Evaluate a Hierarchical Modeling Framework for Decomposing Fluxes at Heterogeneous Eddy-Covariance Sites

Global networks of eddy-covariance towers provide synthesized datasets of energy, water, and carbon fluxes and have been widely used in many modeling and upscaling research. While the eddy-covariance data are recognized for their rich temporal information, their spatially dynamic nature due to the varying source areas from time to time (i.e., so-called flux footprint) is often overlooked. For sites located in a more-or-less heterogeneous or patchy landscape, the spatial variations of land surface characteristics and the temporal dynamics of flux footprints jointly lead to the so-called representativeness issue, i.e., to what extent do the flux measurements taken at the tower location reflect the flux conditions of a specific land-cover type at all times. We developed a modeling framework for decomposing the response functions of CO₂, H₂O, and sensible heat fluxes from distinct land-cover types within the flux tower footprints. The framework incorporated the temporal dynamics of footprints and the spatial variations of land surface characteristics but consisted of a hierarchy of model setups (e.g., time-varying parameter, spatial dependency) and data requirements (e.g., tower-based temporal data, fine-scale remote-sensing-based spatial and spatiotemporal data). We then evaluated the performance at selected AmeriFlux sites with different degrees of heterogeneity and patchiness. Our preliminary results showed that the baseline approach with the least model complexity and data requirements was generally sufficient and robust in decomposing flux at sites with moderate heterogeneity and patchiness. And the derived response functions can inform fluxes at specific land-cover types within the flux footprints. However, the performance and validity degraded with increasing heterogeneity and spatial complexity. Therefore, a more complex modeling framework with extra constraints (either through model structure or spatial data) is required for sites with high heterogeneity to derive robust flux decomposition. Our evaluation provides practical insight and guidance for future research, particularly in interpreting and analyzing eddy-covariance flux data at heterogeneous and patchy ecosystems.