Search USGS⌕ Search

Geology topics

Phillip J. Goodling

Publications and source records attributed to Phillip J. Goodling.

16 recordsLinked to original sources
USGS2026

Predicting streamflow drought in the conterminous United States using machine learning and a donor-gage approach, 1982–2020

Drought is a highly consequential natural disaster that may likely increase in both severity and extent across the conterminous United States (CONUS). The mechanisms affecting the propagation of drought from the atmosphere to streamflow are complex and interactive, making the prediction of streamflow drought difficult with current modeling approaches. Machine learning is an emerging tool in the field of hydrology that may be well-suited to prediction of streamflow drought across large and topographically diverse areas. Here, we train and analyze 3198 random forest models at U.S. Geological Survey streamgages to understand common meteorological drivers of streamflow drought and to define physiographic characteristics of basins sensitive to these drivers. We also develop a novel dynamic regionalization approach using donor gages to predict daily streamflow drought at pseudo-ungaged locations. Our results show that teleconnections, temperature, evaporative demand, and snow–water equivalent are important drivers of streamflow drought in the West, Southwest, and Northern Rocky Mountains (Northern Rockies) regions of the United States, and precipitation and soil moisture are primary drivers of streamflow drought in the Northeast, Southeast, and the Northwest regions. Prediction using dynamic regionalization shows comparable performance to at-site models. This method may be applicable to other hydrologic prediction problems requiring transferability to ungaged locations.

conterminous United States↗
USGS2026

Long-term monotonic trends in water budget components in the contiguous United States: Insights from two hydrologic models

Characterizing changes to water availability for domestic, industrial, agricultural, and other uses is essential to support water management. To better quantify these changes, the U.S. Geological Survey and National Science Foundation National Center for Atmospheric Research produced two hydrologic models simulating water budget components from 1980 to 2021 over the contiguous United States (CONUS). Both hydrologic models were driven by a common atmospheric forcing dataset and aggregated to common spatial and temporal scales, which enables a novel evaluation of congruency between the models. We present annual and seasonal trends in six water budget components (precipitation, evapotranspiration, streamflow, groundwater recharge, soil saturation, and snow water equivalent) based on the Mann–Kendall test for monotonic trend and Theil-Sen slope estimate for the water year 1983–2021 period for ~86,000 catchments in CONUS. Additional components and metrics from our analysis pipeline are available in an associated published dataset, which contains more than 46 million trend results. The water budget trends showed broad agreement with prior observational and modeling studies that indicate increasing trends in the northeast and decreasing trends in southwestern CONUS. We found the seasonal variability in water budget trends was greatest in the southern, central, and northwest CONUS. These findings support integrated trend assessments when coupled with trends in water quality and use.

Contiguous United States↗
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↗
USGS2025

Technical note: A low-cost approach to monitoring relative streamflow dynamics in small headwater streams using time lapse imagery and a deep learning model

Despite their ubiquity and importance as freshwater habitat, small headwater streams are under-monitored by existing stream gage networks. To address this gap, we describe a low-cost, non-contact, and low-effort method that enables organizations to monitor relative streamflow dynamics in small headwater streams. The method uses a camera to capture repeat images of the stream from a fixed position. A person then annotates pairs of images, in each case indicating which image has more apparent streamflow or indicating equal flow if no difference is discernible. A deep learning modeling framework called streamflow rank estimation (SRE) is then trained on the annotated image pairs and applied to rank all images from highest to lowest apparent streamflow. From this result a relative hydrograph can be derived. We found that our modeled relative hydrograph dynamics matched the observed hydrograph dynamics well for 11 cameras at 8 streamflow sites in western Massachusetts. Higher performance was observed during the annotation period (median Kendall's Tau rank correlation of 0.75, with a range of 0.6–0.83) than after it (median Kendall's Tau of 0.59, with range 0.34–0.74). We found that annotation performance was generally consistent across the 11 camera sites and 2 individual annotators and was positively correlated with streamflow variability at a site. A scaling simulation determined that model performance improvements were limited after 1000 annotation pairs. Our model's estimates of relative flow, while not equivalent to absolute flow, may still be useful for many applications, such as ecological modeling and calculating event-based hydrological statistics (e.g., the number of out-of-bank floods). We anticipate that this method will be a valuable tool to extend existing stream monitoring networks and provide new insights on dynamic headwater systems.

