Assessment of Landsat Harmonized sUAS Reflectance Products Using Point Spread Function (PSF) on Vegetation Indices (VIs) and Evapotranspiration (ET) Using the Two-Source Energy Balance (TSEB) Model

One of the main goals in remote sensing is the agreement of remotely sensed information regardless of pixel resolution or sensor. With the availability of Landsat Surface Reflectance (LSR) product and Landsat point spread function (PSF), it is possible to upscale the high-resolution imagery captured by small Unmanned Aerial Systems (sUAS) at Landsat resolution for comparison and bias correction. The bias correction can then be used to adjust the sUAS imagery, leading to a new high-resolution product called "Landsat Harmonized sUAS imagery." In this study, evaluation of Landsat harmonized and upscaled sUAS optical imagery is performed on vegetation indices (VIs) and evapotranspiration (ET) estimates. Specifically, we evaluate the harmonization of different optical sUAS sensors using both PSF and traditional arithmetic averaging methods for upscaling, which includes an iterative approach for reflectance difference minimization. The original and harmonized results are then evaluated for NDVI and for estimated ET using the two-source energy balance (TSEB) model in comparison to eddy covariance measurements. The sUAS imagery, captured at Landsat overpass times by the Utah State University AggieAir sUAS Program, was collected as part of the Grape Remote sensing Atmospheric Profile and Evapotranspiration eXperiment (GRAPEX) field campaigns over a California vineyard. Our results reveal (1) the performance of different sUAS sensors versus Landsat (2) the importance of PSF versus arithmetic averaging, and (3) the impact of the Landsat Harmonized sUAS approach on VIs and ET.