Our Common Future Under Climate Change

International Scientific Conference 7-10 JULY 2015 Paris, France

  • Home
  • Zoom Interactive Programme
Cliquer pour fermer

Wednesday 8 July - 14:30-16:00 UNESCO Fontenoy - ROOM IX

2222 - Semi-Arid Regions Adaptation

Parallel Session

Lead Convener(s): P. Mudliar (Indian Institute of Human Settlements, Bangalore, India)

Convener(s): S. Traore (International Crops Research Institute for the Semi-Arid Tropics, Bamako, Mali), C. Singh (Indian Institute of Human Settlements, Bangalore, India), A. Rahman (Indian Institute for Human Settlements, Bangalore, India)


Keynote speaker

T. Mitchell (Overseas Development Institute, London, United Kingdom)

Abstract details
Keynote speaker
Abstract content

Keynote speaker

E. Totin (International Crops Research Institute for the Semi-Arid Tropics, Bamako, Mali)

Abstract details
Keynote speaker
Abstract content

Adaptation Research in Semi Arid Regions of South Asia

S. Gajjar (Indian Institute for Human Settlements, Bangalore, India)

Abstract details
Adaptation Research in Semi Arid Regions of South Asia

S. Gajjar (1)
(1) Indian Institute for Human Settlements, Environment and Climate, Bangalore, India

Abstract content

Faced with high climate variability and a multitude of development challenges, people in the semi-arid regions of Africa and Asia are particularly vulnerable to climate-related risks. Interventions to date have focused mainly on short-term solutions, leaving large gaps in our understanding of which responses will enable widespread, effective and sustained adaptation over longer time horizons.

The Adaptation at Scale in Semi-Arid Regions (ASSAR) project aims to fill the above knowledge gap. Central to the ASSAR project is a focus on multi-stakeholder processes, which help in identifying key adaptation and development-related vulnerabilities, and determining responses that can yield tangible and lasting benefits to society.

Key insights are emerging from South Asia’s multi-institutional team’s research and interaction with a range of stakeholders. Over the past year, policy makers, senior government officials, community members, academics as well as practitioners were engaged through a variety of formats including key informant interviews, participatory workshops and research validation events.

Initial research findings suggest that the welfare cost of climate change impacts in India varies across geographies and sectors. Given the high prevalence of natural resource based livelihoods, high incidence of poverty and inherent socio-economic inequalities, a significant section of the population is ill-equipped to adapt to current and projected climate variability. While households dependent on agriculture are directly affected, those living in urban areas are also affected by declining agricultural productivity. At the same time, rapid and largely unplanned urbanization manifests in an intense competition for resources and land. Quality of life for marginalized groups in urban areas is characterized by lack of access to social capital, poor quality of jobs and exclusion from public services. This begs the question whether migration for livelihood diversification or improved well-being actually translates into a reduction in vulnerability, or simply into a differently characterized vulnerability.

Practitioners working with urban dwellers and rural households over decades carry unique tacit knowledge of local barriers to effective adaptation, mal-adaptive practices and the unintended outcomes of current institutional, policy and governance structures. As part of ASSAR, climate scientists aim to generate climate messages with the intention to test whether improved climate information heightens adaptive capacity, for those whose lives and livelihoods are directly exposed to climate-related risk. Social scientists are poised to study local governance structures that control the sharing of common resources such as water and forests, as well as patterns of migration across a rural to urban continuum. Research across different domains is driven by the need to produce knowledge which is useful to vulnerable people living in semi-arid regions, either directly or through policy influence.

Using a collaborative research approach, while drawing upon the experiences of practitioners, ASSAR aims to inform climate change adaptation policy and practice in ways that strengthen livelihoods and sustain the well-being of the most vulnerable. Conducting and aligning such research is a complex task, requiring at all levels, leaders who can converse in several disciplinary discourses, becoming conduits of scientific and practical knowledge across different communities. At the same time, the project generates valuable lessons for future research initiatives aspiring to contribute to climate change adaptation. 


