Decision-level integration window strategy in satellite imagery-derived land surface temperature disaggregation

The purpose of this study is to present a new approach for satellite imagery-derived Land Surface Temperature (LST) disaggregation based on a decision level integration of various disaggregation strategies. Firstly, common disaggregation models including Global Window Strategy (GWS), Regular Local Window Strategy (RLWS), Object-based Window Strategy (OWS), and Conceptual Window Strategy (CWS) were used for LST disaggregation. Secondly, the Disaggregated LST (DLST) obtained from these strategies were integrated using the Decision-level Integration Window Strategy (DIWS). Finally, the efficiency of different strategies in LST disaggregation was evaluated using actual LST (ALST) maps and Actual Soil Temperature (AST) based on Pearson correlation coefficient (r) and Root Mean Square Error (RMSE). The mean r (RMSE) between ALST and DLST obtained from GWS, CWS, OWS, RLWS, and DIWS were 0.75 (1.87), 0.76 (1.90), 0.76 (1.80), 0.82 (1.38), and 0.89 (1.09 degrees C), respectively. The RMSE between AST and DLST obtained from these strategies were 3.28, 3.17, 2.87, 2.43, and 2.10 degrees C, respectively. The results showed that the effectiveness of DIWS in LST disaggregation was higher than other strategies.

Automated summary

The study presents a new approach for satellite imagery-derived Land Surface Temperature (LST) disaggregation, achieving higher effectiveness in LST disaggregation by integrating different strategies.