Geology Reports⌕ Search

Geology topics

Jacob Aaron Zwart

Publications and source records attributed to Jacob Aaron Zwart.

At least 19 recordsLinked to original sources
USGS2026

Machine learning generated streamflow drought forecasts for the conterminous United States (CONUS): developing and evaluating an operational tool to enhance sub-seasonal to seasonal streamflow drought early warning for gaged locations

Forecasts of streamflow drought, when streamflow declines below typical levels, are notably less available than for floods or meteorological drought, despite widespread impacts. We apply machine learning (ML) models to forecast streamflow drought 1–13 weeks ahead at 3,219 streamgages across the conterminous United States. We applied two ML methods (Long short-term memory neural networks; Light Gradient-Boosting Machine) and two benchmark models (persistence; Autoregressive Integrated Moving Average) to predict weekly streamflow percentiles with independent models for each forecast horizon. ML models outperformed benchmarks in predicting continuous streamflow percentiles below 30%. ML models generally performed worse than persistence models for discrete classification (moderate, severe, extreme) but exceeded the benchmark models for drought onset/termination. Performance was better for less intense droughts and shorter horizons, with predictive power for 1–4 weeks for severe droughts (10% threshold). This work highlights challenges and opportunities to advance hydrological drought forecasting and supports a new experimental forecasting tool.

conterminous United States↗
USGS2026

Fifty years of riverine harmful algal bloom modeling: A global synthesis of approaches, challenges, and opportunities

This systematic literature review critically examines 162 articles on harmful algal bloom (HAB) modeling in riverine systems to uncover persistent gaps, redefine critical challenges, and propose trackable opportunities to advance future modeling efforts. Articles largely focused on site-specific applications (93%) across more than 80 rivers worldwide. Most modeled systems were large, eutrophic rivers with flow modifications or obstructions. Geographic clustering of modeled systems was pronounced, with South Korea accounting for 26% of articles, followed by Europe (25%), United States (21%), and China (12%). Modeling approaches were led by process-based models (59%), though use of data-driven models (37%) increased over time, reflecting advances in computing and monitoring technology. Modeling endpoints varied widely across the articles with many focused on gross measures of algal abundance and fewer representing more refined endpoints like algal toxins or community composition. Furthermore, inconsistent units and taxonomic resolution hindered comparability between models. Datasets used for model development and calibration typically spanned 5 years, with weekly to monthly sampling at 1–10 sites, though durations and site counts were positively skewed. Quantitative metrics of model skill were often absent and included a diverse set of metrics when reported. Across all models, nutrients, light availability, streamflow, algal physiological processes, and water temperature emerged as key predictors, though algal processes were rarely incorporated in data-driven models. Scenario analyses primarily were conducted with process-based models and addressed flow management, whereas forecasting applications were less common and typically used data-driven models. After almost 50 years of riverine HAB modeling, persistent challenges include underrepresentation of benthic habitats, neglect of side-channel and backwater influences, insufficient documentation of river features, and weak linkages between modeled endpoints and potential harms. Addressing these gaps through reporting of contextual information, models from other aquatic settings, benchmark datasets, and community-driven tools could advance riverine HAB modeling towards increased transferability and ultimately operational forecasts.

Water Research↗
USGS2026

How to accelerate advances in ecological forecasting

Ecological forecasting offers critical insights for managing natural resources and safeguarding public well-being. Despite growing demand for these forecasts, progress is hindered by fragmented systems, redundant workflows, and limited interoperability. Drawing lessons from weather forecasting and recent successes like the NEON Ecological Forecasting Challenge, shared cyberinfrastructure is important for advancing ecological prediction. By adopting common standards, open-source tools, and scalable architectures, and fostering transdisciplinary collaboration, the ecological forecasting community can overcome technical and institutional barriers. Such investments could accelerate scientific understanding, improve forecast reliability, and empower decisionmakers to anticipate environmental change and respond effectively.

Eos, American Geophysical Union↗
USGS2026

A hybrid approach for revealing headwater hydrology

Coordinated work to compile existing data and apply models pairing physical understanding with machine learning could substantially improve streamflow predictions for little-known headwater basins.