Famine, climate, agriculture and desertification in Northeastern Brazil : from historical reconstruction to future adaptation

M.-P. Ledru (Institut de Recherche pour le Développement, Montpellier, France), F. Soares De Araújo (Universidade Federal do Cearà, Fortaleza, Brazil), L. R. Zanette (Universidade Federal do Cearà, Fortaleza, Brazil), L. Bremond (Ecole Pratique des Hautes Etudes, Montpellier, France), C. Favier (Centre National de la Recherche Scientifique, Montpellier, France)

Abstract details
Famine, climate, agriculture and desertification in Northeastern Brazil : from historical reconstruction to future adaptation

MP. Ledru (1) ; F. Soares De Araújo (2) ; LR. Zanette (2) ; L. Bremond (3) ; C. Favier (4)
(1) Institut de Recherche pour le Développement, UMR 226 ISEM, Montpellier, France; (2) Universidade Federal do Cearà, Departamento de biologia, Fortaleza, Brazil; (3) Ecole Pratique des Hautes Etudes, Umr isem, Montpellier, France; (4) Centre National de la Recherche Scientifique, Umr isem, Montpellier, France

Abstract content

The previsions say that the semiarid regions in the world are between the most impacted by the global climate change. In Northeastern Brazil, among the environmental (lost of biodiversity, extreme droughts), social (migrations, diseases exposition) and economics (reduction on income generation, in employment rates, and in the Gross Domestic Product) of global warming, it is affirmed strong pressures specially over the environment, the agriculture and livestock because of the increase of the rain irregularity and increase of evaporation. During the last century Northeastern Brazil and more particularly the state of Ceara, was submitted to successive catastrophic droughts that enhanced severe famine and massive migrations towards the big urban centers in southeastern Brazil. Land use in this semi arid area has increased the size of degraded areas and the intensity of degradation and desertification. Overgrazing by livestock is the main origin for the degradation process. However, with the climate warming and the drastic decrease of livestock production, farmers were forced to revert to more primitive forms of subsistence farming  with a land rotation system (nomadic farming) that relies on fire for vegetation clearing during the dry season. With the increase in human population density and the partitioning of the large landlord farms, the interval between each land use cycle has been decreasing.

The IPCC 2014 report shows that increase in temperature change in seasonality and decrease in precipitation will strongly affect this region althugh a large uncertainty due to the scarcity of environmental datas that prevents a good precision in most of the reconstruction models.

The multidisciplinary project developped at the UFC in the frame of a cooperation with IRD aimed to reconstruct the climate and landscape changes during the 20th century. The project is multidisciplinary and is using an historical reconstruction of land use based on microfossil data, the current state of vegetation cover and the genetic diversity of key indicator species.  

Modern landscapes are described in different types of degradated areas, from well preserved to very degraded, through botanical surveys and remote sensing analyses. The evolution of the landscape during the last century is analysed on the base of paleo ecological methods adapted to shorter time scales. Short sediment cores were collected in selected dams within the different ecosystems and changes in landscape are reconstructed on the base of bio indicators and microcharcoal analyses. These changes are compared with observed climate datas. Climate and man influence and impact on the desertication processes are defined at fine scale and analysed separately. We aim to answer the following questions: (i) what is the history of land use and diversity loss in areas under a degradation/desertification process in the last century?(ii) What is the quality of the biodiversity in the surrounding area of the degraded areas ? iii) How the defined ecosystem services could be used for natural or induced restoration/recovery of the degraded areas?(iv) what will be the resilience of the remaining Caatinga areas for the next century ?

This project started in 2014 and is carried out by researchers at the Federal University of Ceará, Brazil - UFC and the Institute of Evolutionary Sciences of the Montpellier 2 University, France.


Adaptive Capacity and Risk management: Evidence from household-level data from semi-arid tropical villages in India

A. Rahman (Indian Institute for Human Settlements, Bangalore, India), P. Mudliar (iihs, Bangalore, India), C. Singh (Indian Institute of Human Settlements, Bangalore, India)

Abstract details
Adaptive Capacity and Risk management: Evidence from household-level data from semi-arid tropical villages in India

A. Rahman (1) ; C. Singh (2) ; P. Mudliar (3)
(1) Indian Institute for Human Settlements, Bangalore, India; (2) Indian Institute of Human Settlements, Bangalore, India; (3) iihs, Bangalore, India

Abstract content


This paper attempts to provide evidence on how households respond to climatic variability and extreme events such as droughts/floods in the semi-arid regions villages of India. It looks at how vulnerabilities inherent in farming households and the risks faced by them, which shape and affect local adaptive capacity. Two key questions which this paper would be address are – the impact of climate change on food production and the extent of climate induced migration. The paper focusses on semi-arid regions since this region is highly vulnerable to several socio-economic risks apart from rainfall-related production risk.


