Pastoralism in India

The atlas grew out of an effort to map something that, in India, had been known mostly through anecdote and scattered evidence: the seasonal movement of herders with their animals. We work region by region across the three landscapes where mobile herding is concentrated—the Himalaya, the arid west, and the semi-arid Deccan. We cannot sample populations in an attempt to generate maps that are representative of pastoralist distribution across a region. Instead, we have used a combination of opportunistic snowball sampling and key informant based purposive sampling to identify and interview herders, sometimes snowballing from one interview to the next, and at other times using key informants to identify pastoralist households.

Our aim is to ensure that we cover all pastoral communities within a state and, as far as possible, all geographies used by pastoralists. The resulting maps therefore reflect the intensity and reach of our sampling, rather than providing an accurate representation of the distribution or density of pastoralists across a geography. The Atlas, provides a sense of the scale and spatial spread of pastoralism. We see this as a first step towards understanding the extent and diversity of pastoralism within a region.

Within each community we document the routes herders travel and the sites that structure their year—grazing areas, night halts, and water points; places for veterinary care, shearing, and milking; the markets and points of sale for pastoral produce; tenure arrangements and access; issues faced and the herder settlements themselves. The atlas is built with herders rather than only about them. In semi-structured interviews, they identify the routes they travel and the places that matter along them, and this knowledge is recorded on the spot.

At first, locations were noted by name on paper based questionnaires and georeferenced after the interview, which produced entry errors, mismatches with mapped place-names, and duplicate records that were tedious to reconcile. We then built a digital tool on the Open Data Kit (ODK) protocol to move beyond paper forms and collect information at scale, but because locations were still georeferenced after the fact, those inconsistencies persisted. We now use Mergin Maps, an open-source application on which surveyors place each location directly onto a base map on a tablet, and which works offline in remote areas without a signal. Running alongside the mapping, a structured questionnaire captures the economics of herding at the household level—income, expenditure, herd composition, and animal losses.

Data cleaning required place-names be standardised against mapped entries, and datasets are cross-referenced to catch errors, duplicates, and gaps. The cleaned datasets are then analysed in QGIS and ArcGIS alongside land cover, elevation, and administrative, forest, revenue, and protected-area boundaries. These maps are read alongside the interviews, observations, and informal conversations gathered during fieldwork, which supply the context and narratives the maps alone cannot. To understand how pastoralism is changing, a separate strand of work looks closely at charting historical routes alongside current ones to understand how migration patterns have changed over time. Because pastoral movement is seasonal and negotiated, and shifts from one year to the next, we treat each route as a record of a practice at a moment in time rather than a fixed line on the ground.