Massachusetts↗
USGS2025

A low-cost approach to monitoring streamflow dynamics in small, headwater streams using timelapse imagery and a deep learning model

Despite their ubiquity and importance as freshwater habitat, small headwater streams are under monitored by existing stream gage networks. To address this gap, we describe a low-cost, non-contact, and low-effort method that enables organizations to monitor streamflow dynamics in small headwater streams. The method uses a camera to capture repeat images of the stream from a fixed position. A person then annotates pairs of images, in each case indicating which image has more apparent streamflow or indicating equal flow if no difference is discernible. A deep learning modelling framework called Streamflow Rank Estimation (SRE) is then trained on the annotated image pairs and applied to rank all images from highest to lowest apparent streamflow. From this result a relative hydrograph can be derived. We found that our modelled relative hydrograph dynamics matched the observed hydrograph dynamics well for 11 cameras at 8 streamflow sites in western Massachusetts. Higher performance was observed during the annotation period (median Kendall’s Tau rank correlation 0.75 with range 0.6–0.83) than after it (median Kendall’s Tau 0.59 with range 0.34 – 0.74). We found that annotation performance was generally consistent across the eleven camera sites and two individual annotators and was positively correlated with streamflow variability at a site. A scaling simulation determined that model performance improvements were limited after 1,000 annotation pairs. Our model’s estimates of relative flow, while not equivalent to absolute flow, may still be useful for many applications, such as ecological modelling and calculating event-based hydrological statistics (e.g., the number of out-of-bank floods). We anticipate this method will be a valuable tool to extend existing stream monitoring networks and provide new insights on dynamic headwater systems.

EGUSphere↗
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

A spatial machine learning model developed from noisy data requires multiscale performance evaluation: Predicting depth to bedrock in the Delaware River Basin, USA

Spatial machine learning models can be developed from observations with substantial unexplainable variability, sometimes called ‘noise’. Traditional point-scale metrics (e.g., R 2 ) alone can be misleading when evaluating these models. We present a multi-scale performance evaluation (MPE) using two additional scales (distributional and geostatistical). We apply the MPE framework to predictions of depth to bedrock (DTB) in the Delaware River Basin. Geostatistical analysis shows that approximately one third of the DTB variance is at spatial scale smaller than 2 km. Hence, we interpret our point-scale R 2 of 0.3 (testing data) to be sufficient for regional-scale modelling. Bias-correction methods improve performance at two of the three MPE scales: point-scale change is negligible, while distributional and geostatistical performance improves. In contrast, bias correction applied to a global DTB model does not improve MPE performance. This work encourages scale-appropriate performance evaluations to enable effective model intercomparison.

Delaware, New Jersey, New York, Pennsylvania↗
USGS2023

Integrated water resources trend assessments: State of the science, challenges, and opportunities for advancement

Water is vital to human life and healthy ecosystems. Here we outline the current state of national-scale water resources trend assessments, identify key gaps, and suggest advancements to better address critical issues related to changes in water resources that may threaten human development or the environment. Questions like, “Do we have less suitable drinking water now than we had 20 years ago?” or “Are flood events more common now than they were in the past?” prompted improvements in data, trend estimation methods, and modeling frameworks to track changes in, and better understand how land use and climate influence four water resources domains: surface and groundwater quantity and quality. However, continued advancement in trend assessments to better address issues related to changes in water availability is needed. Areas of need include more timely and efficient delivery of water resources trend results and improved capacity to estimate trends at unmonitored locations. Additional integration pieces include increased understanding of groundwater–surface water interactions, incorporation of both quantity and quality trends into water availability estimates, and the refinement of trend metrics to account for the competing needs of society and ecological integrity. Coupled with improved driver attribution studies, these components will better inform current and future water resources management.