Eos, American Geophysical Union↗
USGS2025

Designs for cyanobacterial harmful algal bloom monitoring in the Sacramento–San Joaquin Delta, California

Cyanobacterial harmful algal blooms (CHABs) are a growing concern in freshwater environments. These blooms can lead to degraded water quality, ecosystem disruptions, and public health threats due to the production of potent cyanotoxins. The Sacramento–San Joaquin Delta (California, USA; the Delta) has experienced CHABs since 1999, including CHABs that produce cyanotoxins at concentrations exceeding recreational advisory threshold levels. In response to the CHAB monitoring needs in the Delta, in 2024 a CHAB monitoring strategy for the Sacramento–San Joaquin Delta was released. The CHAB monitoring strategy recommended implementing a comprehensive monitoring program aligned with regional priorities, and this report responds directly to that recommendation. Building on the CHAB monitoring strategy, it describes three example monitoring designs that draw from past studies to address present needs. To develop designs, the report outlines key components and considerations for implementing a CHAB monitoring program in the Delta. It provides: background on monitoring and modeling approaches, an overview of current monitoring programs in the region, guidance for selecting monitoring locations, indicators, and sampling frequencies, and example designs to guide program planning and budgeting. The overarching goal of the report is to help agencies and stakeholders move from high-level recommendations to practical, actionable designs that are tailored to the Delta’s physical, anthropogenic, and ecological landscape.

ESS Open Archive↗
USGS2025

Physics-guided fair graph sampling for water temperature prediction in river networks

This work introduces a novel graph neural networks (GNNs)-based method to predict stream water temperature and reduce model bias across locations of different income and education levels. Traditional physics-based models often have limited accuracy because they are necessarily approximations of reality. Recently, there has been an increasing interest of using GNNs in modeling complex water dynamics in stream networks. Despite their promise in improving the accuracy, GNNs can bring additional model bias through the aggregation process, where node features are updated by aggregating neighboring nodes. The bias can be especially pronounced when nodes with similar sensitive attributes are frequently connected. We introduce a new method that leverages physical knowledge to represent the node influence in GNNs, and then utilizes physics-based influence to refine the selection and weights over the neighbors. The objective is to facilitate equitable treatment over different sensitive groups in the graph aggregation, which helps reduce spatial bias over locations, especially for those in underprivileged groups. The results on the Delaware River Basin demonstrate the effectiveness of the proposed method in preserving equitable performance across locations in different sensitive groups.

Proceedings of the AAAI Conference on Artificial I↗
USGS2025

A systematic literature review of forecasting and predictive models of harmful algal blooms in flowing waters

Occurrences of harmful algal blooms (HABs) in rivers challenge the belief that rivers are not susceptible to HABs because of their short residence times and fluctuating hydrology. Here we present a systematic literature review of predictive and forecasting models for HABs in flowing waters, including rivers, flowing in-stream reservoirs (e.g., run-of-river reservoirs and lock-and-dam systems) and tidal or estuarine systems with riverine processes. The review aimed to understand current and historical modeling approaches for predicting and forecasting river HABs, without restricting to specific taxa, such as cyanobacteria, or modeling endpoints. The review included 162 articles published over nearly 50 years, covering more than 80 rivers worldwide. Eutrophic, non-wadable rivers with in-stream obstruction were commonly modeled, though diverse environmental characteristics were reported. Most articles used algal biomass or chlorophyll as modeling endpoints, with a quarter using novel or unique endpoints. Algal toxins motivated model development in 23% of the articles, however just 5% used algal toxins as an endpoint. Only 6% of the articles modeled benthic HABs; the rest focused on pelagic HABs. There was no standard model used for modeling river HABs. Process-based models were more common (59%) than data-driven approaches (37%), with model formulations ranging from simple to complex, which contrasts with a lake-focused literature review of HAB models that found data-driven models were more common. Models in river settings shared similar input variables as those previously identified for lakes, such as water temperature, nutrients, and light availability. However, streamflow and other transport metrics took prominence in river models compared to lake models. Algal cell physiology (such as growth, predation, and motility) was routinely included as input data or as mathematical formulations in process-based models and these processes were frequently identified as an important predictor by the articles’ authors. Conversely, data-driven models rarely included these processes, instead using predictors related to environmental conditions, such as nutrients, water quality, water temperature, and streamflow. These important proxy predictors have apparent success with modeling overall algal biomass (irrespective of taxa) whereas other factors, such as those related to algal physiology and other biological processes, are likely responsible for more subtle shifts in community composition. These differences highlight the influence of data availability, especially for processes that are difficult, time-consuming, or expensive to measure, on model development and model outcomes, raising questions about the selection of modeling inputs and endpoints. Challenges to advancing river HAB modeling include the lack of site-specific model inputs representing key processes (e.g., photosynthetic parameters and predation rates), overlooked riverine environments like the benthos and side/back-channel areas, lack of information on environmental settings, and poorly reported model performance metrics. This review emphasizes opportunities for advancing river HAB modeling by learning from well-honed estuarine models, supporting current forecasting and operationalization efforts, and developing common datasets for river HAB model development and evaluation.

