Geology topics

Samantha K. Oliver

Publications and source records attributed to Samantha K. Oliver.

At least 19 recordsLinked to original sources

USGS2026

Predicting Minnesota lake ice phenology with deep learning, explainable methods, and a physically based benchmark, 1980-2018

Globally, lakes are losing ice cover, but our understanding of the rates and variability of change are biased toward few lakes with long-term records. Recent works have used remote sensing and predictive modeling to supplement the observational record, but these approaches produce large errors when estimating ice phenology for individual lakes. Accurate, lake-specific estimates of ice formation and breakup through time are important for estimating variability in ice loss across heterogenous lakes and for understanding broader implications. This work explores machine learning (ML) approaches for hindcasting ice phenology using observations from 1980 to 2018 across 625 Minnesota lakes, covering 4359 lake-years of record. We used daily weather and static lake attributes to develop 60 neural networks for hindcasting lake ice time series. We considered LSTMs and attention-based transformers of varying size and initial parameters. We found that the largest LSTM performed most accurately on withheld data, and on a test set of unseen years and lakes, it outperformed a state-of-the-art physically based model in daily accuracy (97% vs. 95%), year-level metrics (RMSE for ice formation = 6.9 vs. 13.2 days; ice breakup = 7.5 vs. 12.2 days; ice duration = 9.6 vs. 13.7 days), and estimating loss of ice cover from 1980 to 2018 (observed = 8.4 days, LSTM = 7.2 days, GLM = 3.4 days). This demonstrates that ML can estimate historical lake ice formation and breakup dates within a week of observed phenology across heterogeneous lakes and recreate broad scale trends in ice cover phenology with limited long-term ice records.

Minnesota↗

USGS2025

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

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., R 2 for annual 7-day maximum = 0.76; R 2 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.

contiguous United States↗

USGS2024

Lake water temperature modeling in an era of climate change: Data sources, models, and future prospects

Lake thermal dynamics have been considerably impacted by climate change, with potential adverse effects on aquatic ecosystems. To better understand the potential impacts of future climate change on lake thermal dynamics and related processes, the use of mathematical models is essential. In this study, we provide a comprehensive review of lake water temperature modeling. We begin by discussing the physical concepts that regulate thermal dynamics in lakes, which serve as a primer for the description of process-based models. We then provide an overview of different sources of observational water temperature data, including in situ monitoring and satellite Earth observations, used in the field of lake water temperature modeling. We classify and review the various lake water temperature models available, and then discuss model performance, including commonly used performance metrics and optimization methods. Finally, we analyze emerging modeling approaches, including forecasting, digital twins, combining process-based modeling with deep learning, evaluating structural model differences through ensemble modeling, adapted water management, and coupling of climate and lake models. This review is aimed at a diverse group of professionals working in the fields of limnology and hydrology, including ecologists, biologists, physicists, engineers, and remote sensing researchers from the private and public sectors who are interested in understanding lake water temperature modeling and its potential applications.

