Quantifying spatial variability of crop water use by combining eddy covariance observations with high-resolution UAV-based remote sensing

Applications of precision agriculture enable water management decisions to be made at increasingly fine resolutionssmaller than which we can routinely estimate the water use from crops. Direct measurement of crop evapotranspiration (ET) using the eddy covariance technique can provide continuous flux observations but cannot provide insight at smaller scales than the measurement footprint (~100-200m). On the other hand, remote sensing techniques based on the energy balance approach are capable of quantifying spatial patterns of ET at high resolutions but are limited to particular instances in time and cannot provide the temporal information required. In this research, we develop and evaluate a method to spatially disaggregate continuous ET measurements from eddy covariance systems. The approach is based on developing high resolution spatial maps of ET derived from a combination of UAV-thermal remote sensing energy balance modeling that is parameterized using UAV-LiDAR to measure crop canopy characteristics (leaf area index, canopy height, and vegetation fraction). Through consideration of the eddy covariance flux footprint, and the remotely-sensed spatial detail, temporally continuous spatial estimates of ET were developed. The application of this method is demonstrated for barley grown in the Canadian prairies under strongly heterogeneous field conditions.