BioRxiv↗
USGS2025

Machine learning generated streamflow drought forecasts for the Conterminous United States (CONUS): Developing and evaluating an operational tool to enhance sub-seasonal to seasonal streamflow drought early warning for gaged locations

Forecasts of streamflow drought, when streamflow declines below typical levels, are notably less available than for floods or meteorological drought, despite widespread impacts. To address this gap, we apply machine learning (ML) models to forecast streamflow drought 1-13 weeks into the future at > 3,000 streamgage locations across the conterminous United States (CONUS). We applied two ML methods (Long short-term memory (LSTM) neural networks; Light Gradient-Boosting Machine - LightGBM) and two benchmark model approaches (persistence; Autoregressive Integrated Moving Average - ARIMA) to predict weekly streamflow percentiles with independent models for each forecast horizon. To explore whether a training focus on dry weeks improved performance, both ML models were trained using all percentiles (LSTM-all, LightGBM-all) and only percentiles below 30% (LSTM<30, LightGBM<30). We evaluated model performance regionally and nationally for drought occurrence (the classification performance for a future date) and for drought onset/termination (performance identifying drought starts and ends). ML models generally performed worse than the persistence model for discrete classification (moderate, severe, extreme drought) of drought occurrence but exceeded the benchmark models for onset/termination. ML models outperformed benchmarks in predicting continuous streamflow percentiles below 30%. Occurrence performance was better for less intense droughts and shorter forecast horizons, with the ML models having predictive power at 1-4 week horizons for severe droughts (10th percentile threshold). All models struggled to forecast onset, though the best ML model was the LSTM<30 (sensitivity of 22%). Termination performance was greater, with the drought termination performance greatest for the LightGBM-all model. When estimating model uncertainty, the LSTM<30 model had the narrowest 90% percentile interval with closest to optimal capture. This work highlights the challenges and opportunities to further advance hydrological drought forecasting and supports an experimental operational streamflow drought assessment and forecast tool.

Continental United States↗
USGS2024

Near-term ecological forecasting for climate change action

A substantial increase in predictive capacity is needed to anticipate and mitigate the widespread change in ecosystems and their services in the face of climate and biodiversity crises. In this era of accelerating change, we cannot rely on historical patterns or focus primarily on long-term projections that extend decades into the future. In this Perspective, we discuss the potential of near-term (daily to decadal) iterative ecological forecasting to improve decision-making on actionable time frames. We summarize the current status of ecological forecasting and focus on how to scale up, build on lessons from weather forecasting, and take advantage of recent technological advances. We also highlight the need to focus on equity, workforce development, and broad cross-disciplinary and non-academic partnerships.

Nature Climate Change↗
USGS2024

Fair graph learning using constraint-aware priority adjustment and graph masking in river networks

Accurate prediction of water quality and quantity is crucial for sustainable development and human well-being. However, existing data-driven methods often suffer from spatial biases in model performance due to heterogeneous data, limited observations, and noisy sensor data. To overcome these challenges, we propose Fair-Graph, a novel graph-based recurrent neural network that leverages interrelated knowledge from multiple rivers to predict water flow and temperature within large-scale stream networks. Additionally, we introduce node-specific graph masks for information aggregation and adaptation to enhance prediction over heterogeneous river segments. To reduce performance disparities across river segments, we introduce a centralized coordination strategy that adjusts training priorities for segments. We evaluate the prediction of water temperature within the Delaware River Basin, and the prediction of streamflow using simulated data from U.S. National Water Model in the Houston River network. The results showcase improvements in predictive performance and highlight the proposed model's ability to maintain spatial fairness over different river segments.

Delaware River basin↗
USGS2024

National-scale remotely sensed lake trophic state from 1984 through 2020

Lake trophic state is a key ecosystem property that integrates a lake’s physical, chemical, and biological processes. Despite the importance of trophic state as a gauge of lake water quality, standardized and machine-readable observations are uncommon. Remote sensing presents an opportunity to detect and analyze lake trophic state with reproducible, robust methods across time and space. We used Landsat surface reflectance data to create the first compendium of annual lake trophic state for 55,662 lakes of at least 10 ha in area throughout the contiguous United States from 1984 through 2020. The dataset was constructed with FAIR data principles (Findable, Accessible, Interoperable, and Reproducible) in mind, where data are publicly available, relational keys from parent datasets are retained, and all data wrangling and modeling routines are scripted for future reuse. Together, this resource offers critical data to address basic and applied research questions about lake water quality at a suite of spatial and temporal scales.