Review of Geophysics↗

USGS2023

GRiMeDB: The Global River Database Methane Database of concentrations and fluxes

Despite their small spatial extent, fluvial ecosystems play a significant role in processing and transporting carbon in aquatic networks, which results in substantial emission of methane ( CH 4 ) into the atmosphere. For this reason, considerable effort has been put into identifying patterns and drivers of CH 4 concentrations in streams and rivers and estimating fluxes to the atmosphere across broad spatial scales. However, progress toward these ends has been slow because of pronounced spatial and temporal variability of lotic CH 4 concentrations and fluxes and by limited data availability across diverse habitats and physicochemical conditions. To address these challenges, we present a comprehensive database of CH 4 concentrations and fluxes for fluvial ecosystems along with broadly relevant and concurrent physical and chemical data. The Global River Methane Database (GriMeDB; https://doi.org/10.6073/pasta/f48cdb77282598052349e969920356ef , Stanley et al., 2023) includes 24 024 records of CH 4 concentration and 8205 flux measurements from 5029 unique sites derived from publications, reports, data repositories, unpublished data sets, and other outlets that became available between 1973 and 2021. Flux observations are reported as diffusive, ebullitive, and total CH 4 fluxes, and GriMeDB also includes 17 655 and 8409 concurrent measurements of concentrations and 4444 and 1521 fluxes for carbon dioxide ( CO 2 ) and nitrous oxide ( N 2 O ), respectively. Most observations are date-specific (i.e., not site averages), and many are supported by data for 1 or more of 12 physicochemical variables and 6 site variables. Site variables include codes to characterize marginal channel types (e.g., springs, ditches) and/or the presence of human disturbance (e.g., point source inputs, upstream dams). Overall, observations in GRiMeDB encompass the broad range of the climatic, biological, and physical conditions that occur among world river basins, although some geographic gaps remain (arid regions, tropical regions, high-latitude and high-altitude systems). The global median CH 4 concentration (0.20 µmol L −1 ) and diffusive flux (0.44 mmol m -2 d -1 ) in GRiMeDB are lower than estimates from prior site-averaged compilations, although ranges (0 to 456 µmol L −1 and − 136 to 4057 mmol m -2 d -1 ) and standard deviations (10.69 and 86.4) are greater for this larger and more temporally resolved database. Available flux data are dominated by diffusive measurements despite the recognized importance of ebullitive and plant-mediated CH 4 fluxes. Nonetheless, GriMeDB provides a comprehensive and cohesive resource for examining relationships between CH 4 and environmental drivers, estimating the contribution of fluvial ecosystems to CH 4 emissions, and contextualizing site-based investigations.

Earth System Science Data↗

USGS2023

Evaluating deep learning architecture and data assimilation for improving water temperature forecasts at unmonitored locations

Deep learning (DL) models are increasingly used to forecast water quality variables for use in decision making. Ingesting recent observations of the forecasted variable has been shown to greatly increase model performance at monitored locations; however, observations are not collected at all locations, and methods are not yet well developed for DL models for optimally ingesting recent observations from other sites to inform focal sites. In this paper, we evaluate two different DL model structures, a long short-term memory neural network (LSTM) and a recurrent graph convolutional neural network (RGCN), both with and without data assimilation for forecasting daily maximum stream temperature 7 days into the future at monitored and unmonitored locations in a 70-segment stream network. All our DL models performed well when forecasting stream temperature as the root mean squared error (RMSE) across all models ranged from 2.03 to 2.11°C for 1-day lead times in the validation period, with substantially better performance at gaged locations (RMSE = 1.45–1.52°C) compared to ungaged locations (RMSE = 3.18–3.27°C). Forecast uncertainty characterization was near-perfect for gaged locations but all DL models were overconfident (i.e., uncertainty bounds too narrow) for ungaged locations. Our results show that the RGCN with data assimilation performed best for ungaged locations and especially at higher temperatures (>18°C) which is important for management decisions in our study location. This indicates that the networked model structure and data assimilation techniques may help borrow information from nearby monitored sites to improve forecasts at unmonitored locations. Results from this study can help guide DL modeling decisions when forecasting other important environmental variables.

Frontiers in Water↗

USGS2023

Pesticide prioritization by potential biological effects in tributaries of the Laurentian Great Lakes