Journal of the American Water Resource Association↗
USGS2023

Hydrogeologic characterization of Area B, Fort Detrick, Maryland

Groundwater in the karst groundwater system at Area B of Fort Detrick in Frederick County, Maryland, is contaminated with chlorinated solvents from the past disposal of laboratory wastes. In cooperation with U.S. Army Environmental Command and U.S. Army Garrison Fort Detrick, the U.S. Geological Survey performed a 3-year study to refine the conceptual model of groundwater flow in and around Area B of Fort Detrick at the site- to regional-scale. The investigation was designed to review the geologic setting, assess the temporal variability of the hydrologic system, evaluate the potential for interbasin groundwater flow, determine the degree of vertical connectivity of the aquifer, characterize the sources and timing of groundwater recharge, and identify if dyes from previous tracer tests continue to drain from the aquifer. This study established a continuous hydrologic monitoring network of 12 water level gages, 2 streamgages, a precipitation gage, and in situ fluorometric monitoring. A water budget analysis was performed using hydrologic monitoring data and a soil-water balance model constructed for the study. In this study each individual water budget term is calculated using available data or through modeling, and a water budget residual term is calculated. If the water budget residual term is small relative to the uncertainty of the underlying data, then an additional import or export of water (in other words, interbasin transfer) is not needed to fully describe the hydrologic system. Groundwater and spring samples from 20 locations were collected in a 2019 synoptic geochemical sampling event and analyzed for a suite of analytes that included groundwater age tracer constituents. The karst groundwater system was found to be highly responsive to hydrologic events, with strong water level and stream base flow responses to individual storm events and a historic wet period in 2017 and 2018. The water budget analysis included historic flooding in May 2018, though more typical hydrologic patterns were observed in 2019 and 2020. During most evaluated intervals, the water budget residual was less than the estimated uncertainty on the residual for the two Carroll Creek watersheds, which suggested no substantial net interbasin flow occurs from these watersheds. The watershed difference area, a region that includes Area B, had a significant negative water budget residual, which may be the result of a net interbasin import of groundwater or the result of focused groundwater recharge not simulated by the soil-water balance model. Geochemical analysis and groundwater age dating reveals shallow groundwater (approximately less than [<] 150 feet deep) appears to be relatively young (approximately <30 years) and to be recharged in the vicinity of Area B. In the deep groundwater sampled in this study (approximately greater than [>] 150 feet deep), older groundwater from a differing recharge source, based on stable isotopes and noble gas analyses, is observed and interpreted to represent less direct connectivity to the surface and increased proportions of water recharged to the north and (or) west of Area B. A clustering analysis to reveal groupings within the suite of geochemical data was used to define seven groups. The groupings generally show that wells in similar depths and lateral aquifer positions generally cluster together, with some exceptions. Although limited by suspended sediments, the in situ fluorometric monitoring at springs did not detect any dye leaving the system above the limit of detection for the method. Dye was only detected above the limit of detection in one well, which was used as an injection well during a previous dye tracer test. The results of this study support and refine the conceptual site model of groundwater hydrology at Area B. The geologic and geophysical log review in this study agrees with prior assessments of physical controls on groundwater flow. A literature review of mid-Atlantic karst studies identified similar controls reported in these environments. The additional characterization of hydrologic responsiveness in this study suggests that hydrologic conditions and events are important considerations when interpreting potentiometric surfaces and contaminant trends over time and highlights the importance of continuous hydrologic monitoring. There is evidence to suggest that either intense focused groundwater recharge occurs in the vicinity of Area B or net along-valley groundwater interbasin flow from the upper study watershed enters the lower watershed and discharges to Carroll Creek. Geochemical analyses also suggest that water recharged from Catoctin Mountain and the elevated areas to the north and (or) west of the site may be present in the older and deeper Area B groundwater.

Maryland↗
USGS2023

Thirty years of regional groundwater-quality trend studies in the United States: Major findings and lessons learned

