H026-07
Added Value of Sentinel-2 and VIIRS Data in Mapping Daily Evapotranspiration at Sub-field Scales

Monday, 7 December 2020: 20:54
Virtual
Jie Xue1, Martha B. Anderson2, Feng Gao3, Christopher Hain4, Yun Yang5, Yang Yang3 and Kyle Knipper6, (1)USDA-ARS Hydrology and Remote Sensing Laboratory, Beltsville, MD, United States, (2)USDA ARS, Hydrology and Remote Sensing Laboratory, Beltsville, MD, United States, (3)USDA-ARS, Hydrology and Remote Sensing Laboratory, Beltsville, United States, (4)NASA Marshall Space Flight Center, Huntsville, AL, United States, (5)University of Maryland College Park, ESSIC, College Park, MD, United States, (6)U. S. Department of Agriculture, Agricultural Research Service, Hydrology and Remote Sensing Laboratory, Beltsville, MD, United States
Recorded Presentation
Abstract:
Evapotranspiration (ET) plays a crucial role in describing crop water use and soil moisture conditions. Accurate monitoring of frequent ET at sub-field scales can be used to identify crop stress and increase crop water use efficiency and production. Land surface temperature (LST) retrieved from thermal infrared (TIR) remote sensing provides an important input for ET estimates. Landsat imaging system with both multispectral and thermal data on board has been widely used in 30-m ET mapping, and fused with MODIS retrievals can generate daily ET at 30-m scales with reasonable accuracy. However, the fusion accuracy may be limited by the relatively low temporal revisit of Landsat data (8 days for combined Landsat 7 and 8), especially in heterogeneous areas with rapid changes, where the changes occur between Landsat overpass dates at scales smaller than a MODIS pixel. The Sentinel-2 (S2) satellite, as a good complement to the Landsat system, provides surface reflectance data at 10-20 m spatial resolution and 5-day revisit period but no TIR data; on the other hand, the Visible Infrared Imaging Radiometer Suite (VIIRS) provides TIR data on a near-daily basis with 375-m resolution, which can be refined through thermal sharpening using S2 reflectances. This study assesses the value added by incorporating S2-sharpened VIIRS ET retrievals into the 30-m Landsat ET timeseries as a means of increasing temporal sampling. The valued added by including VIIRS-S2 is assessed in comparison with flux tower observations collected from various U.S. agricultural sites. In particular, we evaluate the performance of VIIRS-S2 ET estimates as a function of VIIRS view angle, overpass time difference between S2 and VIIRS, and cloud masking approach. Results from this work demonstrate the improved capability of increased temporal frequency of ET in capturing rapid changes, and the incorporation of S2 and VIIRS data into the ET fusion system will facilitate the development of a prototype multi-source system that generates high spatiotemporal resolution ET products.
<< Previous Abstract | Next Abstract >>