Urban heat island and ecological condition modeling using thermal remote sensing in Tigray–Northern Ethiopia

The detection of Urban Thermal Field Variance Index and Ecological condition was generally conducted based on the thermal remote sensing. This study aims to model the Land Surface Temperature (LST), Urban Heat Island (UHI), Urban Heat Hotspot (UHS), Normalized Difference Built–Up Index (NDBI), Enhanced Built–Up and Bareness Index (EBBI), Ecological Evaluation Index (EEI), Urban Thermal Field Variance Index (UTFVI) of Mekelle city, Tigray–Northern Ethiopia. The Landsat 8 Operational Land Imager (OLI) and Thermal InfraRed sensor (TIRS–1) bands were used. The results indicated that the mean LST was highest (38.69 °C) in the Semen sub–city. A total area of 33.4 Km2 (30.56%) was classified as UHI dominantly located in the Semen sub–city. Moreover, about 38 Km2 (34.61%) was characterized with the worst EEI. The UHS zones were covering a total area of 3.03 Km2 (2.8%) and it is mainly concentrated in the Semen sub–city. In addition, the statistical relationship between LST and NDBI is negative and moderately strong (r = – 0.51, p < 0.001) due to the fact that most of the built–up area in the city is mixed with vegetation cover that cools the environment. We also found that an R2 of 0.997 (p < 0.001) in all UTFVI indices. The multiple regression model result indicated that Pv, NDBI, and EBBI control LST negatively, whereas the LST, UTFVI, EEI and NDVI influences LST positively.