Watersheds of the Great Lakes Basin (USA/Canada) are highly modified and impacted by human activities including pesticide use. Despite labeling restrictions intended to minimize risks to nontarget organisms, concerns remain that environmental exposures to pesticides may be occurring at levels negatively impacting nontarget organisms. We used a combination of organismal-level toxicity estimates (in vivo aquatic life benchmarks) and data from high-throughput screening (HTS) assays (in vitro benchmarks) to prioritize pesticides and sites of concern in streams at 16 tributaries to the Great Lakes Basin. In vivo or in vitro benchmark values were exceeded at 15 sites, 10 of which had exceedances throughout the year. Pesticides had the greatest potential biological impact at the site with the greatest proportion of agricultural land use in its basin (the Maumee River, Toledo, OH, USA), with 72 parent compounds or transformation products being detected, 47 of which exceeded at least one benchmark value. Our risk-based screening approach identified multiple pesticide parent compounds of concern in tributaries of the Great Lakes; these compounds included: eight herbicides (metolachlor, acetochlor, 2,4-dichlorophenoxyacetic acid, diuron, atrazine, alachlor, triclopyr, and simazine), three fungicides (chlorothalonil, propiconazole, and carbendazim), and four insecticides (diazinon, fipronil, imidacloprid, and clothianidin). We present methods for reducing the volume and complexity of potential biological effects data that result from combining contaminant surveillance with HTS (in vitro) and traditional (in vivo) toxicity estimates. Environ Toxicol Chem 2022;00:1–18. Published 2022. This article is a U.S. Government work and is in the public domain in the USA. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Great Lakes↗

USGS2022

Near-term forecasts of stream temperature using deep learning and data assimilation in support of management decisions

Deep learning (DL) models are increasingly used to make accurate hindcasts of management-relevant variables, but they are less commonly used in forecasting applications. Data assimilation (DA) can be used for forecasts to leverage real-time observations, where the difference between model predictions and observations today is used to adjust the model to make better predictions tomorrow. In this use case, we developed a process-guided DL and DA approach to make 7-day probabilistic forecasts of daily maximum water temperature in the Delaware River Basin in support of water management decisions. Our modeling system produced forecasts of daily maximum water temperature with an average root mean squared error (RMSE) from 1.1 to 1.4°C for 1-day-ahead and 1.4 to 1.9°C for 7-day-ahead forecasts across all sites. The DA algorithm marginally improved forecast performance when compared with forecasts produced using the process-guided DL model alone (0%–14% lower RMSE with the DA algorithm). Across all sites and lead times, 65%–82% of observations were within 90% forecast confidence intervals, which allowed managers to anticipate probability of exceedances of ecologically relevant thresholds and aid in decisions about releasing reservoir water downstream. The flexibility of DL models shows promise for forecasting other important environmental variables and aid in decision-making.

Journal of the American Water Resources Associatio↗

USGS2022

Prioritizing pesticides of potential concern and identifying potential mixture effects in Great Lakes tributaries using passive samplers

To help meet the objectives of the Great Lakes Restoration Initiative with regard to increasing knowledge about toxic substances, 223 pesticides and pesticide transformation products were monitored in 15 Great Lakes tributaries using polar organic chemical integrative samplers. A screening-level assessment of their potential for biological effects was conducted by computing toxicity quotients (TQs) for chemicals with available US Environmental Protection Agency (USEPA) Aquatic Life Benchmark values. In addition, exposure activity ratios (EAR) were calculated using information from the USEPA ToxCast database. Between 16 and 81 chemicals were detected per site, with 97 unique compounds detected overall, for which 64 could be assessed using TQs or EARs. Ten chemicals exceeded TQ or EAR levels of concern at two or more sites. Chemicals exceeding thresholds included seven herbicides (2,4-dichlorophenoxyacetic acid, diuron, metolachlor, acetochlor, atrazine, simazine, and sulfentrazone), a transformation product (deisopropylatrazine), and two insecticides (fipronil and imidacloprid). Watersheds draining agricultural and urban areas had more detections and higher concentrations of pesticides compared with other land uses. Chemical mixtures analysis for ToxCast assays associated with common modes of action defined by gene targets and adverse outcome pathways (AOP) indicated potential activity on biological pathways related to a range of cellular processes, including xenobiotic metabolism, extracellular signaling, endocrine function, and protection against oxidative stress. Use of gene ontology databases and the AOP knowledgebase within the R-package ToxMixtures highlighted the utility of ToxCast data for identifying and evaluating potential biological effects and adverse outcomes of chemicals and mixtures. Results have provided a list of high-priority chemicals for future monitoring and potential biological effects warranting further evaluation in laboratory and field environments. Environ Toxicol Chem 2023;42:340–366. Published 2022. This article is a U.S. Government work and is in the public domain in the USA. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Great Lakes tributaries↗

USGS2022

Can machine learning accelerate process understanding and decision-relevant predictions of river water quality?

