NGEE Arctic: Integrating Boots-on-the-Ground Observations with the Virtual World of Models to Answer Big Science Questions Across the Arctic

Authors

Colleen Iversen¹* (iversencm@ornl.gov), Charles Abolt³, Katrina Bennett³, W. Robert Bolton¹, Amy Breen⁴, Ryan Crumley³, Baptiste Dafflon², Eugenie Euskirchen⁴, David E. Graham¹, Susan Heinz¹, Forrest M. Hoffman¹, Jennifer Holm², Charles Koven², Hannah Mevenkamp⁴, Scott Painter¹, William J. Riley², Alistair Rogers², Vladimir E. Romanovsky⁴, Verity G. Salmon¹, Fernanda Santos¹, Benjamin N. Sulman¹, Neslihan Taş², Michele Thornton¹, Peter E. Thornton¹, Margaret Torn², Terri Velliquette¹, Daryl Yang¹, Stan D. Wullschleger¹, NGEE Arctic team

Institutions

¹Oak Ridge National Laboratory, Oak Ridge, TN; ²Lawrence Berkeley National Laboratory, Berkeley, CA; ³Los Alamos National Laboratory, Los Alamos, NM; ⁴University of Alaska–Fairbanks, AK

URLs

https://ngee-arctic.ornl.gov

Abstract

The Arctic is a historically understudied biome that holds at least twice as much carbon as the atmosphere. Climate change and Arctic amplification are warming the Arctic faster than the rest of the world, and increasingly frequent wildfires and widespread permafrost degradation are impacting ecosystem structure and function. The Next-Generation Ecosystem Experiments (NGEE) Arctic is a model-driven, multiscale research project with a goal of improving understanding and model representation of complex interactions among climate, landforms, hydrology, biogeochemistry, vegetation, snow, and permafrost in Arctic tundra. NGEE Arctic emphasizes iterative collaboration among interdisciplinary teams of empiricists and modelers to incorporate experiments into models—underscored by open science and a safe, inclusive project culture.

Observations made by the NGEE Arctic team across a gradient of permafrost landscapes in Alaska have improved the representation of tundra processes in the Energy Exascale Earth System Model. The project’s modeling approach was initially driven by the recognition that the current generation of models failed to capture many processes controlling climate feedbacks in Arctic tundra. Model improvements have emphasized unique aspects of permafrost environments and explored reductions in model complexity while retaining predictive power. The team is now leveraging Arctic-informed models to make novel predictions. For example, researchers found that improved model representation of the unique characteristics of tundra plants improved model prediction of carbon-climate feedbacks and underscored the need for dynamic vegetation. In turn, improved model representation of ground subsidence predicted that a sinking tundra surface is unlikely to trigger runaway permafrost thaw. Furthermore, machine learning algorithms trained on thousands of observations improved model representation of snow distribution across the landscape. The next phase of NGEE Arctic proposes to confront models trained on and evaluated against observations from Arctic Alaska with current observations and projected climate responses across the Arctic.