Next-Generation Ecosystem Experiments-Arctic (NGEE-Arctic)

Lead Investigator: 
Participating Staff: 
Brookhaven National Laboratory, Lawrence Berkeley National Laboratory, Los Alamos National Laboratory, Sandia National Laboratory, UIC Science, University of Alaska Fairbanks
US DOE, Office of Science, Biological and Environmental Research
Start Date: 
October 1, 2012
End Date: 

Executive Summary: An important challenge for Earth System Models (ESMs) is to represent land surface and subsurface processes and their interactions in a warming climate. This is true for all regions of the world, but it is especially relevant in the Arctic where surface air temperatures at high latitudes are projected to warm at a rate twice that of the global average in the coming century. While changes in regional temperatures are expected to impact sea ice, snowpack, permafrost, and other components of the Arctic system, these changes are made even more important because they are expected to play, and may already be playing, a role in determining the climate of the rest of the globe.

Ice wedge polygons dominate
the landscape in North Alaska.

The Next-Generation Ecosystem Experiments (NGEE Arctic) is a 10-year project (2012 to 2022) to reduce uncertainty in ESMs through developing a predictive understanding of carbon-rich Arctic system processes and feedbacks to climate. This is achieved through experiments, observations, and synthesis of existing datasets that strategically inform model process representation and parameterization and enhance the knowledge base required for model initialization, calibration, and evaluation. The concept of model-experiment integration (ModEx) requires strong collaboration between scientists developing and testing models and those conducting research in the field and laboratory. Motivated by a ModEx approach, this proposal highlights progress made by the multi-disciplinary NGEE Arctic team in Phase 1 and describes plans for Phase 2. In Phase 1 (2012 to 2014), NGEE Arctic tested and applied a multi-scale measurement and modeling framework in coastal tundra on the North Slope of Alaska. Field plots, transects, and satellite sites near Barrow, Alaska, were chosen to represent a cold, continuous permafrost region at the northern extent of an ecological and climatic gradient. Much of our research focused on subgrid heterogeneity in thermal-hydrology, biogeochemistry, and vegetation structured by topography, landscape, and drainage networks. These efforts provided datasets and derived products and knowledge that meet project requirements for model initialization, parameterization, process representation, and evaluation. Some of these capabilities are now being adopted by DOE’s Earth System Modeling program as fundamental new developments in a next-generation ESM, the Accelerated Climate Model for Energy (ACME).

Building upon research conducted in the first three years of the project, in Phase 2 (2015 to 2018) we will pursue additional field, laboratory, and modeling objectives in Barrow, Alaska. This research will proceed along a natural line of investigation that takes advantage of ongoing data collection, knowledge discovery, and model development. We also propose to establish a southern site which, compared to our research site on the North Slope, is characterized by transitional ecosystems, warm, discontinuous permafrost, higher annual precipitation, and well-defined watersheds with strong topographic gradients. Our selection of the Seward Peninsula is based on a Phase 1 analysis indicating that western Alaska is a proxy for the future ecological and climatic regime of the North Slope of Alaska toward the end of the century. Expanding our activities to the Seward Peninsula will allow us to challenge our Phase 1 scaling strategy with a contrasting environment that will require new process understanding and representation in models. We will use variation in the structure and organization of the Seward Peninsula landscape to guide a series of process-level investigations (Questions 1 through 3) that will be nested at scales ranging from core to plot, landscape, and watershed levels. Knowledge derived in these studies will identify mechanisms controlling carbon, water, nutrient, and energy fluxes, which will then be brought to bear on two integrative questions concerning the future of the Arctic in a changing climate (Questions 4 and 5).

Q1.   How does the structure and organization of the landscape control the storage and flux of carbon and nutrients in a changing climate?

A tram system carries scientific instrumentation for automated data collection.

Q2.   What will control rates of CO2 and CH4 fluxes across a range of permafrost conditions?

Q3.   How will warming and permafrost thaw affect above- and belowground plant functional traits, and what are the consequences for Arctic ecosystem carbon, water, and nutrient fluxes?

Q4.   What controls the current distribution of Arctic shrubs, and how will shrub distributions and associated climate feedbacks shift with expected warming in the 21st century?

Q5.   Where, when, and why will the Arctic become wetter or drier, and what are the implications for climate forcing?

