Field scale crop evapotranspiration monitoring using satellite-driven modeling
Abstract
Evapotranspiration (ET) is a key variable linking the global water and energy cycles, while accurate ET estimates are crucial for understanding cropland water use and drought impacts. Satellite remote sensing provides spatially and temporally continuous information that can be used for ET estimation globally. However, reliable methods for delineating finer field-scale ET patterns in croplands is lacking. In this investigation we modified the MOD16A2 algorithms to better represent croplands by calibrating the model parameters using FLUXNET tower site observations from diverse crop types; incorporating soil moisture as an additional control on soil evaporation and canopy stomatal conductance; and using finer scale satellite vegetation inputs to derive ET at 30m resolution. The 30m 8-day EVI record fused by Landsat 5 and 7 with MODIS data was used to distinguish canopy cover and bare soil fractions for estimating ET at similar scale over all continental US (CONUS) croplands. Soil moisture inputs to the model were obtained from landscape (1-3 km resolution) level retrievals of surface soil moisture conditions from combined SMAP and Sentinel-1 active/passive microwave observations. The model results were evaluated against independent tower ET observations for different crop types and against baseline 500m resolution observations from the MODIS operational (v006) global ET product. The resulting 30m ET estimates corresponded favorably with the MODIS 500m baseline product (0.53<R2<0.86) and against the tower ET observations (0.48<R2<0.81; 0.47<RMSE<1.52 mm/d). The 30m ET results were also significantly improved over the global ET baseline, indicating enhanced utility of the finer scale ET records for complex croplands. The incorporation of soil moisture improved model sensitivity to water supply related constraints affecting ET, while distinguishing impacts from irrigation and active water management. Spatial and temporal patterns of cropland ET and water use efficiency (WUE) were delineated, revealing regional climate impacts on agricultural water use. Our results provide improved spatiotemporal information on cropland water use that are suitable for operational applications and agricultural water management.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.H43G2536H
- Keywords:
-
- 1817 Extreme events;
- HYDROLOGYDE: 1855 Remote sensing;
- HYDROLOGYDE: 1880 Water management;
- HYDROLOGYDE: 1895 Instruments and techniques: monitoring;
- HYDROLOGY