Semi-arid regions constitute almost 53 percent of India’s land cover, where agriculture is the primary source of livelihood. In addition to being drought-prone and being vulnerable to rainfall-related production risks, semi-arid regions are characterized by higher levels of poverty, poorer asset ownership, higher indebtedness, smaller land holdings, and lower levels of land productivity. Poor performance of agriculture and lack of other livelihoods options further limit their capacities to anticipate and respond to climatic variability.


Household level panel data for 17 villages in India has been collected by the International Crops Research Institute for Semi-Arid Tropics (ICRISAT) which is used for the analysis here. This is a unique dataset in the sense that it provides an opportunity to track the economic status of households over time along with the information whether they faced droughts/floods during the period and the response adopted by them.

Relevance of the Research

This research would enable a deeper understanding of the conditions under which household vulnerabilities in semi-arid regions are shaped. This has direct implications for their current and future ability to adapt to climate change. The uniqueness of the data helps us identify which households are vulnerable to climate change, measure the impact of climate change on household assets and production patterns, and measure adaptation responses to climate shocks. In a nutshell, this paper aims to answer questions related to vulnerability, impacts and adaptation within the context of climate change in semi-arid villages in India.


Tropical warm semi-arid regions expanding over temperate latitudes in the projected next century

A. Rajaud (Laboratoire des Sciences du Climat et de l’environnement, Gif-sur-Yvette cédex, France), N. De Noblet-Ducoudré (Laboratoire des Sciences du Climat et de l’environnement, Gif-sur-Yvette cédex, France)

Abstract details
Tropical warm semi-arid regions expanding over temperate latitudes in the projected next century

A. Rajaud (1) ; N. De Noblet-Ducoudré (2)
(1) Laboratoire des Sciences du Climat et de l’environnement, Gif-sur-Yvette cédex, France; (2) Laboratoire des Sciences du Climat et de l’environnement, CEA-CNRS-UVSQ, Gif-sur-Yvette cédex, France

Abstract content

Two billion people today live in so-called dry lands (Millenium Ecosystem Assessment, 2005), where extreme climatic conditions prevail, and natural resources are limited. The future of those drylands is therefore challenging for many human beings. Drylands are expected to expand under several scenarios of climatic change (e.g. Feng and Fu, Atmos. Chem. Phys, 2013). However, no study focuses specifically on the evolution of strictly climatically defined semi-arid zones. These represent the frontier areas where intermediate conditions might shift either towards fully arid (desert), or less water-limited (favorable to rainforests), climate.

Climatic transformations of today's and future warm semi-arid regions raise the following two issues: 1) will these regions expand and thus widen the area where specific water-limited development issues must be addressed? 2) How will climatic conditions evolve in today's semi-arid regions, and what does it imply for development programs?


To address these, we adopted a bioclimatological approach based on the Koppen climate classification (e.g. Rubel and Kottek, Meteorol. Z., 2010). Monthly surface air temperature and precipitation are combined to identify bioclimatic regions. Geographical migration of these regions can thus be followed over time. We focus on the warm semi-arid class, characterized by a mean annual temperature exceeding 18°C and water-limitation criteria calculated from the seasonal distribution of rainfall and temperature.

A multi-model ensemble of CMIP5 projections was selected to draw and analyze the evolution of current and future warm semi-arid regions according to 3 representative concentration pathways of increasing emissions (RCPs 2.6, 4.5 and 8.5). The Koppen climate classification was first applied to key-periods - beginning, middle and end of the 20th and 21st centuries - to localize past and future warm semi-arid regions. Then time-series for the classification were built using a 15-year running average climatology from 1901-1915 to 2086-2100, in order to characterize trends and variability in the evolution of those regions. Averages over a 15-year interval allow to buffer the large interannual variability.


According to the CRU datasets, global expansion of the warm semi-arid area has already started, following the global warming trend since the beginning of the 20th century (~+13%). Climate models are able to capture this trend albeit with a delayed increasing trend.  This expansion will continue according to all projections, especially for the most severe scenarios, reaching ~19% and 51% respectively for RCP2.6 and RCP8.5 (multi-model mean). This will be most significant outside the actual tropical belt towards the North Pole. This poleward migration is of  ~1°, 7° and 10° for respectively RCP2.6, 4.5, and 8.5.

