Publications

You, WJ; Huang, CL; Hou, JL; Zhang, Y; Dou, P; Han, WX (2024). Reconstruction of MODIS LST Under Cloudy Conditions by Integrating Himawari-8 and AMSR-2 Data Through Deep Forest Method. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 62, 4407017.

Abstract
Land surface temperature (LST) plays a crucial role in Earth's energy balance and ecosystems. Various gap-filling methods have been developed to reconstruct seamless LST datasets to deal with the effect of data gaps caused by cloud cover; however, existing studies mainly focus on LST reconstruction under clear-sky conditions, rather than generating actual cloud-impacted LST. This study treats MODIS cloud-free pixels as known sample points. The deep forest (DF) algorithm is employed to establish a nonlinear relationship model between Himawari-8 cumulative downward surface shortwave radiation (DSSR), AMSR2 brightness temperature (TB) data, and other influencing factors on the sample points, as well as LST. This model is applied to cloud-covered pixels to obtain the LST of the underlying pixels, thereby reconstructing the real MODIS LST under the cloud over the Yellow River source region. The feasibility of this approach lies in the fact that cumulative DSSR incorporates the impact of cloud coverage on incoming solar radiation, and there exists a correlation between AMSR2 TB data and LST. The reconstruction results for January, April, July, and October of 2021 were validated against in situ 0 cm LST measurements from five meteorological stations. The results show that the reconstructed LST exhibits high consistency with in situ measurements, with R-2 hof 0.86, bias of 0.62K, and RMSE of 4.48K. The results demonstrate the effectiveness of using DSSR and microwave data in LST reconstruction, accurately representing actual cloud-impacted LST.

DOI:
10.1109/TGRS.2024.3388409

ISSN:
1558-0644