Vegetation Indices as Indicators of Drought, a Remote Sensing Perspective: a Case Study of Narok County

Abstract

Vegetation cover is essential in determining the health of ecosystems and critical in planning and management of environmental and land resources. This study aimed at establishing the correlation between vegetation cover and drought indicators in Narok County. Using the Enhanced Vegetation Index (EVI), the Normalized Difference Vegetation Index (NDVI), Soil Adjusted Vegetation Index (SAVI), and the Atmospherically Resistant Vegetation Index (ARVI), areas with potential vegetation cover were delineated. Within the same timestamp, the following drought indices were computed; the Temperature Condition Index (TCI), Vegetation Condition Index (VCI), Standardized Precipitation Index (SPI), Land Surface Temperature (LST), Normalized Difference Water Index (NDWI) and Soil Moisture Index (SMI). A correlation analysis was done to establish the relationship that exists between vegetation cover and drought indicators over a period of 35 years (1987-2022). The findings revealed that SMI, VCI, and NDWI exhibited a correlation with the vegetation indices. However, LST, TCI, and SPI calculated over a period of 1 month (SPI-1) showed no significant correlation with the vegetation indices. Furthermore, a drought model was developed based on these findings, utilizing regression analysis techniques. The model exhibited strong performance, with an R-squared value of 0.86, indicating a high level of accuracy in predicting drought conditions. The validation of the model using independent data confirmed its reliability and robustness. The study will benefit a wide range of stakeholders, including farmers, government agencies, environmentalists, and researchers. The advantages will encompass enhanced agricultural output, heightened efficacy in disaster management, refined conservation approaches, and the progression of remote sensing as a discipline. This study recommends further research and validation studies to strengthen the understanding and applicability of vegetation indices as indicators of drought. Conducting field studies, comparing remote sensing data with ground observations, and evaluating the performance of vegetation indices under different climatic and ecological conditions will enhance the reliability and confidence in their use for drought monitoring and prediction