Relationships analysis of land surface temperature with vegetation indicators and impervious surface fraction by fusing multi-temporal and multi-sensor remotely sensed data

Abstract

It is known that vegetation, and impervious surface are very important factors to affect the LST distribution in surface urban heat island (SUHI) analysis. However, the trade-off between temporal resolution and spatial resolution and/or the influence of cloud covering, make it difficult to obtain fine-scale spatial-temporal relationship analysis. To relieve these difficulties, this study employs multi-temporal and multi-sensor fusion methods for summer spatial-temporal relationships of Land surface temperature (LST) with normalized difference vegetation index (NDVI), vegetation fraction (VF) and impervious surface fraction (ISF) analysis on Wuhan city of China. Here, the correlation analysis was extended from two-dimensional to three-dimensional by using the continuous fused data (from 1988 to 2013). Our analysis indicates there is a strong negative relationship between LST and NDVI as well as VF, whereas the relationship between LST and ISF is obvious positive correlation. In addition, we also find that all these relationships are spatial-temporal steady. This result suggest that increasing impervious surface area means enhance LST, whereas increasing vegetation means weaken LST in summer, especially in the “warm edge” area. We believe the use of continuous long-term data weakened the interference of data quality and improve the reliability.