either have low or high deficit, e.g. that of Brewa (Fig. 3(b)). This is relatedto the strong seasonality in Alpine catchments, where the threshold level is high in summer anddroughts can develop freely and low in winter and drought deficit is limited by zero. A similareffect was found in soil moisture droughts in semi-arid seasonal climates by Van Loon et al.(2013), albeit with contrasting seasonality.

oM

(a) (b)--e-- Luuv41-.D.- 208058+ 208447

208637

o 20 40 60 80 100 12C

oM

o

(b)

Duration [days]

----&-,,' 200154..c:. 200204

200287

o 20 40 60 80 100 120

Duration [days]

Fig. 3 Example of duration-deficit plots of drought events for two contrasting hot spots (includinglinear regression line); (a) Weinv (four catchments), (b) Brewa (three catchments). Symbols refer toindividual catchments within a hot spot.

16 Gregor Laaha et al.

Finally, we screened whether the duration-deficit relationships depended on climate andcatchment factors. Van Lanen et at. (2013) found that the strength of the relationship mainlydepended on climate type (Koppen-Geiger classification) and on groundwater storage. We wantedto assess if we could find similar effects in the Austrian study area, where Alpine catchmentsbelong to cold climate (Type D) and the lower situated catchments belong to warm temperateclimate (Type C). The analysis yielded a strong negative relationship with elevation (the spreadincreased with altitude) which is likely an effect of climate. The effect might be due to theseasonal threshold levels, but also due to the fact that summer and winter droughts are generatedby different processes with differing recession behaviour. Besides the effect of elevation, we founda strong positive relationship between the density of the stream network and the slope of theregression line. The important effect of the stream density in these hot spots and theinterrelationship between stream density and other geological and climate parameters on differenttime scales were previously discussed in Gaal et at. (2012). This is an issue that will further beexplored in a more detailed analysis of within- and between-hot spot variability.

SUMMARY AND CONCLUSIONS

We presented two ongoing pilot studies initiated from the EURO FRlEND- Water Low Flow group.The first study gave an example of bi-lateral assessment of regionalization methods for predictinglow flows at ungauged sites aiming to develop improved predictive models. The second studycombined different expertise in drought and low flow analysis to foster process understanding. Theexamples illustrated that FRIEND-Water provides a very precious network which facilitatesinternational collaboration of water experts to tackle wicked water problems.

REFERENCESAERM-UL Report (2012) Determination des Ecoulements et des Phases Hydrologiques (DEPHy). Rapport final, 63 p.

http://www.univ-metz.fr/recherche/labos/cegumIRapportl 7Sept20 12 DEPHy.pdfBloschl, G; Sivapalan, M; Wagener, T; Viglione, A & Savenije, H (2013) (eds) Runoff Prediction in Ungauged Basins:

Synthesis across Processes. Places and Scales. Cambridge University Press; ISBN 978-1-107-02818-0.Gaal L., Szolgay, L, Kohnova, S., Parajka, L, Merz, R., Viglione, A. & Bloschl, G. (2012) Flood timescales: Understanding the

interplay of climate and catchment processes through comparative hydrology. Water Resour. Res. 48, W04511.Laaha, G. & Bloschl, G. (2006) Seasonality indices for regionalizing low flows. Hydrol. Processes 20, 3851-3878.Laaha, G., Demuth, S., Hisdal, H., Kroll, C. N., Van Lanen, H. A. l., Nester, T., Rogger, M-, Sauquet, E., Tallaksen, L. M.,

Woods, R. A. & Young, A. (2013b) Prediction of low flows in ungauged basins. In: Runoff Prediction in UngaugedBasins: Synthesis across Processes. Places and Scales (ed. by G. Bloschl, M. Sivapalan, T. Wagener, A. Viglione & H.Savenije), Cambridge University Press; ISBN 978-1-107-02818-0.

Laaha, G., Skeien, J, & Bloschl, G. (2013a). Spatial prediction on river networks: comparison of top-kriging with regionalregression. Hydrological Processes doi: I0.1002/hyp.9578.

Lang, C., Gille, E., Franois, D. & Drogue, G. (2008) Improvement of a lumped rainfall-runoff structure and calibrationprocedure for predicting daily low flow discharges. Journal 0/ Hydrology and Hydromechanics 56, 59-71.

Oudin, L., Andreassian, V., Perrin, C., Michel, C. & Le Moine, N. (2008) Spatial proximity, physical similarity, regression andungaged catchments: A comparison of regionalization approaches based on 913 French catchments. Water Resour. Res.44, W03413.

