Satellites Show Impact of USAID Investments in Tigray, Ethiopia


Analysis by Ashutosh Limaye, NASA/SERVIR; text by Chris Hillbruner and Kiersten Johnson, USAID Bureau for Resilience and Food Security
In Tigray, Ethiopia, the production of rain-fed crops and livestock are important livelihoods, with more than 80 percent of the population engaged in agriculture. However, land holdings are small, soil is degraded, and the eastern part of the region regularly experiences drought. For the past 15 years, USAID’s Office of Food for Peace has supported community-level resilience-building activities in eastern Tigray, including the construction of water management infrastructure, through the Government of Ethiopia-led Productive Safety Net Program (PSNP). A local partner, the Relief Society of Tigray (REST) has implemented these programs for over a decade. A recent analysis used 34 years of satellite data (1984-2018) to explore changes in land productivity and water availability associated with the construction of this infrastructure at 36 sites in eastern Tigray. Key findings of this analysis include the following:
1. Project sites show increased water availability throughout the year
High-resolution satellite imagery suggests that vegetation conditions changed following water infrastructure interventions (Figure 1). To explore this change more systematically, researchers used the Normalized Difference Vegetation Index (NDVI), a satellite-derived measure of vegetative greenness and a proxy for water availability and land productivity. NDVI has been regularly collected over Ethiopia since the early 1980s and is therefore much more available than high resolution satellite imagery for historical analyses. A comparison of project-site NDVI before and after the construction of water infrastructure indicated an average increase in greenness of 15 percent. Increased greenness was found throughout the year, in both dry (Oct-Mar) and wet (Apr-Sep) seasons (Figure 2).
2. Increased water availability persists in drought years
Not only did greenness improve on average following the implementation of water management interventions, this improvement remained even during drought years. This finding is illustrated by Figure 3. The hatched green bar reflects the average level of greenness at project sites for all years prior to interventions, the solid green bar reflects the average level of greenness at project sites for all years after interventions. The hatched yellow bar reflects the average level of greenness at project sites for drought years prior to interventions and the solid yellow bar reflects the average level of greenness at project sites for all drought years after interventions.


3. In most sites, water infrastructure interventions likely strengthened resilience by improving water access during the main cropping season
The graphs in Figure 4 show average annual NDVI curves at three sites with water-related interventions. NDVI curves reflect the level of vegetation over a one year period, with higher values reflecting higher levels of vegetation. The average NDVI observed during the years prior to the intervention is plotted in red, while the average NDVI for the years after the intervention is plotted in blue.
In Tigray, the main rainy season, the Kiremt, occurs from June to September. Main season, or Meher, harvests then occur from October to January. In 25 of 36 study sites, average peak NDVI was higher after the construction of water infrastructure than before these interventions. As an example, see the circled area in charts 3 and 6 in Figure 4. These findings suggest that the construction of water infrastructure likely strengthened the resilience of farmers in these areas by improving water access during the region’s main cropping season, increasing crop yields.

4. In some areas, water infrastructure interventions have also strengthened resilience by allowing for a second crop to be planted
The dry season in Tigray typically runs from October to February/March, and crops are not typically planted during this period. However, in roughly one-third of project sites, the post-intervention pattern of greenness suggests that farm households had begun planting a second season of crops. As an example, see the circled areas of chart 10 in Figure 4. This additional crop production may contribute to household resilience by smoothing production and consumption over the course of the year, allowing these households to better withstand shocks and stresses.