Accurate spatiotemporal predictions of daily stream temperature from statistical models accounting for interactions between climate and landscape

Abstract

Spatial and temporal patterns in stream temperature are primary factors determining species composition, diversity and productivity in stream ecosystems. The availability of spatially and temporally continuous estimates of stream temperature would improve the ability of biologists to fully explore the effects of stream temperature on biota. Most statistical stream temperature modeling techniques are limited in their ability to account for the influence of variables changing across spatial and temporal gradients. We identified and described important interactions between climate and spatial variables that approximate mechanistic controls on spatiotemporal patterns in stream temperature. With identified relationships we formed models to generate reach-scale basin-wide spatially and temporally continuous predictions of daily mean stream temperature in four Columbia River tributaries watersheds of the Pacific Northwest, USA. Models were validated with a testing dataset composed of completely distinct sites and measurements from different years. While some patterns in residuals remained, testing dataset predictions of selected models demonstrated high accuracy and precision (averaged RMSE for each watershed ranged from 0.85–1.54 °C) and was only 17% higher on average than training dataset prediction error. Aggregating daily predictions to monthly predictions of mean stream temperature reduced prediction error by an average of 23%. The accuracy of predictions was largely consistent across diverse climate years, demonstrating the ability of the models to capture the influences of interannual climatic variability and extend predictions to timeframes with limited temperature logger data. Results suggest that the inclusion of a range of interactions between spatial and climatic variables can approximate dynamic mechanistic controls on stream temperatures.

Which environmental factors control extreme thermal events in rivers? A multi-scale approach (Wallonia, Belgium)

“…Higher elevation monitoring stations tend to record lower air temperatures. WT is therefore generally lower ( Siegel & Volk, 2019 ) when elevation is higher. Stream slope , another topographic characteristic of watercourses, also influences WT ( Mayer, 2012 ).…”

“…Some research shows that shade strongly influences stream WT ( Brown & Krygier, 1970 ; Beschta, 1997 ; Poole & Berman, 2001 ) by influencing solar energy inputs into the water ( Brazier & Brown, 1973 ; Brosofske et al, 1997 ). It is observed that shaded streams tend to be cooler in summer ( Siegel & Volk, 2019 ). Of all those enumerated above, land cover is one of the few variables on which humans can act to mitigate high WT without altering the properties of the stream itself.…”

Modeling Seasonal Effects of River Flow on Water Temperatures in an Agriculturally Dominated California River

“…We used cyclic cubic regression splines ("cc") as the smoother for D and thin plate regression splines ("tp") as smoothers for other covariates. To improve prediction under new conditions and avoid overfitting (Jackson et al, 2018;Siegel & Volk, 2019), we limited smoothers for air temperature and flow to a maximum of three knots, except in the one-covariate model "GAM11" where air temperature was allowed six knots. D was allowed up to six knots, except in three-dimensional tensors where it was restricted to five knots.…”

“…While time-varying covariate effects can be represented using separate models for each season (Mohseni et al, 1998;Sohrabi et al, 2017), this may cause unnatural, abrupt changes at seasonal transitions. Time-varying coefficients, including those used in generalized additive models (GAMs) (Pedersen et al, 2019;Wood, 2017) use continuous functions that avoid these abrupt changes (Li et al, 2014;Jackson et al, 2018;Siegel & Volk, 2019). While GAMs have been used in daily stream temperature modeling for single-site prediction (Boudreault et al, 2019;Coleman et al, 2021;Glover et al, 2020;Laanaya et al, 2017;Siegel et al, 2022), spatiotemporal prediction (Jackson et al, 2018;Siegel & Volk, 2019), identifying extreme events (Georges et al, 2021), and trend assessment (Yang & Moyer, 2020), few studies have used GAMs to model seasonally varying flow effects or identify when stream temperatures are most affected by flow variation (Glover et al, 2020;Yang & Moyer, 2020).…”

Examining Scale Dependent Environmental Effects on American Lobster (Homarus americanus) Spatial Distribution in a Changing Gulf of Maine

“…Environmental and spatial variables were also explored via GAM response curves for each significant predictor variable. Latitude and longitude variables were combined as an interaction term in each model to help account for spatial autocorrelation (Siegel and Volk, 2019). Response curves varied greatly depending on independent variable, season, sex, size, and spatial scale of the model.…”