Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Hydrology and Remote Sensing Laboratory » Research » Publications at this Location » Publication #329121
Title: Reconstructing daily clear-sky land surface temperature for cloudy regions from MODIS data

Author
item SUN, L. - Collaborator
item CHEN, ZHONGXIN - Chinese Academy Of Agricultural Sciences
item Gao, Feng
item Anderson, Martha
item SONG, LISHENG - Beijing Normal University
item WANG, LIMIN - Chinese Academy Of Agricultural Sciences
item HU, BO - Collaborator
item YANG, Y. - Collaborator

Submitted to: IEEE Transactions on Geoscience and Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/24/2017
Publication Date: 4/29/2017
Citation: Sun, L., Chen, Z., Gao, F.N., Anderson, M.C., Song, L., Wang, L., Hu, B., Yang, Y. 2017. Reconstructing daily clear-sky land surface temperature for cloudy regions from MODIS data. IEEE Transactions on Geoscience and Remote Sensing. 105:10-20. doi:10.1016/j.cageo.2017.04.007.

Interpretive Summary: Landsat surface temperature (LST) is a key parameter for mapping land surface energy fluxes and evapotranspiration, and monitoring drought. Landsat surface temperature can be retrieved from thermal infrared band imagery. However, thermal infrared bands are affected by clouds which lead to more than half LST. This paper presents a new daily land surface temperature reconstruction technique. The reconstructed Moderate Resolution Imaging Spectrometer (MODIS) LST were validated using ground measurements as well as MODIS LST under clear-sky conditions. In general, the reconstructed LST agrees with ground measurements and captures spatial details of LST. This study presents a new approach for producing gap-free LST imagery and has potential uses in crop water use and drought monitoring that will greatly benefit the USDA National Agricultural Statistics Service (NASS) and Foreign Agricultural Service (FAS) for more accurate yield assessments and predictions.

Technical Abstract: Land surface temperature (LST) is a critical parameter in environmental studies and resource management. The MODIS LST data product has been widely used in various studies, such as drought monitoring, evapotranspiration mapping, soil moisture estimation and forest fire detection. However, cloud contamination affects thermal band observations and will lead to inconsistent LST results. In this study, we present a new Remotely Sensed DAily land Surface Temperature reconstruction (RSDAST) model that recovers LST for pixels covered by cloud or cloud shadows using clear-sky neighboring pixels from nearby dates. The reconstructed LST was validated using field measurements from the HiWATER experiment and LST measurements from 15 meteorological stations administered by the China Meteorological Administration (CMA) in Northwest China. For the HiWATER sites, the reconstructed MODIS LST showed correlation ® of 0.93, bias of -1.94K and Root Mean Squared Error (RMSE) of 4.83K. Most of the low values of LST resulting from the change of either air temperature or soil moisture at two HiWATER sites were correctly modeled by RSDAST. The validation based on the 15 CMA stations shows that reconstructed LST has a good correlation with daily maximum and averaged LST. In addition, the reconstructed LST were also compared to the original LST images. Results show a better accuracy for flat areas with R of 0.85-0.94, bias of -0.05-0.09K, and RMSE of 0.95-1.08K, comparing to mountain areas with R of 0.92-0.93, bias of -0.54-1.01K, and RMSE of 1.38-2.24K. The monthly reconstructed LST reduced the bias from 3.27K to -0.53K and RMSE from 4.75K to 1.98K compared to the original clear-sky MODIS LST. Accordingly, the resulting monthly, yearly, and multi-year averaged LSTs are 1-10K smaller than the averaged values of clear-sky LST from the original MODIS LST product. The reconstructed areas show spatial and temporal patterns that are consistent with the clear neighbor areas. Our approach shows a great potential to reconstruct LST under cloudy conditions and provides consistent daily LST which are critical for daily evapotranspiration mapping and drought monitoring.