Analysis of spatial-temporal pattern of Land Surface Temperature (LST) due to NDVI and elevation in Ilorin, Nigeria

Sammanfattning: The rate of global urbanization in the 21st century has been unprecedented especially in the African and the Asian continents. Urbanization typically causes a removal of natural vegetation and replacement of same with impervious and non-evaporative urban materials with high heat capacity and low solar reflectivity. The alteration of the natural land cover due to urbanization causes a change in the thermal properties of the urban landscape, with the effect that the urban land surface typically has a relatively higher thermal capacity than that of the surrounding rural areas; a phenomenon known as Urban Heat Island (UHI). Studies have shown that, in the last few decades, Ilorin city has undergone significant changes in landscape, population structure, and urban form and the city has witnessed a rise in urban temperature. Similarly, Nigerian Meteorological Agency (NIMET) and epidemiologists have warned of possible outbreak of some diseases that thrive in hot weather such as meningitis, tuberculosis, malaria, dengue fever, etc. Given this scenario, UHI effects is estimated to affect a large number of persons in the city. This study investigates to what extent the observed Land Surface Temperature (LST) could be accounted for by vegetation cover measured via Normalized Difference Vegetation Index (NDVI) in the study area from 2000 to 2016. The spatial pattern and trend of LST, and the magnitude of UHI within the period was also estimated. Result shows a statistically significant clustering of LST values in all the investigated epochs with Global Moran’s I index showing values greater than 0.6 and z-scores greater than 220 in all cases at p < 0.001. Analysis of regression reveals that NDVI explained between 50% - 71% of the variation in LST between 2000 and 2002, and NDVI increasingly diminished in importance as an explanatory variable for LST in 2016, where only 29% - 49% explanation was provided by NDVI. For all the seasons and years, the coefficients of the regression analysis results for NDVI in the study area are negative indicating that NDVI is inversely related to LST. Research result shows a presence of UHI with an intensity ranging between 0.2OC to 4.6OC, with peaks in the wet seasons and tending towards a decreasing trend.