Samantha K. Oliver Galen Gorski Jeremy Alejandro Diaz 2025 <p><span>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 &gt;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 &lt;2&nbsp;°C with 90&nbsp;% prediction intervals that contain 90.7&nbsp;% of observations. We also assessed how the model captured variability in ecologically relevant metrics (e.g., R</span>²<span>&nbsp;for annual 7-day maximum&nbsp;=&nbsp;0.76; R</span>²<span>&nbsp;for days exceeding 25&nbsp;°C&nbsp;=&nbsp;0.75). This model does not outperform state-of-the-art machine learning efforts (e.g., RMSE ≤1.5&nbsp;°C) due to a limited input set but does provide the most spatially complete modeling to date to support water availability assessments.</span></p> application/pdf 10.1016/j.envsoft.2025.106655 en Elsevier Evaluation of daily stream temperature predictions (1979-2021) across the contiguous United States using a spatiotemporal aware machine learning algorithm article