Perrin, C., Michel, C. & Andreassian, V. (2003) Improvement of a parsimonious model for streamflow simulation. J. Hydrol.279(1-4),275-289.

Quintana-Segui, P., Le Moigne, P., Durand, Y., Martin, E., Habets, F., Baillon, M., Canellas, C., Franchisteguy L. & Morel, S.(2008) Analysis of Near-Surface Atmospheric Variables: Validation of the SAFRAN Analysis over France. J. App.Meteorol. Climatol. 47,92-107.

Salinas, J, L., Laaha, G., Rogger, M., Parajka, L, Viglione, A., Sivapalan, M. & Bloschl, G. (2013) Comparative assessment ofpredictions in ungauged basins - Part 2: Flood and low flow studies, Hydrol. Earth System Sci. 17, 2637-2652,doi: I0.5 I94/hess- I7-2637-2013.

Van Lanen, H. A. l., N. Wanders, L. M. Tallaksen & A. F. Van Loon (2013) Hydrological drought across the world: impact ofclimate and physical catchment structure. Hydrol. Earth System Sci. 17, 1715-1732, doi: 1O.5194/hess-17-1715-2013.

Van Loon, A. F (2013) On the propagation of drought. How climate and catchment characteristics influence hydrologicaldrought development and recovery, PhD Thesis, Wageningen University [http://edepot.wur.nI/249786, last access: 28May2013].

Van Loon, A. F. & Van Lanen, H. A. J, (2012) A process-based typology of hydrological drought. Hydrol. Earth System Sci. 16(7),1915-1946, doi:10.5194/hess-16-342 1915-2012.

Van Loon, A.F., Tijdeman, E., Wanders, N., Van Lanen, H. A. L, Teuling, A. J. & Uijlenhoet, R. (2013) How ClimateSeasonality Modifes Drought Characteristics. Journal ofGeophysical Research (submitted).

Vidal, l.-P., Martin, E., Franchisteguy, L., Baillon, M. & Soubeyroux, l.-M. (2010) A 50-year high-resolution atmosphericreanalysis over France with the Safran system. Int. J. Climatol. 30: 1627-1644.

Hydrology in a Changing World: Environmental and Human DimensionsProceedings of FRIEND-Water 2014, Montpellier, France, October 2014 (IAHS Pub!. 363, 2014).

Progressive aridity impact on the hydrological regime in theVolta River basin in Benin (West Africa)

HENRI s.v.TOTIN',2, ARNAUD ZANNOU 3, ERNEST AMOUSSOU"4,ABEL AFOUDA S & MICHEL BOKO'I Department ofGeography, University ofParakou, BP 123, Parakou, Benin2 Laboratory Pierre PAGNEY, Climate, Water, Ecosystems and Development, 03 BP I 122 Jericho, Cotonou,University ofAbomey-Calavi, Beninsourouhenri@yahoo.fr; totinsourouhv@gmail.com3 National Hydrological Service, Ministry of Water Resources, Benin4 Climatology Research Centre (CRC), University ofBourgogne, 6 Boulevard Gabriel, 2 I000 Dijon, France5 Laboratory ofApplied Hydrology, University ofAbomey-Calavi, Benin, BP 526 Cotonou, Benin

17

Abstract The Volta River basin within the Soudano-Sahelian climate zone in Benin is experiencingprogressive aridity, of which the impacts on surface water runoff are large. In this context, it is necessary tounderstand 'he hydrological regime change to provide the best decision tools for water managers. For thispurpose, methods of statistical variability analysis, calculation of aridity index based on the UNEPclassification and discontinuity detection on rainfall and flow series were used over the period 1961-20 IO.The water deficit index, the base flow index and the seasonal irregularity index were calculated. This studyrevealed that since 1975 the Volta basin in Benin has experienced a drought persistence leading to anincreased climate aridity (20% from 1961 to 2010). Rainfall decreased by 13% while the aridity index variesfrom 0,01 to 2.66, indicating dry sub-humid climate to hyperarid climate at the monthly scale. Indeed,annual variation in flow decreased by 41% in the sub-basin of Porgaand 32% in that of Tiele over the period1975-20 I0 compared with 1961-1975. Also, an increased base flow index of 0.6 to 0.8% at Porga, 0 to0.1 % at Tiele supported by very marked seasonal irregularity were relevant indicators of hydrologicaldrought and surface water scarcity. The sensitivity of the hydrological regimes to the change of aridity in theVoila basin should be a basis for sustainable water management strategies development in the Sudano-Sahelian area of Africa.Key words hydrological regime; aridity index; surface water scarcity; sustainable management; Volta River basin;Benin

INTRODUCTION

Surface water resources, dependent on very irregular rains become increasingly limited because ofclimate change resulting in decreasing precipitations, rising temperature and consequently higherevaporation. According to Smakhtin (200 I), long-term periods of little rain in catchment areasbring about hydrological drought that manifests in rivers as low flows.