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 for 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.

Hydrological Processes↗

USGS2022

Modeling reservoir release using pseudo-prospective learning and physical simulations to predict water temperature

This paper proposes a new data-driven method for predicting water temperature in stream networks with reservoirs. The water flows released from reservoirs greatly affect the water temperature of downstream river segments. However, the information of released water flow is often not available for many reservoirs, which makes it difficult for data-driven models to capture the impact to downstream river segments. In this paper, we first build a state-aware graph model to represent the interactions amongst streams and reservoirs, and then propose a parallel learning structure to extract the reservoir release information and use it to improve the prediction. In particular, for reservoirs with no available release information, we mimic the water managers' release decision process through a pseudo-prospective learning method, which infers the release information from anticipated water temperature dynamics. For reservoirs with the release information, we leverage a physics-based model to simulate the water release temperature and transfer such information to guide the learning process for other reservoirs. The evaluation for the Delaware River Basin shows that the proposed method brings over 10% accuracy improvement over existing data-driven models for stream temperature prediction when the release data is not available for any reservoirs. The performance is further improved after we incorporate the release data and physical simulations for a subset of reservoirs.

Conference Paper↗

USGS2022

Teams, networks, and networks of networks advancing our understanding and conservation of inland waters

Networks are defined as groups of interconnected people and things, and by this definition, networks play a major role in the science of inland waters. In this article, we bring the latest social network research to understand and improve inland waters science and conservation outcomes. What we found is that relationships matter. Different teams and networks have different objectives and lifespans. Consider this: Data collection networks may persist for decades, whereas knowledge-generating teams may exist only for months. The structure of connections in a network determines how easily information or resources can flow or pass through a network, which then influences the ability of the network to accomplish work like creating and applying new knowledge, integrating knowledge across fields, or coordinating collective action. When independent networks designed around different purposes become connected to achieve new goals, a network of networks is formed, where each layer is a unique network defined by social, geographic, and temporal boundaries and distinct types of connections. This structure has a lot of potential for transformative work, but is especially susceptible to failure if one of the cross-network connections fails. From the smallest of inland waters research teams to the largest, multi-institutional, international collaborations, an understanding of how the connections between people are created and maintained can be used to set up conditions for success.

Book chapter↗

USGS2022

Machine learning for understanding inland water quantity, quality, and ecology

This chapter provides an overview of machine learning models and their applications to the science of inland waters. Such models serve a wide range of purposes for science and management: predicting water quality, quantity, or ecological dynamics across space, time, or hypothetical scenarios; vetting and distilling raw data for further modeling or analysis; generating and exploring hypotheses; estimating physically or biologically meaningful parameters for use in further modeling; and revealing patterns in complex, multidimensional data or model outputs. An important research frontier is the injection of limnological knowledge into machine-learning models, which has shown great promise for increasing such models’ accuracy, trustworthiness, and interpretability. Here we describe a few of the most powerful machine learning tools, describe best practices for employing these tools and injecting knowledge guidance, and give examples of their applications to advance understanding of inland waters.

