loading page

What can we learn from 100,000 freshwater forecasts? A synthesis from the NEON Ecological Forecasting Challenge
  • +10
  • Freya Olsson,
  • Cayelan C Carey,
  • Carl Boettiger,
  • Gregory Harrison,
  • Robert Ladwig,
  • Marcus F Lapeyrolerie,
  • Abigail S L Lewis,
  • Mary E Lofton,
  • Felipe Montealegre-Mora,
  • Joseph S Rabaey,
  • Caleb J Robbins,
  • Xiao Yang,
  • R Quinn Thomas
Freya Olsson
Department of Biological Sciences, Virginia Tech, Center for Ecosystem Forecasting, Virginia Tech

Corresponding Author:[email protected]

Author Profile
Cayelan C Carey
Department of Biological Sciences, Virginia Tech, Center for Ecosystem Forecasting, Virginia Tech
Carl Boettiger
Department of Environmental Science, Policy, and Management, University of California Berkeley
Gregory Harrison
Center for Ecosystem Forecasting, Virginia Tech
Robert Ladwig
Department of Ecoscience, Aarhus University, Center for Limnology, University of Wisconsin-Madison
Marcus F Lapeyrolerie
Department of Environmental Science, Policy, and Management, University of California Berkeley
Abigail S L Lewis
Department of Biological Sciences, Virginia Tech
Mary E Lofton
Department of Biological Sciences, Virginia Tech, Center for Ecosystem Forecasting, Virginia Tech
Felipe Montealegre-Mora
Department of Environmental Science, Policy, and Management, University of California Berkeley
Joseph S Rabaey
Large Lakes Observatory, University of Minnesota
Caleb J Robbins
Institute of Arctic Biology, University of Alaska Fairbanks, Center for Reservoir and Aquatic Systems Research, Baylor University
Xiao Yang
Department of Earth Sciences, Southern Methodist University
R Quinn Thomas
Department of Biological Sciences, Virginia Tech, Center for Ecosystem Forecasting, Virginia Tech, Department of Forest Resources and Environmental Conservation, Virginia Tech

Abstract

Near-term, iterative ecological forecasts can be used to help understand and proactively manage ecosystems. To date, more forecasts have been developed for aquatic ecosystems than other ecosystems worldwide, likely motivated by the pressing need to conserve these essential and threatened ecosystems and increasing availability of high-frequency data. Forecasters have implemented many different modelling approaches to forecast freshwater variables, which have demonstrated promise at individual sites. However, a comprehensive analysis of the performance of varying forecast models across multiple sites is needed to understand broader controls on forecast performance. Forecasting challenges (i.e., community-scale efforts to generate forecasts while also developing shared software, training materials, and best practices) present a useful platform for bridging this gap to evaluate how a range of modelling methods perform across axes of space, time, and ecological systems. Here, we analysed forecasts from the aquatics theme of the National Ecological Observatory Network (NEON) Forecasting Challenge hosted by the Ecological Forecasting Initiative. Over 100,000 probabilistic forecasts of water temperature and dissolved oxygen concentration for 1-30 days ahead across seven NEON-monitored lakes were submitted in 2023. We . For water temperature, we found that forecast skill degraded with increases in forecast horizons, process-based models and models that included air temperature as a covariate generally exhibited the highest forecast performance, and that the most skillful forecasts often Overall, the NEON Forecasting Challenge provides an exciting opportunity for a model inter-comparison to learn about the relative strengths of a diverse suite of models and advance our understanding of freshwater ecosystem predictability.
08 Oct 2024Submitted to ESS Open Archive
09 Oct 2024Published in ESS Open Archive