Surface water reduction on the mean West African basins reached 40-60%, much moremarked than that of precipitation (15-30%) (Afouda et al., 2007). Climate warming may lead toimportant changes in river flow (Kunstmann & lung, 2005; Opoku-Ankomah & Minia, 2005;lung, 2006; Idieti, 2009; etc.). Changes in water flux between the surface of the Earth and theatmosphere are not expected to be spatially uniform but to vary much like the current daily meanvalues of precipitation and evaporation (IPCC, 2007). The impact of a drought persistence (Totinet al., 2009) is indeed largely amplified in the flow which records a reduction three to four timesmore significant (30-40%) than that of rainfall in the Benin basin of the Volta River (Totin et al.,2010). In this basin, the evidence of drought impact on the surface water resources is its increasingwater scarcity.

Decreasing rainfall does not only affect flowing river water within catchments, but it alsocauses considerable changes in the hydrological regime, especially on the Sudano-Sahelian climatezone of Africa including the study area. But the effects of drought or aridity on the streams andrivers regime are less on the Volta River basin in Benin.

The Volta basin, situated between latitudes 0040'N and 2°IO'N and longitudes II 040'E and9°21'E (Fig. I) in northern West Benin (12.1%) covers an area of 13590 km2 representing 3.4% ofthe whole basin extended to Burkina Faso, Cote d'Ivoire, Ghana, Mali and Togo.

Copyright © 2014 IAHS Press

18 Henri S. V. Tatin et al.

,,.\JIQUf::$~II.V(R

t,,,,,

t

_ . l..smlIbf !U le

- • _ - U IM (), VOIla DU!Iln

r:::::::l SlI'JOMn I)'Ma-gour::J ~'fl 01P~n

~ I

]

,.

«r

.~ . t\

"w ·P « " J IIL ~

lJ'TI)"O

+~.

.. t

,,' ...,. -

T C ~

I' '-.,

.-

'- -, • • _ ,.",,, -v , 'o r ......".. ," '_".! -..'

~ r~ (

.,,'\.,.,., BASIN OF VOLTA RIVER -r.' ,( \, 1".EPUBLlC OF r,------------,... , (J

1-(

..:--'

~ ,

I

Fig. 1 The geographical location of the Volta River basin in Benin.

It is formed by the gauged sub-basins of the River Pendjari (22 280 km2 include Burkina part)and Magou (836 krrr') and the non-gauged streams sub-basins of Koumongou, Keran and Kara.Within the Inter Tropical Convergence Zone, its climate is dry tropical (Sudano-Sahelian climatezone), featuring one dry and one rainy season from April to October and November to Marchrespectively. Climate variation at all time scales influences the hydrological process which directswater availability necessary for the ecological and economic systems in this basin.

The aim of this study is to assess the change of the Volta hydrological regime in the context ofa drying climate in northern Benin.

DATA AND METHODS

Data

To investigate the sensitivity of the Volta River regime to changing climate in Benin, rainfal lrecords, potential evapotranspiration data and the Pendjari and Magou River flows data were usedover the period 1961-2010.

Also, existing literature (Moron, 1994; Mahe et al., 2001; Afouda et al., 2007; Idieti, 2009:Totin et al., 2009, 2010) helped to assemble complementary data necessary to analyse thevulnerability of the study area to climate variation. Field work enables the identification of theenvironmental tracers of the hydroclimatic variation in the Volta basin.

METHODS

Estimation of the aridity index

The aridity index (AI) (UNEP, 1992; Middleton & Thomas, 1997; Turkes, 1999; Tsakiris &Vangelis, 2005; Tsakiris et al., 2007) is determined by the ratio of mean precipitation (P) to meanpotential evapotranspiration (PET) on a monthly scale for each monitoring station.