Book chapter↗

USGS2022

Multi-task deep learning of daily streamflow and water temperature

Deep learning (DL) models can accurately predict many hydrologic variables including streamflow and water temperature; however, these models have typically predicted hydrologic variables independently. This study explored the benefits of modeling two interdependent variables, daily average streamflow and daily average stream water temperature, together using multi-task DL. A multi-task scaling factor controlled the relative contribution of the auxiliary variable's error to the overall loss during training. Our experiments examined the improvement in prediction accuracy of the multi-task approach using paired streamflow and water temperature data from sites across the conterminous United States. Our results showed that for 56 out of 101 sites, the best performing multi-task models performed better overall than the single-task models in terms of Nash-Sutcliffe efficiency for predicting streamflow with single-site models. For 43 sites, the best multi-task, single-site models made no significant difference in predicting streamflow. The multi-task approach had a smaller effect when applied to a model trained with data from 101 sites together, significantly improving performance for only 17 sites. The multi-task scaling factor was consequential in determining to what extent the multi-task approach was beneficial. A naïve selection of this factor led to significantly worse-performing models for 3 of 101 sites when predicting streamflow as the primary variable, and 47 of 53 sites when predicting stream temperature as the primary variable. We conclude that a multi-task approach can make more accurate predictions by leveraging information from interdependent hydrologic variables, but only for some sites, variables, and model configurations.

Water Resources Research↗

USGS2022

Physics-guided machine learning from simulation data: An application in modeling lake and river systems

This paper proposes a new physics-guided machine learning approach that incorporates the scientific knowledge in physics-based models into machine learning models. Physics-based models are widely used to study dynamical systems in a variety of scientific and engineering problems. Although they are built based on general physical laws that govern the relations from input to output variables, these models often produce biased simulations due to inaccurate parameterizations or approximations used to represent the true physics. In this paper, we aim to build a new data-driven framework to monitor dynamical systems by extracting general scientific knowledge embodied in simulation data generated by the physics-based model. To handle the bias in simulation data caused by imperfect parameterization, we propose to extract general physical relations jointly from multiple sets of simulations generated by a physics-based model under different physical parameters. In particular, we develop a spatio-temporal network architecture that uses its gating variables to capture the variation of physical parameters. We initialize this model using a pre-training strategy that helps discover common physical patterns shared by different sets of simulation data. Then we fine-tune it using limited observation data via a contrastive learning process. By leveraging the complementary strength of machine learning and domain knowledge, our method has been shown to produce accurate predictions, use less training samples and generalize to out-of-sample scenarios. We further show that the method can provide insights about the variation of physical parameters over space and time in two domain applications: predicting temperature in streams and predicting temperature in lakes.

Conference Paper↗

USGS2021

Deep learning approaches for improving prediction of daily stream temperature in data-scarce, unmonitored, and dammed basins

Basin-centric long short-term memory (LSTM) network models have recently been shown to be an exceptionally powerful tool for stream temperature (T s ) temporal prediction (training in one period and predicting in another period at the same sites). However, spatial extrapolation is a well-known challenge to modelling T s and it is uncertain how an LSTM-based daily T s model will perform in unmonitored or dammed basins. Here we compiled a new benchmark dataset consisting of >400 basins across the contiguous United States in different data availability groups (DAG, meaning the daily sampling frequency) with and without major dams, and studied how to assemble suitable training datasets for predictions in basins with or without temperature monitoring. For prediction in unmonitored basins (PUB), LSTM produced a root-mean-square error (RMSE) of 1.129°C and an R 2 of 0.983. While these metrics declined from LSTM's temporal prediction performance, they far surpassed traditional models' PUB values, and were competitive with traditional models' temporal prediction on calibrated sites. Even for unmonitored basins with major reservoirs, we obtained a median RMSE of 1.202°C and an R 2 of 0.984. For temporal prediction, the most suitable training set was the matching DAG that the basin could be grouped into (for example, the 60% DAG was most suitable for a basin with 61% data availability). However, for PUB, a training dataset including all basins with data was consistently preferred. An input-selection ensemble moderately mitigated attribute overfitting. Our results indicate there are influential latent processes not sufficiently described by the inputs (e.g., geology, wetland covers), but temporal fluctuations can still be predicted well, and LSTM appears to be a highly accurate T s modelling tool even for spatial extrapolation.

