Evaluation of daily stream temperature predictions (1979-2021) across the contiguous United States using a spatiotemporal aware machine learning algorithm

Environmental Modelling & Software
By: , and 

Links

Abstract

Stream temperature controls a variety of physical and biological processes that affect ecosystems, human health, and economic activities. We used 42 years (1979–2021) of data to predict daily summary statistics of stream temperature across >50,000 stream reaches in the contiguous United States using a recurrent graph convolution network. We comprehensively documented the performance – both across all reaches and by stream type (e.g., reservoir or groundwater influence) – as a baseline for future improvement. The model showed reach-level RMSE of <2 °C with 90 % prediction intervals that contain 90.7 % of observations. We also assessed how the model captured variability in ecologically relevant metrics (e.g., R2 for annual 7-day maximum = 0.76; R2 for days exceeding 25 °C = 0.75). This model does not outperform state-of-the-art machine learning efforts (e.g., RMSE ≤1.5 °C) due to a limited input set but does provide the most spatially complete modeling to date to support water availability assessments.

Study Area

Publication type Article
Publication Subtype Journal Article
Title Evaluation of daily stream temperature predictions (1979-2021) across the contiguous United States using a spatiotemporal aware machine learning algorithm
Series title Environmental Modelling & Software
DOI 10.1016/j.envsoft.2025.106655
Volume 193
Year Published 2025
Language English
Publisher Elsevier
Contributing office(s) WMA - Integrated Information Dissemination Division
Description 106655, 16 p.
Country United States
Other Geospatial contiguous United States
Additional publication details