Progress and opportunities in advancing near‐term forecasting of freshwater quality

Abstract

Near‐term freshwater forecasts, defined as sub‐daily to decadal future predictions of a freshwater variable with quantified uncertainty, are urgently needed to improve water quality management as freshwater ecosystems exhibit greater variability due to global change. Shifting baselines in freshwater ecosystems due to land use and climate change prevent managers from relying on historical averages for predicting future conditions, necessitating near‐term forecasts to mitigate freshwater risks to human health and safety (e.g., flash floods, harmful algal blooms) and ecosystem services (e.g., water‐related recreation and tourism). To assess the current state of freshwater forecasting and identify opportunities for future progress, we synthesized freshwater forecasting papers published in the past 5 years. We found that freshwater forecasting is currently dominated by near‐term forecasts of water quantity and that near‐term water quality forecasts are fewer in number and in the early stages of development (i.e., non‐operational) despite their potential as important preemptive decision support tools. We contend that more freshwater quality forecasts are critically needed and that near‐term water quality forecasting is poised to make substantial advances based on examples of recent progress in forecasting methodology, workflows, and end‐user engagement. For example, current water quality forecasting systems can predict water temperature, dissolved oxygen, and algal bloom/toxin events 5 days ahead with reasonable accuracy. Continued progress in freshwater quality forecasting will be greatly accelerated by adapting tools and approaches from freshwater quantity forecasting (e.g., machine learning modeling methods). In addition, future development of effective operational freshwater quality forecasts will require substantive engagement of end users throughout the forecast process, funding, and training opportunities. Looking ahead, near‐term forecasting provides a hopeful future for freshwater management in the face of increased variability and risk due to global change, and we encourage the freshwater scientific community to incorporate forecasting approaches in water quality research and management.

Data assimilation experiments inform monitoring needs for near‐term ecological forecasts in a eutrophic reservoir

“…Among ecosystems, freshwater lakes and reservoirs are particularly important systems for developing near‐term forecasts because they provide essential ecosystem services, including drinking water, food, irrigation, and recreation (Carpenter et al, 2011; Meyer et al, 1999; Williamson et al, 2016). Because freshwaters are experiencing greater variability and adverse water quality issues in response to land use and climate change (e.g., O'Reilly et al, 2015; Paerl & Paul, 2012; Woolway et al, 2021), some water managers have used forecasts to preemptively address poor water quality events (reviewed by Lofton et al, 2023). To date, iterative, near‐term freshwater forecasts have been developed for a number of water quality variables, including water temperature (e.g., Carey, Woelmer, et al, 2022; Thomas, McClure, et al, 2023), dissolved oxygen (e.g., Wang et al, 2016), and phytoplankton (e.g., Page et al, 2017; Woelmer et al, 2022).…”

Modelling the potential for local management practices to offset climate change impacts on freshwater macroinvertebrate communities

“…Manipulative experiments in the laboratory or in the field are used to quantify the individual and combined effects of multiple stressors for specific endpoints (Jackson et al., 2016; Orr et al., 2020; Verberk, Durance, et al., 2016), while remote sensing and observational studies are used to correlate anthropogenic activity with changes in biodiversity across larger temporal and spatial scales (Birk et al., 2020; Gilarranz et al., 2022). To complement these approaches, numerical modelling can be used to simulate different scenarios and to make mechanistically informed forecasts at a landscape scale (Lofton et al., 2023). Although anthropogenic stressors may arise from either global or local activities, their effects interact with each other and with the environment at local or regional scales, which can be captured by process‐based modelling.…”

“…Indeed, these process‐based models have been widely employed to understand the impacts of climate change and land‐use change on the physics and chemistry of river systems and are increasingly being used to make biological forecasts and to inform ecosystem management (e.g., Bussi, Whitehead, et al., 2016; Guse et al., 2015; Kuemmerlen et al., 2015; Nelson et al., 2009; Sultana et al., 2020). When these models have been used to predict biological responses, they have typically focused on fish and phytoplankton, whereas other important components of food webs, such as macroinvertebrates, have received less attention (Lofton et al., 2023).…”

A Multi‐Model Ensemble of Baseline and Process‐Based Models Improves the Predictive Skill of Near‐Term Lake Forecasts

“…Despite their simplicity, simple empirical models such as persistence and climatology (historical day‐of‐year mean and variance) models can also provide useful forecasts (Ward et al., 2014). Often used as null models to test the skill of emerging forecasting approaches (Lofton et al., 2023; Pappenberger et al., 2015), these simple baseline models include information on current conditions and seasonal trends that influence lake temperature dynamics. For example, a persistence model can be useful for forecasting dynamics in systems with high inertia that exhibit small changes across the forecast horizon (i.e., time into the future; Ward et al., 2014), which is common in lakes and reservoirs that exhibit seasonal thermal stratification.…”

“…To date, process-based models (Baracchini et al, 2020;Clayer et al, 2023;Mercado-Bettín et al, 2021;Thomas et al, 2020), machine learning and data-driven models (Di Nunno et al, 2023;Read et al, 2019;Zwart et al, 2023), as well as "hybrid" approaches (e.g., Saber et al, 2020) have been used to forecast near-term dynamics (days to seasons ahead) in lake and reservoir water temperatures, with varying levels of performance (reviewed by Lofton et al, 2023). Of these modeling approaches, process-based models (hereafter, PMs) have shown substantial promise, especially in near-term forecast horizons (Baracchini et al, 2020;Carey et al, 2022b;Mercado-Bettín et al, 2021;Thomas et al, 2020), with a performance of 0.4-1.4°C RMSE (daily root mean square error) for reservoir water temperature forecasted 1-16 days-ahead (Thomas et al, 2020).…”