Species-specific hydraulic traits play a critical role in determining the response of ecosystem carbon and water fluxes to water stress. Improving the representation of plant hydraulic behavior in vegetation and land-surface models is critical for improving our predictions of the impacts of water stress on ecosystem carbon and surface fluxes given that biodiverse representation of forest canopies remain challenging for land-surface models. Here, we use FETCH3.14, a multispecies, canopy-level, hydrodynamic transpiration model which builds upon the previous versions of the Finite-difference Ecosystem-scale Tree Crown Hydrodynamics model (FETCH). FETCH3.14 is parameterized by our newly developed package, Bayesian Optimization for Anything (BOA), which facilitates and eases hyperparameter optimization using multi-scale and multi-variate observations.
BOA incorporates multiple sources of data easily, reduces optimization setup time, and eases advanced use cases such as High-Performance Computing (HPC) parallelization and optimization restarting. BOA facilitates multi-source data assimilation for FETCH3.14 from a disparate range of sources including ET observations, soil and stem water potential observations, and carbon flux observations to provide insights about species-specific hydraulic traits. We use flux data from representative model trees that get scaled to the plot level based on the composition of species and structure of the canopy in the plot, which allows parameterization using tree level observations (sap flux, stem water storage) and plot level observations (eddy covariance evapotranspiration). We use BOA to set up a multi-objective optimization inverse problem with little overhead or extra boilerplate code. This approach allows us to utilize multi-scale observations to resolve information about species-specific hydraulic parameters, including parameters that are difficult or impossible to measure in the field.