Skip to main content
Knowledge4Policy
Knowledge for policy
Supporting policy with scientific evidence

We mobilise people and resources to create, curate, make sense of and use knowledge to inform policymaking across Europe.

Enhancing climate resilience: A data-driven north rift weather prediction system for real-time forecasting and agricultural decision support

Global Food and Nutrition Security

Highlights:

  • XGBoost Boosts NRWPS: NDVI and BSI integration improves predictive accuracy.
  • Comprehensive Monitoring: NRWPS integrates environmental data for land management.
  • Climate Impacts Vegetation: Post-2000 temperature rise causes loss; rainfall aids recovery.
  • Land Degradation Effect: Rising BSI, declining NDVI indicate soil exposure, vegetation loss.
  • Sustainability & Resilience: NRWPS supports adaptation, food security, and conservation.

Abstract:

This study presents the development and integration of predictive models for the Normalized Difference Vegetation Index (NDVI) and Bare Soil Index (BSI) using the XGBoost algorithm within the North Rift Weather Prediction System (NRWPS) to enhance ecosystem monitoring in Kenya's North Rift region. Trained on a comprehensive dataset spanning 1995 to 2020, which includes precipitation (from the Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS)), temperature (TerraClimate), historical NDVI (Landsat 4–5 Thematic Mapper (from 1995 to 2013) and Landsat 7 Enhanced Thematic Mapper plus (ETM+) (from 2014 to 2020)), and BSI (SoilGrids) data, the models effectively capture the complex relationships between environmental factors and vegetation health. The BSI model achieved an MSE of 0.029, an MAE of 0.019, and an R-squared score of 0.93, while the NDVI model yielded an MSE of 0.002, an MAE of 0.024, and an R-squared score of 0.945. These results demonstrate the models' strong predictive accuracy, enabling precise assessments of vegetation health and bare soil exposure. By analyzing temporal variations in vegetation health and land degradation from 1995 to 2020, the study identifies a significant inverse relationship between NDVI and BSI, where increasing bare soil exposure corresponds to declining vegetation health. The analysis also reveals that climatic factors particularly temperature (minimum and maximum) and precipitation play a critical role in shaping these trends, with high temperatures after 2000 associated with reduced NDVI, while regions with higher precipitation show healthier vegetation and lower BSI. The successful development of the NRWPS model provides significant opportunities for informing land management strategies, conservation efforts, and agricultural practices, enabling data-driven decision-making. Moreover, its integration into larger decision support systems allows for proactive interventions to mitigate land degradation and climate change stressors. This study emphasizes the importance of sustainable land-use practices and climate adaptation strategies to preserve vegetation health and manage ecosystem vulnerabilities effectively in the wake of regional climate change with the North Rift region most affected.