Impact of Climate Change on Soil Properties of Farms in Southern Guinea Savanna Using Machine Learning and Remote Sensing Approach.
Loading...
Date
2026-02-02
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Nigerian Journal of Soil and Environmental Research.
Abstract
This study quantitatively assessed the impact of climate change on soil properties in Asomu Farm and
Malete Teaching and Research Farm using remote sensing and machine learning approaches. Landsat 8
satellite imagery from 2018, 2020, and 2024 was processed to derive the Normalized Difference Vegetation
Index (NDVI), Land Surface Temperature (LST), and Soil Moisture Index (SMI). The NDVI increased from
0.23 in 2018 to 0.42 in 2024, indicating improved vegetation density and health. Correspondingly, LST
decreased from 35.6°C in 2018 to 32.1°C in 2024, suggesting a reduction in surface heat stress over time.
SMI values revealed a consistent improvement in soil moisture levels, rising from 0.31 in 2018 to 0.47 in
2024. These changes suggest that the soil in both locations has become more resilient to moisture loss, likely
due to increased vegetation cover and possibly improved rainfall patterns. Machine learning algorithms
were applied to identify patterns and correlations between NDVI, LST, and SMI, enhancing the prediction of
soil condition trends under varying climatic scenarios. The study concludes that integrating remote sensing
with machine learning offers a reliable, data-driven approach to monitor the effects of climate change on
soil systems. The quantitative results provide a strong basis for recommending climate-smart agriculture
and adaptive land management practices in semi-arid regions of Nigeria.