Pastoralism is an important livelihood for hundreds of millions of people across the globe. In many parts of Africa, mobile pastoralism is the dominant livelihood of dryland regions. Moving their herds with the seasons is a critical part of that livelihood, giving pastoralist communities an edge against climate shocks and drought. Understanding these movements is key to understanding pastoral livelihoods. While international aid agencies are increasingly interested in serving pastoral communities, they often have little understanding of their mobility, making it difficult to assist them.
HSS has a long history of mapping transhumance movements across the Sahel using innovative technologies and tools that center the experience of pastoralists in the mapping process. New advancements in satellite imagery and artificial intelligence (AI) have facilitated the detection of herds and pastures in a cost-effective manner. When combined with local knowledge, the result is a set of powerful tools and data that reflect the spatial conditions of pastoral livelihoods.
Our methods rely on a mix of high tech and participatory methods. We rely on systems that build on local knowledge, rather than trying to replace it. In addition, our team has used satellite data, ground data and Tensorflow AI algorithms to track herds. In 2021, in partnership with SNV, we piloted the use of machine learning to automatically detect clusters of cattle concentrations with satellite imagery in central Mali. Some of the technologies we employ include:
- Participatory methods;
- AI pattern detection for finding herds of cattle;
- Very High Resolution (VHR) imagery;
- Satellite imagery of climatic data to better understand the relationship between transhumance and climate shocks.
Unlike many actors in the development space, we do not view this as an extractive data exercise. By mapping transhumance movements with pastoralist communities, we provide pastoral communities with powerful advocacy tools and maps that they have often used with governments and aid actors to call for greater recognition of grazing rights and resource allocation.
Our methods produce accurate and reliable maps that are unparalleled by using a blend of local knowledge and cutting edge technology.
