OPS 2-1
Modeling long-term continuous stream metabolism across the continent: NEON opportunities and approaches

Tuesday, August 12, 2014
Exhibit Hall, Sacramento Convention Center
Claire K. Lunch, National Ecological Observatory Network, Boulder, CO
Keli Goodman, National Ecological Observatory Network, Boulder, CO
Gordon W. Holtgrieve, School of Aquatic and Fishery Sciences, University of Washington, Seattle, WA
Robert O. Hall Jr., Department of Zoology & Physiology, University of Wyoming, Laramie, WY
Amy M. Marcarelli, Department of Biological Sciences, Michigan Technological University, Houghton, MI
Madeleine M. Mineau, Plant, Soil and Environmental Sciences, University of Maine, Orono, ME
Jerard Bales, United States Geological Survey
Melody Bernot, Biology, Ball State University, Muncie, IN
Brian J. Roberts, Louisiana Universities Marine Consortium, Chauvin, LA
Background/Question/Methods

Stream metabolism, the measurement of primary productivity (GPP) and ecosystem respiration (ER) in stream ecosystems, is an important measure of stream health and indicator of climatic and local environmental change. Long-term, continuous monitoring of stream metabolism can enable better understanding of the integrated and complex effects of large-scale change (e.g., land-use, climate, atmospheric deposition, invasive species, etc.) on stream ecosystem function. Similarly, parallel measurements of metabolism in streams across a broad range of ecosystems can inform understanding of local and regional controls over ecosystem function, and illustrate differing responses to shared climate changes.

Long-term, continuous monitoring across a large number of sites creates challenges in both data management and modeling. A standardized framework is needed for error and uncertainty estimation and propagation, and for modeling approaches and parameterization. Challenges to modeling GPP and ER from long-term stream data include 1) obtaining unbiased, low-error estimates of daily fluxes, 2) interpreting GPP and ER estimates over extended time periods, and 3) developing an automated, standardized model.

Results/Conclusions

The National Ecological Observatory Network (NEON) is a national-scale research platform that will use consistent procedures and protocols to standardize measurements across the United States, providing long-term, high-quality, open-access data from a connected network to address large-scale change. NEON is currently preparing for the collection, processing, and delivery to the public of 30 years of continuous whole-ecosystem stream metabolism data from 29 stream and river sites across the US. In partnership with academic and government scientists, NEON is developing a Bayesian inverse modeling framework for partitioning metabolism into GPP and ER. Automating metrics of both data and model quality is a major goal. The model developed by NEON must be flexible enough to accommodate NEON’s diverse aquatic sites, and thus will be usable at other sites as well. The model itself will be made available to the community along with all associated data.