Using LiDAR data to spatially scale and examine the accuracy of evapotranspiration estimates in the Western Boreal Plains, Canada

In Canada a large portion of the boreal forest is comprised of the mosaic landscape of the Western Boreal Plains (WBP)—a region characterized by consistent water deficit conditions. Because potential evapotranspiration (PET) generally exceeds precipitation annually, evapotranspiration (ET) is the primary driver of the hydrologic balance in the WBP. Owing to this, the WBP is a hydrologically sensitive region, and future changes to the climate will significantly impact the region's water balance. This region is also an economic hub for the country's natural resource extraction (e.g. forestry, conventional oil and gas, and oilsands development) creating significant disturbance on the landscape. However, this sensitive landscape is characterized by sparse measurement stations, making it challenging to gain information and drive models outside of the tower-footprint scale. This scenario makes remote sensing an ideal method of acquiring spatial information pertaining to the hydro-meteorology of this region—though further investigation is needed to determine the most appropriate resolution at which remote sensing data is best collected. At a study site representative of the WBP, this research uses high-resolution (1m x 1m) Light Detection and Ranging (LiDAR) data of the vegetation structure as an input to the Penman-Monteith model to provide a spatially explicit estimate of ET. The accuracy of high-resolution spatially explicit and spatially static vegetation parameters are examined relative to eddy covariance (EC) validation data. Subsequently, high-resolution, spatially explicit vegetation parameters were resampled to lower resolutions, providing ET estimates from input data representative of the resolution of modern global satellite systems, i.e. SPOT (10m), Landsat (30m), and MODIS (250m, 500m, 1000m). The accuracy of these lower resolution estimates of ET are examined relative to high-resolution estimates and validation data. Understanding the relative accuracy of ET estimates with increasingly low-resolution input data will help determine the validity of using data acquired from low-resolution global satellite systems to drive regional-scale climate models.