B109-0012
Introducing the NEON Ecological Forecasting Challenge hosted by the Ecological Forecasting Initiative Research Coordination Network.

Wednesday, 16 December 2020
Poster
Quinn Thomas1, Carl Boettiger2, Cayelan Carey3, Michael Dietze4, Andrew Fox5, Melissa A. Kenney6, Christine M Laney7, Jason S McLachlan8, Jody Peters8, Jake F Weltzin9, Whitney M Woelmer3, John R Foster4, James P. Guinnip10, Anna Spiers11, Sadie Ryan12, Kathryn Isabel Wheeler13, Alexander R Young14 and Leah R Johnson15, (1)Virginia Polytechnic Institute and State University, Forest Resources and Environmental Conservation, Blacksburg, VA, United States, (2)University of California, Berkeley, Davis, CA, United States, (3)Virginia Polytechnic Institute and State University, Biological Sciences, Blacksburg, VA, United States, (4)Boston University, Boston, MA, United States, (5)Joint Center for Satellite Data Assimilation, Boulder, CO, United States, (6)University of Minnesota Twin Cities, Minneapolis, United States, (7)Battelle, National Ecological Observatory Network (NEON), Boulder, CO, United States, (8)University of Notre Dame, Notre Dame, IN, United States, (9)USA National Phenology Network, Tucson, AZ, United States, (10)Kansas State University, Manhattan, United States, (11)University of Colorado at Boulder, Boulder, United States, (12)University of Florida, Geography, Ft Walton Beach, FL, United States, (13)University of Delaware, Geography, Newark, DE, United States, (14)SUNY College of Environmental Science and Forestry, Syracuse, United States, (15)Virginia Polytechnic Institute and State University, Blacksburg, United States
Abstract:
The 21st century has been - and will continue to be - characterized by major environmental changes to the ecosystem services upon which society depends. Anticipating and responding to these changes requires developing novel approaches that integrate data and models to make explicit predictive forecasts in near real time. Furthermore, the cycle of creating forecasts, evaluating them using observations, and revising ecological hypotheses and models based on that feedback has the potential to accelerate learning across many disciplines within ecology. This is especially true when observations used to develop models and evaluate forecasts are readily available, include quantified uncertainty, and are standardized across time and space - as is now available through the National Ecological Observatory Network (NEON). To create a community of practice that builds capacity for ecological forecasting by leveraging NEON data, we are introducing the NEON Ecological Forecasting Challenge. The Challenge is hosted by the Ecological Forecasting Initiative Research Coordination Network (EFI-RCN) with two primary goals: 1) provide the community with a formal opportunity to develop iterative near-term ecological forecasts using different NEON data products, and 2) compare forecasts spanning multiple ecosystems to answer fundamental ecology questions on predictability. In the first round of the Challenge, we are focusing on five themes that include ecosystem, community, and population dynamics: terrestrial carbon and water fluxes, freshwater temperature and dissolved oxygen, tick population, terrestrial phenology, and beetle diversity. The Challenge is designed to be collaborative and open to anyone, which is aligned with the EFI-RCN objectives to increase the number, diversity, and quality of forecasts. In leading the Challenge, broad needs of the ecological forecasting community are addressed by: (1) defining the standards for archiving ecological forecasts, (2) providing training opportunities to encourage broad participation, (3) collaboratively developing software tools, (4) engaging forecast stakeholders in the development of the Challenge rules, scope, and potential use, and (5) using submissions as the basis for a multi-forecast synthesis that will highlight emergent patterns across forecasts.