Spatial assessment of the effects of in situ and neighbourhood factors on urban land surface temperature mitigation in a rapidly developing region

Abstract

This study investigates the effect of a set of in situ and neighbourhood factors on urban land surface temperature (LST) in the city of Sari, northern Iran. 200 urban pixels were randomly selected to investigate how LST values in these points linked to their corresponding normalized difference built-up index (NDBI) values as an in situ factor and the percentage of green pixels (%Green), the mean normalized difference vegetation index (M-NDVI), mean patch size (MPS), mean nearest neighbour distance (MNN), and mean shape index (MSI) of green patches located within a set of circular rings around the sample points as neighbourhood factors. The results showed that M-NDVI in the 150-m-radius ring had the highest impact (r = –0.63), and %Green had the longest distance impact (r = 0.24) on mitigating urban LST. MSI and MNN showed the highest association with LST within 300-m to 1000-m-radius rings with the highest r values of about −0.34 and 0.34, respectively. These results indicate that maximum integration of green cover patches with the least MNN value, and the highest degree of MSI is the most effective approach to alleviate urban LST; however, small but dense green cover patches can be suggested as an alternative approach for areas with restricted availability of open spaces.

Keywords

• land use
• land cover
• land surface temperature
• landscape metrics
• iran