Multi-temporal characterization of land surface temperature and its relationships with normalized difference vegetation index and soil moisture content in the Yellow River Delta, China

Land surface temperature (LST) is an important indicator of resource and environment changes in coastal wetlands and exhibits considerable spatiotemporal variations under multiple influences of intensive landesea interaction and various human activities. Revealing the multi-temporal characteristics of LST and its relationships with normalized difference vegetation index (NDVI) and soil moisture content (MC) in coastal wetlands is important for identifying the LST variations under complex factors, the heat effects of different human exploitations, and the corresponding changes in vegetation and soil water conditions. The Yellow River Delta, a typical and important coastal wetland in China, was selected to demonstrate the study. Remote sensing and field investigation served as the data source, and the LST spatiotemporal patterns and their intrinsic correlations with NDVI and MC in the last 30 years and across different seasons were analyzed. Results indicated that LST spatial distributions varied in different years and seasons. In 1987, LST was low along the Yellow River and high in alongshore areas; in 2016, LST was distinctly low in most of the alongshore areas and high in urban and industrial areas. Across seasons, LST differences between alongshore and inner areas were more distinct in spring that in the other seasons, and LST in urban and industrial areas was especially higher than that in other areas in summer. Among different land covers, salt pond generated cool island effect, whereas urban and industrial areas generated heat island effect. Significant correlations were observed between LST and NDVI at regional scale and among LST, NDVI, and MC at site scale, and their relationships varied in different land cover types across different times. The changes in shorelines, land covers, and vegetation and soil conditions jointly determined the LST spatiotemporal patterns, and human activity has been the main driving factor. (c) 2020 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).