The global decline of water quality in rivers and streams has resulted in a pressing need to design new watershed management strategies. Water quality can be affected by multiple stressors including population growth, land use change, global warming, and extreme events, with repercussions on human and ecosystem health. A scientific understanding of factors affecting riverine water quality and predictions at local to regional scales, and at sub‐daily to decadal timescales are needed for optimal management of watersheds and river basins. Here, we discuss how machine learning (ML) can enable development of more accurate, computationally tractable, and scalable models for analysis and predictions of river water quality. We review relevant state‐of‐the art applications of ML for water quality models and discuss opportunities to improve the use of ML with emerging computational and mathematical methods for model selection, hyperparameter optimization, incorporating process knowledge into ML models, improving explainablity, uncertainty quantification, and model‐data integration. We then present considerations for using ML to address water quality problems given their scale and complexity, available data and computational resources, and stakeholder needs. When combined with decades of process understanding, interdisciplinary advances in knowledge‐guided ML, information theory, data integration, and analytics can help address fundamental science questions and enable decision‐relevant predictions of riverine water quality.

This paper provides initial results from a multi-model ensemble analysis based on the volc-pinatubo-full experiment performed within the Model Intercomparison Project on the climatic response to Volcanic forcing (VolMIP) as part of the sixth phase of the Coupled Model Intercomparison Project (CMIP6). The volc-pinatubo-full experiment is based on an ensemble of volcanic forcing-only climate simulations with the same volcanic aerosol dataset across the participating models (the 1991–1993 Pinatubo period from the CMIP6-GloSSAC dataset). The simulations are conducted within an idealized experimental design where initial states are sampled consistently across models from the CMIP6-piControl simulation providing unperturbed preindustrial background conditions. The multi-model ensemble includes output from an initial set of six participating Earth system models (CanESM5, GISS-E2.1-G, IPSL-CM6A-LR, MIROC-E2SL, MPI-ESM1.2-LR and UKESM1). The results show overall good agreement between the different models on the global and hemispheric scales concerning the surface climate responses, thus demonstrating the overall effectiveness of VolMIP's experimental design. However, small yet significant inter-model discrepancies are found in radiative fluxes, especially in the tropics, that preliminary analyses link with minor differences in forcing implementation; model physics, notably aerosol–radiation interactions; the simulation and sampling of El Niño–Southern Oscillation (ENSO); and, possibly, the simulation of climate feedbacks operating in the tropics. We discuss the volc-pinatubo-full protocol and highlight the advantages of volcanic forcing experiments defined within a carefully designed protocol with respect to emerging modelling approaches based on large ensemble transient simulations. We identify how the VolMIP strategy could be improved in future phases of the initiative to ensure a cleaner sampling protocol with greater focus on the evolving state of ENSO in the pre-eruption period.

Researchers in ecology and evolutionary biology are increasingly dependent on computational code to conduct research. Hence, the use of efficient methods to share, reproduce, and collaborate on code as well as document research is fundamental. GitHub is an online, cloud‐based service that can help researchers track, organize, discuss, share, and collaborate on software and other materials related to research production, including data, code for analyses, and protocols. Despite these benefits, the use of GitHub in ecology and evolution is not widespread. To help researchers in ecology and evolution adopt useful features from GitHub to improve their research workflows, we review 12 practical ways to use the platform. We outline features ranging from low to high technical difficulty, including storing code, managing projects, coding collaboratively, conducting peer review, writing a manuscript, and using automated and continuous integration to streamline analyses. Given that members of a research team may have different technical skills and responsibilities, we describe how the optimal use of GitHub features may vary among members of a research collaboration. As more ecologists and evolutionary biologists establish their workflows using GitHub, the field can continue to push the boundaries of collaborative, transparent, and open research.