Two processes mainly explain this expansion: warming and drying. Drying is mostly responsible for the conversion from equatorial, subhumid climates within the tropical belt. Beyond 30° of latitude, warming is directly responsible for the conversion of cold semi-arid towards warm semi-arid climates; particularly so in RCP8.5. A third category concerns the conversion of temperate climates to warm semi-arid, outside of the tropics. In this case, both warming and drying processes are at stake, as drying effects are amplified in warmer conditions.


Those various transitions may have drastic but also very distinct ecological and sociological impacts. We therefore stress that many of the warm semi-areas that will appear in the future may need to be dealt with differently than tropical semi-arid regions in the past.


(Re)conceptualising maladaptation in policy and practice: towards an evaluative user-focused framework for semi-arid regions

L. Jones, (Overseas Development Institute, London, United Kingdom), E. Carabine (Overseas Development Institute, London, United Kingdom), L. Schipper (Overseas Development Institute, Stockholm Environment Institute, Berkeley, United States of America)

Abstract details
(Re)conceptualising maladaptation in policy and practice: towards an evaluative user-focused framework for semi-arid regions

E. Carabine (1) ; L. Jones, (1) ; L. Schipper (2)
(1) Overseas Development Institute, Climate and Environment Programme, London, United Kingdom; (2) Overseas Development Institute, Stockholm Environment Institute, Berkeley, United States of America

Abstract content

Maladaptation is a concept that has received notable policy attention in recent years, but has yet to be fully explored in both conceptual and practical terms. As a consequence it suffers from a lack of consensus around its definition and application in policy and programming. While there is general agreement that maladaptation involves action to adapt to change that increases vulnerabilities or reduces adaptive capacity, there is disagreement over what causes such action.

One of the reasons why maladaptation is such a powerful term is that it encourages practitioners and policy-makers to recognise that the decisions they take now to address climate change can backfire and inadvertently make people more vulnerable in the longer-term. However, a lack of consensus and clarity on how to characterise maladaptation currently prevents decision makers from being able to apply the concept in practice.

In the context of semi-arid regions, the immediacy of many development challenges, the susceptibility of these areas to existing climate variability and the inherent non-linearity of these systems contribute to a higher risk of maladaptive outcomes arising from development or adaptation strategies.

We outline five areas of conceptual clarity needed in understanding and evaluating maladaptation in semi-arid regions and elsewhere.

First, deliberate non-action should, if contributing to increased climate risks and negative outcomes for people and communities, be considered as maladaptation. The semi-arid lands of East Africa, for example, have long been actively marginalised from the economy and politics. Over the past decades, a lack of development in these regions over long periods has contributed to high levels of vulnerability.

Second, a strategy that has not considered the impacts of climate change should constitute maladaptation. For example, many pastoralist communities have been encouraged to diversify livelihoods as a result of persistent drought. But moving away from semi-nomadic pastoralism can erode existing adaptive capacity, leading to greater vulnerability.

Third, a fundamental component of maladaptation is time. It is only with time that the success or failure of interventions will become evident. This trait is well illustrated in the context of groundwater abstraction for adaptation in semi-arid regions, which may be maladaptive in the longer term.

Fourth, semi-arid ecosystems, livelihoods and economies are not static. Moreover, under climate change, climate risks and vulnerabilities to particular climate variables are likely to shift. Semi-arid systems are characterised by non-linear and unpredictable dynamics in both their socio-economic and ecological components.

Fifth, distributional aspects of adaptation are important because climate change is likely to affect segments of the population differently, both in terms of direct impacts as well as influences on wider drivers of development, and the act of implementing (or choosing not to implement) an adaptation strategy can fail to uniformly reduce climate risks across all social groups. For example, strategies to enhance and diversify livelihoods for adaptation to climate change and variability can reinforce inequalities in terms of gender, or livelihood types.

Building on this reconceptualisation of maladaptation, we present the groundwork for a framework that can lend itself to qualitative and quantitative assessment of adaptation strategies, and clarify the differences between four distinct types of adaptation outcomes - ranging from optimal adaptation to maladaptation. Most importantly, we have used the framework to highlight a number of different ‘symptoms’ that can act as early warnings for maladaptive outcomes, hoping to guide policy makers in achieving early-diagnosis. 

In doing so, our aim has been to make this onerous concept more tractable and applicable to planners and practitioners so as to diagnose strategies likely to lead to maladaptation. It is our hope that this work will stimulate debate and galvanise interest in advancing efforts to understand, and critically, to avoid maladaptation in semi-arid regions in the face of increasing climate risks in the coming decades.