Japan, South Africa, Kenya, Botswana

Case Studies

Geo-Spatial Risk Index Tool

South Africa

Research focus

The South African component is aligned with the same integrated questions used across the project:

Kenya

TSAVO CONSERVATION AREA, KENYA

Open Tsavo dashboard ↗

Why Tsavo?

Tsavo is a major conservation landscape where wildlife, pastoral and agricultural livelihoods, settlements, climate variability and limited water resources interact. This makes it an important setting for studying how ecological stress and human pressures translate into HWC risk.

Data foundation

The GSRIT team compiled and digitized historical HWC records covering 1995–2023 (28 years). The project presentation reports more than 37,900 HWC incidents in this historical dataset. Recorded conflict types include crop raiding, human attacks, livestock predation, property damage and accidents.

Climate and land-cover analysis

The Tsavo work combines HWC records with precipitation, temperature, vegetation and land-cover information. The current climate analysis applies SPI, VCI, elevation and Random Forest modelling to classify drought and wetness conditions from 2000–2024. The project also uses Landsat and Sentinel-2 to map changes in vegetation, agriculture, settlements, water bodies and bare land.

Preliminary project findings

Botswana

OKAVANGO DELTA, BOTSWANA

A dynamic human–wildlife landscape

The Okavango Delta is a biodiversity-rich inland delta shaped by strong seasonal hydrological and climatic variability. Human–wildlife conflict is unevenly distributed across Botswana, with important hotspots in northern districts and around protected areas and wildlife corridors.

Land-use and ecological change

GSRIT has used Landsat imagery and other datasets to examine land-use/land-cover change in the Okavango Delta, including classifications for 2001, 2013 and 2023. Project analysis indicates growth of agricultural land and settlements, even where their overall spatial share remains relatively small.

The Botswana component also evaluates ecological indicators including NDVI, MNDWI, land-surface temperature and the Remote Sensing Ecological Index (RSEI), together with HWC records.

Preliminary project observations