Improving predictive understanding of Earth system variability requires not only models of ever-increasing mechanistic complexity and resolution, but understanding how ecological processes contribute to model uncertainty. This requires increasing integration across multiple disciplines to determine key mechanisms and controlling model parameters, computational infrastructure to run large model ensembles, and how to deal efficiently with large model output datasets. Uncertainty quantification (UQ) algorithms that use machine learning and artificial intelligence link observations and models together to produce credible simulations of Earth system behavior with uncertainty estimates. Deep neural networks (DNNs) combined with dimension-reduction techniques can be used to build fast-to-evaluate surrogate models of spatiotemporally varying model output fields using a smaller number of ESM ensemble members and higher accuracy than traditional methods. These surrogate models can then be used for uncertainty quantification techniques such as global sensitivity analysis, data assimilation or model calibration, leading to more rapid advances in model development, prediction of climate processes, and targeting which processes need more observations or model algorithm development.
Results/Conclusions
Here we focus on uncertainty quantification of the Energy Exascale Earth System land model (ELM) parameters related to key ecosystem processes considering uncertainty in 20 key model parameters related to fluxes of carbon and energy, phenology and drought response. Parameter uncertainty ranges are determined from trait databases and the literature across a range of 13 naturally occurring plant functional types. A 200-member ensemble of ELM simulations is performed at 2x2 spatial resolution. Global sensitivity analysis indicates different parameters drive model prediction uncertainty depending on time of year and environmental conditions. In warmer and drier climates, parameters controlling stomatal conductance and rooting depth distribution are strong drivers of productivity, while in colder climates phenology and temperature sensitivity parameters are more important. We investigate how these parameter sensitivities change under a warming climate and under a doubling of atmospheric CO2 concentrations. The ensemble of ELM simulations is also used to assess the vulnerability of ecosystems to climate change in a probabilistic sense.