Stream temperature ($T_s$) is an important water quality parameter that affects ecosystem health and human water use for beneficial purposes. Accurate $T_s$ predictions at different spatial and temporal scales can inform water management decisions that account for the effects of changing climate and extreme events. In particular, widespread predictions of $T_s$ in unmonitored stream reaches can enable decision makers to be responsive to changes caused by unforeseen disturbances. In this study, we demonstrate the use of classical machine learning (ML) models, support vector regression and gradient boosted trees (XGBoost), for monthly $T_s$ predictions in 78 pristine and human-impacted catchments of the Mid-Atlantic and Pacific Northwest hydrologic regions spanning different geologies, climate, and land use. The ML models were trained using long-term monitoring data from 1980–2020 for three scenarios: (1) temporal predictions at a single site, (2) temporal predictions for multiple sites within a region, and (3) spatiotemporal predictions in unmonitored basins (PUB). In the first two scenarios, the ML models predicted $T_s$ with median root mean squared errors (RMSE) of 0.69–0.84 °C and 0.92–1.02 °C across different model types for the temporal predictions at single and multiple sites respectively. For the PUB scenario, we used a bootstrap aggregation approach using models trained with different subsets of data, for which an ensemble XGBoost implementation outperformed all other modeling configurations (median RMSE 0.62 °C).The ML models improved median monthly $T_s$ estimates compared to baseline statistical multi-linear regression models by 15–48% depending on the site and scenario. Air temperature was found to be the primary driver of monthly $T_s$ for all sites, with secondary influence of month of the year (seasonality) and solar radiation, while discharge was a significant predictor at only 10 sites. The predictive performance of the ML models was robust to configuration changes in model setup and inputs, but was influenced by the distance to the nearest dam with RMSE <1 °C at sites situated greater than 16 and 44 km from a dam for the temporal single site and regional scenarios, and over 1.4 km from a dam for the PUB scenario. Our results show that classical ML models with solely meteorological inputs can be used for spatial and temporal predictions of monthly $T_s$ in pristine and managed basins with reasonable (<1 °C) accuracy for most locations.

Data standardization combined with descriptive metadata facilitate data reuse, which is the ultimate goal of the Findable, Accessible, Interoperable, and Reusable (FAIR) principles. Community data or metadata standards are increasingly created through an approach that emphasizes collaboration between various stakeholders. Such an approach requires platforms for collaboration on the development process that centers on sharing information and receiving feedback. Our objective in this study was to conduct a systematic review to identify data standards and reporting formats that use version control for developing data standards and to summarize common practices, particularly in earth and environmental sciences. Out of 108 data standards and reporting formats identified in our review, 32 used GitHub as the version control platform, and no other platforms were used. We found no universally accepted methodology for developing and publishing data standards. Many GitHub repositories did not use key features that could help developers to gather user feedback, or to create and revise standards that build on previous work. We provide guidance for community‐driven standard development and associated documentation on GitHub based on a systematic review of existing practices.

Strong, strato-volcanic eruptions are a substantial, intermittent source of natural climate variability. Initial atmospheric and oceanic conditions, such as El Niño Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO), also naturally impact climate on interannual timescales. We examine how initial conditions of ENSO and NAO contribute to the evolution of climate in the period following a Pinatubo-type eruption using a large (81-member) ensemble of model simulations in GISS model E2.1-G. Simulations are initialized from sampled conditions of ENSO and NAO using the protocol of the coordinated CMIP6 Volcanic Model Intercomparison Project (VolMIP) – where aerosols are forced with respect to time, latitude, and height. We analyze paired anomalous variations (perturbed – control) to understand changes in global and regional climate responses under positive, negative, and neutral ENSO and NAO conditions. In particular, we find that for paired anomalies there is a high probability of strong (∼1.5 ^∘C) warming of northern Eurasia surface air temperature in the first winter after the volcanic eruption for negative NAO ensembles coincident with decreased lower stratospheric temperature at the poles, decreased geopotential height, and strengthening of the stratospheric polar vortex. Climate anomalies (relative to average conditions across the control period), however, show no mean warming and suggest that the strength of this response is impacted by conditions present in the selected period of the control run. Again using paired anomalies, we also observe that under both +ENSO and −ENSO ensembles sea surface temperature decreases in the first post-eruptive boreal winter coinciding with surface cooling from volcanic aerosols. Neutral ENSO ensembles, on the other hand, show variability in their response with no clear trend in post-eruptive warming or cooling. In general, paired anomalies from unperturbed simulations give insight into the evolution of the climate response to volcanic forcing; however, when compared with anomalies from climatological conditions, it is clear that paired anomalies are significantly affected by sampled initial conditions occurring at the time of the volcanic eruption.