Climate Model Projections of the Water and Carbon Cycles

Transforming the mechanistic representation of key carbon and water cycle processes in DOE’s flagship Energy Exascale Earth System Model (E3SM)

EESA is conducting path breaking research on climate modeling through several Berkeley Lab-led and collaborative projects that provide enhanced mechanistic representations of key carbon and water cycle processes to Earth System Models. These efforts include:

CASCADE SFA (Calibrated and Systematic Characterization, Attribution, and Detection of Extremes Scientific Focus Area), to measure, model, and understand climate extremes in the historical record and to project their evolution in future, warmer climates.
Energy Exascale Earth System Model (E3SM) project to develop a climate model for DOE with better resolution of complex, multi-scale earth system interactions, more realistic representation of the two-way interactions between human and natural processes, and ensemble modeling to quantify earth system uncertainties.
RUBISCO SFA (Reducing Uncertainties in Biogeochemical Interactions through Synthesis and Computation Scientific Focus Area), to identify and quantify interactions between biogeochemical and hydrological cycles and the earth system, and to quantify and reduce uncertainties in ESMs associated with those interactions.
HyperFACETS, to advance understanding of regional processes at the atmosphere-water-energy-land interface, and to quantify and improve our ability to perform credible climate modeling at regional scales across the contiguous United States.
The Next Generation Ecosystem Experiments (NGEEs), focused in the Arctic and Tropics. Berkeley Lab leads the NGEE-Tropics project and is a key participant in the NGEE-Arctic project.
The International Land Model Benchmarking Project (ILAMB), which uses AmerFlux data. LBNL participates in ILAMB and leads the AmeriFlux Management Project.
An Early Career Research Project (ECRP) on vegetation dynamical responses to multivariate extremes in the western U.S. to better understand the role of climate extremes on shaping ecosystem-level responses and feedbacks to climate change via evolutions in the coverage, size, and functional composition of vegetation.

Relevant science, program advances, & PROJECTS

CASCADE

CASCADE (Calibrated and Systematic Characterization, Attribution, and Detection of Extremes Scientific Focus Area)

Developed a new first-principles theory and model for aggregation of deep tropical convection as well as the first analytical theory for the spatial extent of blocking events that can induce major droughts in the western U.S. (CASCADE)
Developed the world’s first exascale climate application, a machine learning algorithm for identifying storms in climate model simulations, which was awarded the Association for Computing Machinery Gordon Bell prize at Supercomputing 2018 (CASCADE)

NGEE-Tropics/NGEE-Arctic

The Next Generation Ecosystem Experiments (NGEEs), focused in the Arctic and Tropics. Berkeley Lab leads the NGEE-Tropics project and is a key participant in the NGEE-Arctic project.

E3SM/FATES

E3SM (Energy Exascale Earth System Mode)

NGEE-Tropics investigators have introduced the Functionally Assembled Terrestrial Ecosystem Simulator (FATES), a model of forests that dynamically treats the size, age, and succession of trees, into DOE’s flagship Energy Exascale Earth System Model (E3SM) (E3SM/FATES)
FATES produces drastically different projections of terrestrial carbon stores relative to traditional ‘big leaf’ models used in the majority of Earth System Model simulations evaluated in previous and current IPCC Assessments. (E3SM/FATES)

RUBISCO

RUBISCO SFA (Reducing Uncertainties in Biogeochemical Interactions through Synthesis and Computation Scientific Focus Area)

Projected that the majority of ecosystems currently limited by temperature will no longer be so as soon as 2050. CMIP5 models were unable to capture these dynamics, motivating needed improvements to land model representations of temperature sensitivity (RUBISCO)
Through a collaboration with NGEE-Arctic, we have shown that uptake of nitrogen by vegetation during non-growing periods leads to N₂O emissions that produce a radiative forcing comparable to current net land-carbon exchanges (RUBISCO)

HyperFACETS

Found that most regional climate model simulations systematically overestimate melt-rate biases and that these biases must be addressed to improve the utility of future projections from these models for water stakeholders (HyperFACETS)

North American Coordinated Regional Climate Downscaling Experiment

Using the North American Coordinated Regional Climate Downscaling Experiment, projected that by the end of the 21st century all models show a significant and similar decline in Sierra Mountain snowpack that supplies one-third of California’s potable water.

Early Career Research Project (ECRP)

Demonstrated that an extension of FATES that captures root-water table interactions can replicate with high fidelity the response of the Sierran forest ecosystem to the extreme 2012-2015 drought (ECRP)
Developed approaches for handling the complexity of modern land surface models, in particular how to develop modeling systems that enable easy adjustment of the level of structural and process complexity to ask specific scientific questions (ECRP)

AmeriFlux