Location: Hydrology and Remote Sensing LaboratoryTitle: Effect of the revisit interval on the accuracy of remote sensing-based estimates of evapotranspiration at field scales
Submitted to: Hydrology and Earth System Sciences
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/7/2016
Publication Date: 1/5/2017
Citation: Alfieri, J.G., Anderson, M.C., Kustas, W.P. 2017. Effect of the revisit interval on the accuracy of remote sensing-based estimates of evapotranspiration at field scales. Hydrology and Earth System Sciences. 21:83-98.
Interpretive Summary: Satellites imagery can be used to estimate evapotranspiration (ET) over large areas. While this information is critical for water management in agricultural systems, among innumerable other activities, many satellites pass over a target area infrequently. To fill the data gaps, “temporal upscaling” methods are commonly used, but these methods can introduce errors into the ET estimates. Using surface-based data as a proxy, several upscaling methods were evaluated to characterize the accuracy of the ET estimates with increasing return interval (gap size). The analyses indicated the effective decreased rapidly with increasing return interval. It also showed that errors in the estimated ET were largest for landscapes, such as forest and cropland, which typically exhibit high ET. Using 20% as a threshold, the study found that the longest return interval that can reliably estimated ET is five days. These results are useful for climate and weather forecasters and watershed managers.
Technical Abstract: Accurate spatially distributed estimates of evapotranspiration (ET) derived from remotely sensed data are critical to a broad range of practical and operational applications. However, due to lengthy return intervals and cloud cover, data acquisition is not continuous over time. To fill the data gaps, interpolation methods that take advantage of the relationship between ET and other environmental properties that can be continuously monitored are often used. This study sought to evaluate the impacts of this approach, which is commonly referred to as temporal upscaling, on the accuracy of the flux estimates. Using data collected at 20 Ameriflux sites distributed throughout the contiguous United States and representing 4 distinct land cover types (cropland, grassland, forest, and open canopy) as a proxy for satellite imagery, this study compared flux estimate derived using 5 different reference quantities (incident solar radiation, net radiation, available energy, reference ET, and equilibrium latent heat flux) and 3 different interpolation methods (linear, cubic spline, and hermite spline). Not only did the analyses find that the autocorrelation, i.e. persistence, of all of the reference quantities was short, it also found that those land cover types with the greatest ET also exhibited the least persistence. This carries over to the error associated with both the various scaled quantities and flux estimates. In terms of both the root mean square error (RMSE) and mean absolute error (MAE), the errors increased rapidly with increasing return interval following a logarithmic relationship. Again, those land cover types with the greatest error showed the largest errors. Moreover, using a threshold of 20% relative error, this study indicates that the maximum return interval that allows for accurate ET estimates is 5 days. It also found that the spline interpolation methods performed erratically for long return intervals and should be avoided.