TECO modeling, data assimilation, and ecological forecasting at Sevilleta (SEV) LTER site. NSF, 2018-2023 (Co-PI, PI: Jennifer Rudgers, University of New Mexico)

We are doing modeling in close collaboration with the Mean-Variance experiment and observations from flux towers, remote sensing, and biogeochemical measurements at the SEV site. The dryland version of TECO explicitly includes microbial processes and enables simulations of biogeochemical processes in biocrusts. This TECO model permits us to explore plant-microbe interactions at SEV. We further develop TECO by incorporating the proposed SEV plant traits database to predict ecosystem dynamics during biome transitions. We use data assimilation techniques to assimilate data from the experiments and observations to train the TECO model before we use it to explore how ecosystem productivity and biogeochemical processes respond to (1) changes in climate means and variability, (2) disturbances, and (3) resulting biome transitions. Moreover, we adopt the Ecological Platform for Assimilating Data into models (EcoPAD) to develop real- or near-time ecological forecasting, first at our flux tower sites, and later for our Mean-Variance experiment. This ecological forecasting system feeds back to inform both what datasets are needed to improve predictive ability and what modeling improvements are required.