Our model-inspired vision implemented in Phase 1, and now extended into Phase 2, strengthens the connection between process studies in Arctic ecosystems and high-resolution scaling strategies that form the foundation of DOE’s land surface modeling for climate prediction. The NGEE Arctic project supports the BER mission to advance a robust predictive understanding of Earth’s climate and environmental systems by delivering a process-rich ecosystem model, extending from bedrock to the top of the vegetative canopy/atmospheric interface, in which the evolution of Arctic ecosystems in a changing climate can be modeled at the scale of a high-resolution, next-generation ESM grid cell. Implicit in our expanded scope of research in Phase 2 is the need to build upon the scaling and modeling framework established in Phase 1 and to populate that framework with knowledge derived from experiments and observations from new and existing sites. This will facilitate upscaling our field and landscape-scale observations to regional scales, and encourage continued interactions on the North Slope of Alaska with the DOE’s ARM and Atmospheric System Research (ASR) programs and cross-agency collaborations with NSF (NEON), NOAA, USGS, and NASA through their CARVE and ABoVE campaigns. The new ASCR-BER project Interoperable Design of Extreme-scale Application Software (IDEAS) is using an NGEE Arctic thermal hydrology model to simulate polygonal tundra at the Barrow field site as one of its two use cases. The IDEAS Arctic use case will focus on refactoring of the software developed in Phase 1 to extend the current thermal hydrology capability to much larger spatial regions. Our research task on plant traits and trait-enabled modeling (i.e., Q3) is a direction that is consistent with that of the NGEE Tropics and ACME projects and represents an area where close collaboration among our projects will be encouraged. We will continue to collaborate with the TES SFA at Argonne National Laboratory as together we share knowledge and samples that can be used by the SFA to develop regional maps of soil carbon stocks and their intrinsic decomposability for model benchmarking. We will coordinate with the SPRUCE effort within the TES SFA at Oak Ridge National Laboratory to make use of new sub-grid models of wetland hydrology and microtopography.

In Phase 3 (2019 to 2022), we expect to be in a strong position to conduct pan-Arctic simulations using a model with unparalleled sophistication in its cross-scale process representation that is parameterized and evaluated against a multi-scale, nested hierarchy of measurements and synthesis products. Integration and a truly interdisciplinary perspective, forged by our team in Phase 1, will be foundational to Phase 2 activities and beyond as we use model sensitivity and uncertainty analysis and new process knowledge to guide computational, experimental, and observational efforts toward improved climate predictions in high-latitude ecosystems. Safety, collaboration, communication and outreach, and a strong commitment to data management, sharing, and archiving are key underpinnings of our model-inspired research in the Arctic.

Significance: Prediction of global climate change at decadal to century time scales requires Earth system models that couple physical, biological, and ecological processes, capturing the most significant climate change drivers and representing critical feedback mechanisms. Unfortunately, while current models represent many of the processes that govern important land-atmosphere interactions, there is a continuing need for data to represent new processes in models, constrain model predictions, and test models against experimentally-derive datasets. Scientific investigation of the response of Arctic ecosystems to climate is critical to our overall understanding of the impacts of global environmental change. This is because this region is currently experiencing rapid changes in the climate system, the biophysical system of the Arctic is particularly vulnerable to these changes, and there is currently large uncertainty in the net effect of these biophysical changes and their feedbacks to climate. Current changes observed in the Arctic system may be a harbinger of near-future

Conducting research in North Alaska presents challenges in accessing the field sites. Snow machines are the vehicles of choice.

Interesting Findings: Investigating linkages between surface and subsurface properties is critical for understanding the fate of terrestrial ecosystems in a changing climate. Near-surface soil hydrological and biogeochemical processes are widely known to be influenced by strong surface and subsurface interactions in ice-rich landscapes, but seldom are studies designed to examine co-variation in active layer and permafrost properties at scales appropriate for inclusion into Earth System Models. Scientists in the NGEE Arctic project are investigating linkages between soil and landscape property dynamics in the Arctic polygonal tundra near Barrow, AK. This is being done along transects that traverse a range of geomorphological conditions, including ice-wedge polygons, interstitial tundra, and drained-haw lake basins. Landscape characteristics are identified from topographic, multi-spectral and thermal-infrared imaging measurements using either a kite-, pole- and tram-based platform at various temporal and spatial scales from continuous monitoring along a 35 m long transect to occasional campaigns along 500x40 m corridors. Soil properties are inferred using electrical resistivity tomography, time-domain reflectometry, temperature measurements, and soil sample analysis. Overall, scientists are gaining an unprecedented appreciation for the spatiotemporal linkages that exist between various soil and landscape properties, including water inundation, vegetation, topography, thaw layer thickness, soil water content, temperature, electrical conductivity, and snow thickness. These emerging conclusions confirm the complementary nature of various ground- and aerial-based approaches and proxies to estimate soil properties within a framework that considers uncertainty, resolution, and spatial coverage. Among other results, a relatively strong relationship is observed between changes in soil electrical conductivity, water content, active layer thickness, and vegetation state. These associations reinforce the importance of water distribution on various processes including vegetation dynamics, thermal conductivity, surface-subsurface energy exchange, redox reactions, and biogeochemical mechanisms. Identifying such linkages is crucial if we are to extrapolate knowledge from point-scale and core-based biogeochemical measurements at specific sites over larger scales to ultimately improve parameterization of models simulating ecosystem feedbacks to climate.