An analysis of the influence of preceding precipitation on the signature of Baltimore’s surface urban heat island

Abstract

Surface Urban Heat Islands (SUHI) are urban areas where temperatures exceed that of their rural counterparts. SUHIs are caused by alterations of the land surface by humans, generally from natural vegetated surfaces to urban surfaces (Zhou et al., 2014). SUHIs have been studied using multiple methods (Voogt & Oke, 2003). One method involves collecting satellite images using specific criteria. This includes percent cloud cover (CC), time of day, season, wind speed, and synoptic setup. These studies overlook preceding precipitation and its impact on the overall SUHI analysis. This study collected eight satellite images of Baltimore County, Maryland from 1994 to 2016. Four satellite scenes had preceding precipitation occur within three days of satellite acquisition. Four satellite scenes had no preceding precipitation. Land surface temperature (LST), Normalized Difference Vegetation Index (NDVI), Normalized Difference Built-Up Index (NDBI), and Normalized Difference Water Index (NDWI) were calculated for each scene. LST was regressed on NDVI, NDBI, and NDWI. Both NDBI and NDWI were positively correlated with LST. NDVI had a negative correlation. On average the explanatory power of NDVI, NDBI, and NDWI decreased for scenes with preceding precipitation. The Getis-Ord Gi* statistic showed no difference in spatial clustering of hot and cold LST for all images.