Scientific Data↗
USGS2024

Evaluation of metrics and thresholds for use in national-scale river harmful algal bloom assessments

The spatiotemporal distribution of harmful algal blooms (HABs) in rivers remains poorly understood, and there is an urgent need to develop a consistent set of metrics to better document HAB occurrences and forecast future events. Using data from seven sites in the Illinois River Basin, we computed metrics focused on HAB conditions related to excess algal growth and hypoxia. Daily mean chlorophyll and dissolved oxygen (DO) concentrations, gross primary productivity (GPP), and net ecosystem productivity (NEP) rates, focused on water quality status, identifying the timing of the transition from a clear-water to an algal dominated state. Early warning indicators (EWIs), the first-order autoregressive process (Ar1) and standard deviation (SD) of chlorophyll concentrations, focused on future events, forecasting blooms. Metrics were compared to either literature-derived or statistical-based thresholds and were normalized by total number of daily samples for an exceedance rate. Exceedances of a daily mean chlorophyll concentration averaged 50 % across all sites using a 10 µg L −1 threshold but increasing the threshold to 50 μg L −1 reduced the average exceedance rate to 5 %. The average exceedance rate for GPP (∼8 g O 2 m 2 d −1 threshold) was 15 %, similar to the daily amplitude DO concentration (∼3 mg L −1 threshold), but the average for NEP (0 g O 2 m 2 d −1 threshold) was higher, at 28 %. The number of days with at least 1 continuous DO concentration below the threshold of 5, 3, or 2 mg L −1 , had basin wide exceedance rates of 9 %, 3 %, and 2 %, respectively. Thresholds for EWIs, Ar1 and SD, were exceeded at 5 of the 7 sites with high chlorophyll concentrations and GPP rates. The correlation between proxies for algal biomass (chlorophyll concentration) and productivity (GPP) was strongest for sites in the middle region of the basin, with R 2 values between 0.54 and 0.74. Although, cyanotoxin concentrations are the most commonly used metrics by states to define an inland water HAB, there is a paucity of publicly available data. The wider availability of chlorophyll and oxygen data combined with the results from this study suggest that biomass and productivity state and event-based metrics may be a promising way to assess and predict the vulnerability of rivers to some of the deleterious effects of HABs at broad spatial scales.

Illinois, Indiana, Wisconsin↗
USGS2023

Response of lake metabolism to catchment inputs inferred using high-frequency lake and stream data from across the northern hemisphere

In lakes, the rates of gross primary production (GPP), ecosystem respiration (R), and net ecosystem production (NEP) are often controlled by resource availability. Herein, we explore how catchment vs. within lake predictors of metabolism compare using data from 16 lakes spanning 39°N to 64°N, a range of inflowing streams, and trophic status. For each lake, we combined stream loads of dissolved organic carbon (DOC), total nitrogen (TN), and total phosphorus (TP) with lake DOC, TN, and TP concentrations and high frequency in situ monitoring of dissolved oxygen. We found that stream load stoichiometry indicated lake stoichiometry for C : N and C : P ( r 2 = 0.74 and r 2 = 0.84, respectively), but not for N : P ( r 2 = 0.04). As we found a strong positive correlation between TN and TP, we only used TP in our statistical models. For the catchment model, GPP and R were best predicted by DOC load, TP load, and load N : P ( R 2 = 0.85 and R 2 = 0.82, respectively). For the lake model, GPP and R were best predicted by TP concentrations ( R 2 = 0.86 and R 2 = 0.67, respectively). The inclusion of N : P in the catchment model, but not the lake model, suggests that both N and P regulate metabolism and that organisms may be responding more strongly to catchment inputs than lake resources. Our models predicted NEP poorly, though it is unclear why. Overall, our work stresses the importance of characterizing lake catchment loads to predict metabolic rates, a result that may be particularly important in catchments experiencing changing hydrologic regimes related to global environmental change.

Limnology & Oceanography↗
USGS2023

A synergistic future for AI and ecology

Research in both ecology and AI strives for predictive understanding of complex systems, where nonlinearities arise from multidimensional interactions and feedbacks across multiple scales. After a century of independent, asynchronous advances in computational and ecological research, we foresee a critical need for intentional synergy to meet current societal challenges against the backdrop of global change. These challenges include understanding the unpredictability of systems-level phenomena and resilience dynamics on a rapidly changing planet. Here, we spotlight both the promise and the urgency of a convergence research paradigm between ecology and AI. Ecological systems are a challenge to fully and holistically model, even using the most prominent AI technique today: deep neural networks. Moreover, ecological systems have emergent and resilient behaviors that may inspire new, robust AI architectures and methodologies. We share examples of how challenges in ecological systems modeling would benefit from advances in AI techniques that are themselves inspired by the systems they seek to model. Both fields have inspired each other, albeit indirectly, in an evolution toward this convergence. We emphasize the need for more purposeful synergy to accelerate the understanding of ecological resilience whilst building the resilience currently lacking in modern AI systems, which have been shown to fail at times because of poor generalization in different contexts. Persistent epistemic barriers would benefit from attention in both disciplines. The implications of a successful convergence go beyond advancing ecological disciplines or achieving an artificial general intelligence—they are critical for both persisting and thriving in an uncertain future.

Proceedings of the National Academy of Sciences↗
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↗
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↗