Cultivation in grazing reserves

Using a 3-period timescan model (developed by Boudinaud, 2019), I mapped cropland expansion within the pastoral reserve of Sideradougou. The blue pixels represent active cropland during the 2020 growing season.

Grazing reserves in Burkina Faso are being devoured by agriculture. While these areas are typically protected by state decree and reserved for herds, the need for increased cultivation places enormous pressure on these zones. The dominant narrative of “farmer-herder” conflict in the Sahel often places the blame on mobile herders for encroaching on sedentary farmers’ fields. But it’s far more complex than that as we also see farmers moving into grazing lands.

By creating an NDVI time series during the growing season, this method isolates cropland by identifying pixels corresponding to a spectral signature common to Sahelian agriculture. This was performed using Google Earth Engine.

STAMP: A call center for pastoralists

One of my current projects is the “Sustainable Technology Adaptation for Malian Pastoralists” initiative. It consists of a call center based in Bamako, Mali that fields calls from pastoralists in Gao, Tombouctou and Mopti who seek information on pasture, water and market prices. The call center is powered by an open platform that uses field data and satellite imagery to gather this data. You can actually use the platform yourself.

As a consultant, I work mostly on the interface, building the GIS products that go into it. I also work directly with pastoralists and other project staff to test the products, facilitate the collection of field data and provide GIS trainings as needed.

Screenshot of the STAMP interface

More information on the product can be found here.

Surface Water in Mauritania (2017)

Client/Employer: ACF

During the 2017/2018 Sahelian drought, surface water in pastoral zones was acutely scarce, worsening conditions for livestock herds who depended on lakes and ponds for water.

Using 1km imagery from Proba-V, an “Accessibility” anomaly was calculated (visible in the lower-right corner of the infographic). This showed the presence of surface water for the current year, compared with the average area of surface water for the same zone between the 1998-2017 period. Areas shaded in blue were in excess of the average and red areas in deficit. The darkest red areas were completely absent of surface water.

This was then complemented with Sentinel 2 imagery taken from Sentinel Hub’s EO Browser to show the difference between 2016 and 2017.

Rift Valley Fever Monitoring (2016)

Client/employer: ACF, 2016

In partnership with the Minsitry of Health of Niger, my team built an SMS-based monitoring service to track cases of Rift Valley Fever in Tahoua, Niger. The data was used by the Ministry of Health to quickly track and respond to suspected outbreaks and reduce the number of costly field visits. A network of community respondents (veterinarians, livestock production technicians) regularly provided information on local outbreaks via SMS. The responses were pushed to a cloud server, where GPS coordinates were attached, allowing for the cases to be easily mapped.