Using Spatial Eddy Covariance to Investigate Energy Balance Closure over a Heterogeneous Ecosystem

Land surface heterogeneity influences patterns of sensible and latent heat flux, which in turn affect processes in the atmospheric boundary layer. However, gridded atmospheric models often fail to incorporate the influence of land surface heterogeneity due to differences between the temporal and spatial scales of models compared to the local, sub-grid processes. Improving models requires the scaling of surface flux measurements; a process made difficult by the fact that surface measurements usually find an imbalance in the energy budget.

The Chequamegon Heterogeneous Ecosystem Energy-balance Study Enabled by a High-density Extensive Array of Detectors (CHEESEHEAD19) was an observational experiment designed to investigate how the atmospheric boundary layer responds to scales of spatial heterogeneity in surface-atmosphere heat and water exchanges. The campaign was conducted from June – October 2019, measuring surface energy fluxes over a heterogeneous forest ecosystem as fluxes transitioned from latent heat-dominated summer through sensible heat-dominated fall. Observations were made by ground, airborne, and satellite platforms within the 10 x 10 km study region, which was chosen to match the scale of a typical model grid cell. The spatial distribution of energy fluxes was observed by an array of 20 eddy covariance towers and a low-flying aircraft. Mesoscale atmospheric properties were measured by a suite of LiDAR and sounding instruments, measuring winds, water vapor, temperature, and boundary layer development. Plant phenology was measured in-situ and mapped remotely using hyperspectral imaging.

The dense set of multi-scale observations of land-atmosphere exchange collected during the CHEESEHEAD field campaign permits combining the spatial and temporal distribution of energy fluxes with mesoscale surface and atmospheric properties. This provides an unprecedented data foundation to evaluate theoretical explanations of energy balance non-closure, as well as to evaluate methods for scaling surface energy fluxes for improved model-data comparison. Here we show how fluxes calculated using a spatial eddy covariance technique across the 20-tower network compare to those of standard temporal eddy covariance fluxes in order to characterize of the spatial representativeness of single tower eddy covariance measurements. Additionally, we show how spatial EC fluxes can be used to better understand the energy balance over heterogeneous ecosystems.