Strengthening Resilience in Fragile Contexts - A Geospatial Impact Evaluation in Mali

If development policy is to achieve good results in climate-vulnerable and fragile contexts, evidence-based policy design takes on particular importance. Geospatial techniques can be of great benefit in this respect. In conflict-affect regions especially, the use of geographical data offers an objective and low-cost adjunct or even an alternative to data collection in the field. As the technology has advanced and been driven forward by the restrictions on mobility and travel caused by the global COVID-19 pandemic, remote sensing data is being used considerably more frequently for the production of rigorous evidence. Many evaluations based on geographical information confine themselves to measuring simple outcome indicators, such as raised agricultural production. On the other hand, the use of geospatial information for analyses and assessments of more complex concepts such as climate resilience still tends to be the exception.

As part of an evaluation of measures for climate change adaptation, DEval considered measures on the irrigation infrastructure in Mali, which is a typical intervention to strengthen climate resilience in fragile contexts.

Since the main form of agriculture in Mali – and in the Sahel zone generally – is rain-fed farming, climate change is having substantial consequences for food production. Rainfall is becoming less and less predictable, added to which, rising temperatures createn a growing need for agricultural irrigation and increase people’s vulnerability to extreme weather events such as prolonged droughts and severe rainfall events. Against this backdrop, efficient and ecologically sound irrigation of agricultural land has a high potential to increase Malian society’s resilience to climate change. Irrigation improves agricultural productivity and makes it less dependent on rain. This not only has the potential to reduce poverty but ultimately also to support the social stability of Malian
society.

German development cooperation in Mali has been supporting the installation and maintenance of agricultural irrigation infrastructure since the end of the 1990s. KfW Development Bank promotes pumpbased small-scale irrigation, large-scale gravity-based irrigation systems and enhanced management of floodplains via a variety of projects. The objective of the adaptation measures is to strengthen climate resilience. In a country affected by violent conflicts, the projects also include conflict-sensitive measures aimed at strengthening resilience on a broader scale.

The rigorous geospatial impact evaluation conducted by DEval focuses on how infrastructure interventions contribute to the ecological, economic and social resilience of the Malian population. It employs a quasi-experimental design (difference-in-difference analysis) and the interpretation of high-resolution aerial photographs. The data basis consists of almost 1,000 geocoded project locations, remote sensing data, geocoded survey data and incident statistics as well as interviews and focus group discussions. The analysis makes use of both panel data and repeated crosssectional data. Geospatial impact analyses until now have usually only examined shortterm effects with reference to simple indicators at outcome level.

Thanks to the availability of large amounts of geocoded data covering a long – 20-year – period of time, the evaluation not only rates the development effectiveness of the measures but also their sustainability. This it does by measuring their long-term effects on food security, child health, incomes, women’s empowerment, social cohesion (in relation to conflict risk and conflict intensity) and the environment.

Until now, impact analyses on irrigation have usually compared project locations with similar areas that were not part of the intervention (“treatment” and “control” units). This approach, however, is problematic in regions where many different donors are active because there is no way of guaranteeing that a control area has not been “contaminated” by other activities. If artificial irrigation is also taking place in supposed control areas – unbeknown to researchers – for example, then a comparison of intervention and control units could result in mistaken assessments of the results. The DEval evaluation takes advantage of the staggered introduction of the measures by comparing the locations before the start of the project (control group) with the same locations after the start of the project (intervention group). At the same time, the difference-in-difference design controls for
different temporal trends and non-temporal differences between locations.

The evaluation finds that supporting the irrigation infrastructure contributes to a significant and long-term rise in agricultural production. By increasing crop yields, the measures have contributed to better child health and to lowering the conflict risk in the surrounding communities. The findings indicate that despite a protracted high-conflict situation, in semi-arid areas sustainable irrigation systems can be an effective instrument for improving the resilience of communities to cope with current and future negative socioeconomic impacts of climate change.

The DEval approach to geospatial impact analysis of measures for irrigation infrastructure in Mali shows that even in conflict contexts where data collection on the ground is not possible due to the constraints of the security situation, development impacts (such as climate resilience-building) or even unintended effects (such as the escalation of conflicts due to the measures) can be reliably determined. However, rigorous analyses oriented to complex systems of objectives and using geo-referenced data are still rare in conflict-affected contexts.

At the same time, in conflict-affected contexts in particular, the repercussions of climate change put increasing pressure on development cooperation, which has to meet effectiveness criteria. The task now is to reduce this gap between rising relevance and a poor evidence base. At the national and international level, the number of rigorous evaluations using geospatial data to examine the consequences of climate change adaptation measures in contexts of fragile statehood should therefore be increased.

DEval has presented the availablen evidence in a variety of formats geared towards user-friendliness because the implementing development organisations must now put the evidence to good use and incorporate it at an early stage in the planning of measures.