Al=~ (1)PET

Progressive aridity impact on the hydrological regime in the Volta River basin in Benin 19

The Aridity Index was used to delineate the dry and wet months based on the meaning of the termshyperarid (AI < 0.05), arid (0.05 < AI < 0.20), semi-arid (0.20 < AI

20 Henri S. V. To/in et at.

homogeneity tests of Pettitt for time series leads to a decrease of J3% and drought persistencesince 1970. According to Totin et ut. (2009) the occurrence of the dry climatic conditions is due tothe influence of the Sahelian climate from the north on the basin.

This annual rainfall trend is evidence of change in the monthly precipitations over the studyperiod, as shown in Fig. 3. At the monthly scale. except January with increasing rain (6%) butnegligible rainfall anywa y, the Volta basin registered decreasing rainfall from 1% (May) to 58%(November). There is a clear link between drought in the Sahel, sea surface temperature (SST) andupwellings in the Gulf of Guinea (Le Houerou, 1996) and the climate of the study area.

zso -

10 1961-1% 9 1 I• 1970-2010

I

I

II

IT:I

II

II

I I II

II II

III I

I

rII I

II I

I I IrlI..I

" .\ \ 1 A u N I>

Fig. 3 Variation of the mean precipit ations regime on the Volta Riverbasin in Benin ( 1961-20 10).

Table I Evolution of the mean aridity index over the period 1961-20 10 on the Voila River basin in Benin.

F M M J J A S 0 N D

1961-1 96'.1 0.84 1.62 2.39 3.13 2.72 1.031970-2010 0.81 1.21 2.00 2.60 2. 17 O. 81961- 2010 0.82 1.27 2.06 2.66 2.24 0.74

Moreover, seasonal rainfall vanati on (1970 · 20 10 compared to 1961-1969) is - 12% and-3 2% for the rainy season and the dry season, respectivel y. Indeed, this signature of the dryingclimate postulates that the Volta River basin in Benin faces drought conditions and certainlyincreasing aridity.

Aridity indicators (Table I) show that the Benin basin of the Volta River experienced withineach year six climate contexts such as hyperarid climate (December to February), arid (March andNovember), semi-arid (April over 1970-20 10), dry sub-humid (Apri lover 1961-1969 and1961-20 I0), sub-humid (October) and humid/hyper-humid (May to September).

The aridity index variation over 1961 -2010 (0.01 to 2.66) illustrates a slide of thc hyper-humid climate context (.fAS) to that of hyperarid climate (DJF). This index of the period1970-2010 compared to 1961-1969 (Table I) decreased by 20%, explaining the drying climate.This is related to the fact that the Volta River basin in Benin, near southwestern Burkina Faso ismore and more influenced by the Sahelian dry climate . Furthermore, change of temperatu re(lPCC. 2007) , however moderate, would correspond with a significant increase in climatic aridity(Le Houerou, 1996) in this basin.

Hydrological regime sensitivity to drying climate

The year 1975 is the break point in the hydrological time series stationarity, six Yl:3rS after that ofthe heavy rainfall of 1969. The observed increasing level of aridity affects hydrological processes

Progressive aridity impact on the hydrological regime in the Voila River basin in Benin 21

in the Volta River basin. Indeed , variation in the hydrological regime (Fig. 4) shows a decrease inmean annual flow by 41% in the sub-basin of Porga and by 32% in that of Tie le 71 to 41.5 m3/sand 5.5 to 3.7 m' /s, respecti vely, from the period 1961-1975 to 1975-2010 .

Correlation between the aridity index (Table I ) and monthl y flow (Fig. 4) highlights theeffects of the dry climate on the Volta River. The hydrological change is marked at the monthlysca le by higher variability and decreasing flow (8--47% at Porga and 3 1-62% at Tiele). Thishydrological variation certainl y results from a spatiotemporal slide of the humid climate to the aridclimate on the basin.

(a) Pendjari Rlver at Porga

I .1.50 -11- 1 'l6 1 - 197~ 1I 300 . GJ 1976-2010

I ~250 -

21VI

~ec" 150 -

I0<

".." irn

~>-e

50 .

~ l,0

J F M A M J J A S 0 N 0

-- - - -Fig . 4 Variation of the mean hydrological regime on the Volta River basin of Benin ( 1961-20 10):(a) River Pendjari and (b) River Magou.

Table 2 Variation of the base flow index (8FI) and the seasonal irregularity index (SII).

Sub-basins Magou River (Tiele) Pendjari River (Porga)Periods 196 1- 1975 1976--2010 1961-20 I0 1961-19 75 1976--20 108 FI (%) O.DO 0.06 0.03 0.64 0.85SIl (%) 575.7 543.0 555.6 476.0 483.0

1961- 20100.8 1480 .2

Analysis or other hydrological indicators such as the base flow index and the seaso nalirregularity index (Table 2) reveals the sensitivity of the Volta basin to its progressive aridificat ion.

