GSRIT

Geo-Spatial Risk Index Tool (GSRIT)

Monitoring Human–Nature Conflicts through Earth Observation, GeoAI and Local Knowledge

Japan · South Africa · Kenya · Botswana

Introduction

Human–nature conflicts are increasingly shaped by interacting pressures: climate variability, land-use change, habitat fragmentation, competition for water and food, settlement expansion, socio-economic conditions, and policy decisions. GSRIT brings these dimensions together within a geospatial framework to understand where conflict risks occur, why they change, and how evidence can support more targeted monitoring and intervention.

The project combines satellite Earth observation, climate information, field and socio-economic evidence, human–wildlife conflict (HWC) records, policy analysis, artificial intelligence, and GIS-based visualization. Its current focal landscapes include the Tsavo Conservation Area in Kenya, the Okavango Delta in Botswana, and Hluhluwe-iMfolozi National Park in South Africa.

At a glance

• 4-country collaboration: Japan, South Africa, Kenya and Botswana

• 3 current focal landscapes: Tsavo, Okavango Delta and Hluhluwe-iMfolozi

• 4 integrated work packages: drivers; HWC evidence; policy; GSRIT integration

What GSRIT integrates?

• Climate & ecological stress: Drought, rainfall variability, temperature change, vegetation stress, water scarcity and extreme events.

• Land-use & human pressure: Agricultural expansion, settlement growth, habitat loss and fragmentation, land conversion and resource competition.

• Human–wildlife conflict: Historical incident records, field evidence, surveys, media and unstructured text.

• Policy, governance & decision support: National and international policy analysis, institutional coordination, risk indicators, AI-assisted knowledge organization and GIS dashboards.

Current progress

GSRIT has progressed from identifying climate and land-use drivers toward integrated risk analysis. In Tsavo, the project has compiled a long historical HWC record and is linking climate and land-cover dynamics with conflict patterns. In Botswana, the team has analysed land-cover and ecological change in the Okavango Delta together with HWC patterns. AI research has advanced from knowledge graphs and retrieval-augmented generation toward case-based solution recommendation, while the project dashboard is being developed to combine outputs from the different work packages.

Featured case studies

• Tsavo Conservation Area, Kenya: A long-term HWC dataset, climate analysis, LULC mapping and conflict-hotspot analysis support development of the GSRIT risk framework.

• Okavango Delta, Botswana: Remote-sensing indicators, land-cover change and HWC records are being combined to examine how ecological conditions and human activity shape conflict patterns.

• Hluhluwe-iMfolozi, South Africa: The South African component contributes HWC, environmental and policy evidence to the comparative GSRIT framework. Detailed public case-study results should be added when verified project outputs are available.