Influence of climate forecasts, data assimilation, and uncertainty propagation on the performance of near-term phenology forecasts

Authors: Taylor, Shawn; WHITE, ETHAN - University Of Florida

Submitted to: Peer J Computer Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/21/2020
Publication Date: 8/20/2020
Citation: Taylor, S.D., White, E.P. 2020. Influence of climate forecasts, data assimilation, and uncertainty propagation on the performance of near-term phenology forecasts. Peer J Computer Science. Article. https://doi.org/10.1101/2020.08.18.256057.
DOI: https://doi.org/10.1101/2020.08.18.256057

Interpretive Summary: Ecological forecasting is becoming more commonplace, but evaluation of those forecasts is rare. Evaluation measures the accuracy of past forecasts and helps build trust in the system. It also shows where improvements can be made. Here we performed an evaluation of a near-term plant phenology forecast which has been operating for several years. We evaluated overall accuracy and reliability, and also how well upstream climate forecasts improved the predictions. We found that the climate forecasts contributed little overall to phenology forecast accuracy, though some species did benefit from them. Using observed winter and spring temperature provided the largest improvement of forecast accuracy throughout the spring. We also found that phenology forecast uncertainty, eg. the upper and lower bounds of predictions, were most robust when uncertainty from all available sources was used, as opposed to using climate uncertainty alone. Our analysis points the way toward several potential improvements to the forecasting system, which can be re-evaluated at a future date in a continuous cycle of forecast refinement.

Technical Abstract: Evaluation of ecological forecasts is a vital step in the continuous improvement of near-term ecological forecasts. Here we performed a thorough evaluation of a near-term phenological forecast system which has been operating for several years. We evaluated point forecast accuracy and the reliability of the prediction intervals. We also tested the contribution of upstream climate forecasts on phenology forecast proficiency. We found that 9 month climate forecasts contributed little skill overall, though some species did benefit from them. The assimilation of observed winter and spring temperature provided the largest improvement of forecast skill throughout the spring. We also found that phenology forecast prediction intervals were most robust when uncertainty was propagated from climate, phenological model, and model parameters as opposed to using climate uncertainty alone. Our analysis points the way toward several potential improvements to the forecasting system, which can be re-evaluated at a future date in a continuous cycle of forecast refinement.