Within the context of the hyperarid climate (Dece mber to Februa ry) and of the arid climate(March and November) ami over the period 1976-20 I0 compar ed with 1961-1975, the base nowindex increase from 0.6 to 0.8% at Porga and 0 to 0.1% at Tiele. This pattern of the hydrologicalrecession and the surface water scarcity is supported by a very marked seasonal irregularity linkingthe aridity severity, as shown by the environmental state in Fig. 5.

(a)

Fig. 5 Slate of the Volta River basin in time of the hyperarid climate: (a) on the Magou sub-basin and(b) on the Pendjari sub-basin in 8enin.

22 Henri S. V. Totin et al.

Reduction or destruction of the perennial plant cover also decreases the rugosity of thelandscape resulting in higher wind speeds at the soil surface and hence higher rates ofevapotranspiration and increased aridity (Le Houerou, 1996) as observed on the Volta River basinin Benin.

CONCLUSIONS

Since 1970 on the Volta River basin in Benin, annual rainfall and flow have changed by -13%,and -30% to --40%, respectively, leading to the climatic and hydrological droughts persistence.Indeed increase in climatic aridity (20%) results from the precipitation and flow regimessensitivity to the effects of the dry Sahelian climate.

As the monthly aridity index showed, it is evident that the Volta basin in Benin is affected andfaces problems of water supply.

Hydrological regime sensitivity to the increasing aridity, in the Volta basin, should be anindicator to develop adequate and sustainable environment protection and water managementstrategies in the Sudano-Sahelian area of Africa.

REFERENCESAfouda, A., Ould Baba Sy, M., Gaye, A. T., Cabral, A., Nazoumou, Y., Compaore, l. A. & Sanoussi, R. (2007) Impacts du

changement et de la variabilite climatiques sur les ressources en eau des bassins versants ouest africains: Quellesperspectives? Adaptation aux changements c1imatiques et gestion des res sources en eau en Afrique de !'ouest. Rapport desynthese de Writeshop.

Humbert, J. & Kaden, U. (1994) Detection des modifications de l'ecoulement fluvial au moyen de I'indice de debit de base.Rev. de Geogr. Alpine 2, 25-36.

ldieti, E. M. (2009) Variabilite hydro-c1imatique dans le bassin versant de la Pendjari (Nord-Ouest du Benin), Mernoire deDEA, Universite d' Abomey-Calavi, Cotonou, Benin,

IPCC: Climate Change (2007) The Physical Science Basis. In: Contribution of Working Group I to the Fourth AssessmentReport of the Intergovernmental Panel on Climate Change (ed. by S. Solomon, D. Qin, M. Manning, Z. Chen, M.Marquis, K. B. Averyt, M. Tignor & H. L. Miller). Cambridge University Press, Cambridge, UK and New York, NY,USA.

lung, G. (2006) Regional climate change and the impact on hydrology in the Volta Basin of West Africa. Dissertation,Garmisch-Partenkirchen, University of Augsburg, Germany.

Kunstmann, H. & lung, G. (2005) Impact of regional climate change on water availability in the Volta basin of West Africa. In:Regional Management ofWater Resources (ed. by T. Wagener), 1-11. IAHS Publ. 295. IAHS Press, Wallingford, UK.

Le Houerou, H. N. (1996) Climate change, drought and desertification. J. Arid Environ. 34, 133-185.Mahe, G., L'Hote, Y., Olivry, J. C. & Wotling, G. (2001) Trends and discontinuities in regional rainfall of west and central

Africa - 1951-1989. Hydrol. Sci. J. 46(2), 211-226.Middleton, N. & Thomas, D. (1997) World Atlas ofDesertijication. Amold, London.Moron, V. (1994) Guinean and Sahelian rainfall anomaly indices at annual and monthly scales (1933-1990) Int. J. Climatol. 14,

325-341.Opoku-Ankomah, Y. & Minia, Z. (2005) Climate change scenarios and impacts on surface water resources of the Volta River

Basin. In: Climate and Anthropogenic Impacts on the Variability of Water Resources, IHPIUNESCO, TechnicalDocument in Hydrology 80, 165-172.

Perrin, C. (2000) Vers une amelioration d'un rnodele global pluie-debit au travers d'une approche comparative. These deDoctorat, Inst