Hydrological Processes↗

USGS2021

Integrating ecosystem metabolism and consumer allochthony reveals nonlinear drivers in lake organic matter processing

Lakes process both terrestrial and aquatic organic matter, and the relative contribution from each source is often measured via ecosystem metabolism and terrestrial resource use in the food web (i.e., consumer allochthony). Yet, ecosystem metabolism and consumer allochthony are rarely considered together, despite possible interactions and potential for them to respond to the same lake characteristics. In this study, we compiled global datasets of lake gross primary production (GPP), ecosystem respiration (ER), and zooplankton allochthony to compare the strength and shape of relationships with physicochemical characteristics across a broad set of lakes. GPP was positively related to total phosphorus (TP) in lakes with intermediate TP concentrations (11–75 μ g L −1 ) and was highest in lakes with intermediate dissolved organic carbon (DOC) concentrations. While ER and GPP were strongly positively correlated, decoupling occurred at high DOC concentrations. Lastly, allochthony had a unimodal relationship with TP and related variably to DOC. By integrating metabolism and allochthony, we identified similar change points in GPP and zooplankton allochthony at intermediate DOC (4.5–10 mg L −1 ) and TP (8–20 μ g L −1 ) concentrations, indicating that allochthony and GPP may be coupled and inversely related. The ratio of DOC:nutrients also helped to identify conditions where lake organic matter processing responded more to autochthonous or allochthonous organic matter sources. As lakes globally face eutrophication and browning, predicting how lake organic matter processing will respond requires an updated paradigm that incorporates nonlinear dynamics and interactions.

Limnology and Oceanography↗

USGS2021

Predicting water temperature dynamics of unmonitored lakes with meta-transfer learning

Most environmental data come from a minority of well-monitored sites. An ongoing challenge in the environmental sciences is transferring knowledge from monitored sites to unmonitored sites. Here, we demonstrate a novel transfer-learning framework that accurately predicts depth-specific temperature in unmonitored lakes (targets) by borrowing models from well-monitored lakes (sources). This method, meta-transfer learning (MTL), builds a meta-learning model to predict transfer performance from candidate source models to targets using lake attributes and candidates' past performance. We constructed source models at 145 well-monitored lakes using calibrated process-based (PB) modeling and a recently developed approach called process-guided deep learning (PGDL). We applied MTL to either PB or PGDL source models (PB-MTL or PGDL-MTL, respectively) to predict temperatures in 305 target lakes treated as unmonitored in the Upper Midwestern United States. We show significantly improved performance relative to the uncalibrated PB General Lake Model, where the median root mean squared error (RMSE) for the target lakes is 2.52°C. PB-MTL yielded a median RMSE of 2.43°C; PGDL-MTL yielded 2.16°C; and a PGDL-MTL ensemble of nine sources per target yielded 1.88°C. For sparsely monitored target lakes, PGDL-MTL often outperformed PGDL models trained on the target lakes themselves. Differences in maximum depth between the source and target were consistently the most important predictors. Our approach readily scales to thousands of lakes in the Midwestern United States, demonstrating that MTL with meaningful predictor variables and high-quality source models is a promising approach for many kinds of unmonitored systems and environmental variables.

Water Resources Research↗

USGS2021

Physics-guided recurrent graph model for predicting flow and temperature in river networks

This paper proposes a physics-guided machine learning approach that combines machine learning models and physics-based models to improve the prediction of water flow and temperature in river networks. We first build a recurrent graph network model to capture the interactions among multiple segments in the river network. Then we transfer knowledge from physics-based models to guide the learning of the machine learning model. We also propose a new loss function that balances the performance over different river segments. We demonstrate the effectiveness of the proposed method in predicting temperature and streamflow in a subset of the Delaware River Basin. In particular, the proposed method has brought a 33%/14% accuracy improvement over the state-of-the-art physics-based model and 24%/14% over traditional machine learning models (e.g., LSTM) in temperature/streamflow prediction using very sparse (0.1%) training data. The proposed method has also been shown to produce better performance when generalized to different seasons or river segments with different streamflow ranges.

Conference Paper↗