Changes in groundwater quality have been evaluated for more than 2,200 wells in 25 Principal Aquifers in the United States based on repeated decadal sampling (once every 10 years) from 1988 to 2021. The purpose of this study is to identify contaminants with changing concentrations, the locations and magnitude of those changes, the factors driving those changes, the obstacles to interpreting the changes, and approaches to ameliorate those obstacles. Sampling was conducted in 89 networks of 20–30 wells each that represent various geographic regions, aquifer types and land use types. Each network, and the wells that comprise them, are sampled on a rotating basis once every 10 years. Of the 28 constituents evaluated for trends, concentrations of Na, Cl, dissolved solids, SO 4 , and NO 3 exhibited statistically significant increases at the network level more frequently than other constituents. Factors affecting trends in Cl and NO 3 are emphasized in this study. Regional patterns show large increases of Cl in urban areas in the Northeast and Midcontinent, where road-deicing salt application rates are 10 to 100 times greater than in other regions of the country, and in semiarid and arid regions of the western United States, where evaporation concentrates solutes in recharge. The largest increases in NO 3 were in agricultural areas of the semiarid west, arid west and Pacific regions which are characterized by oxic groundwater, long-term increases in nitrogen fertilizer usage, and high rates of irrigation. However, finding a direct relation between increasing contaminant sources and corresponding groundwater quality response, particularly when sampling once every 10 years, can be complicated by factors such as uncertainty in the timing, mass, and location of contaminant sources, groundwater residence time (recharge date), geochemical conditions in the aquifer that affect contaminant transport, and variability in water quality due to climatic factors such as seasonality and hydrologic conditions. Understanding groundwater residence time allows the changes in groundwater quality to be evaluated in the context of recharge date rather than the sample date. Likewise, information on geochemical characteristics of the aquifer can be helpful for understanding relations between contaminant source inputs and groundwater concentrations. For example, oxic geochemical conditions in the aquifer may allow for conservative transport and accumulation of NO 3 in groundwater, whereas reducing environments could favor NO 3 degradation. Differences in hydrologic conditions (wetter or drier than average) on the date of sampling could impact the statistical results of sampling at decadal intervals and obscure long-term patterns. Samples collected under substantially different hydrologic conditions can be identified, and high-frequency sampling can improve interpretation of measured results in these cases. Although decadal sampling and associated water-quality interpretations have limitations, repeated, scheduled sampling of thousands of wells over multiple decades has great value for identifying and understanding long-term, regional groundwater-quality trends. Despite these limitations, the concepts presented herein provide options that could be used to interpret trends or changes when sampling over longer timespans, which is less common than trend networks with higher frequency sampling intervals.

Conterminous United States↗
USGS2023

Regional streamflow drought forecasting in the Colorado River Basin using Deep Neural Network models

Process-based, large-scale (e.g., conterminous United States [CONUS]) hydrologic models have struggled to achieve reliable streamflow drought performance in arid regions and for low-flow periods. Deep learning has recently seen broad implementation in streamflow prediction and forecasting research projects throughout the world with performance often equaling or exceeding that of process-based models. Deep learning models are a possible approach to increase the accuracy of streamflow drought predictions and to expand the spatial coverage of river locations with available streamflow drought forecasts. As part of a multi-component Data-Driven Drought Prediction project, the U.S. Geological Survey is developing and testing deep learning models for streamflow drought forecasting. In this work, we present preliminary results of a deep learning model capable of predicting streamflow drought occurrence at ungaged locations for the Colorado River Basin (CRB). A long short-term memory (LSTM) neural network model was trained using 40 years (1980-2020) of daily streamflow data from 425 streamgages within and surrounding the CRB using static watershed attributes as well as meteorological and remotely sensed dynamic forcing inputs. Model tests were performed to evaluate model accuracy for now-casting streamflow drought conditions at ungaged locations and for forecasting drought conditions at lead times ranging from 0 to 14 days. Nearly all model configurations showed behavioral performance for predicting daily streamflow percentiles. Comparisons of LSTM model performance for predicting drought using fixed drought thresholds (calculated over all days and years) and variable drought thresholds (unique threshold calculated for each day of the year) identify differences in model skill between locations with implications for model design.

Colorado River basin↗
USGS2022

Bedrock depth influences spatial patterns of summer baseflow, temperature and flow disconnection for mountainous headwater streams

In mountain headwater streams, the quality and resilience of summer cold-water habitat is generally regulated by stream discharge, longitudinal stream channel connectivity and groundwater exchange. These critical hydrologic processes are thought to be influenced by the stream corridor bedrock contact depth (sediment thickness), a parameter often inferred from sparse hillslope borehole information, piezometer refusal and remotely sensed data. To investigate how local bedrock depth might control summer stream temperature and channel disconnection (dewatering) patterns, we measured stream corridor bedrock depth by collecting and interpreting 191 passive seismic datasets along eight headwater streams in Shenandoah National Park (Virginia, USA). In addition, we used multi-year stream temperature and streamflow records to calculate several baseflow-related metrics along and among the study streams. Finally, comprehensive visual surveys of stream channel dewatering were conducted in 2016, 2019 and 2021 during summer low flow conditions (124 total km of stream length). We found that measured bedrock depths along the study streams were not well-characterized by soils maps or an existing global-scale geologic dataset where the latter overpredicted measured depths by 12.2 m (mean) or approximately four times the average bedrock depth of 2.9 m. Half of the eight study stream corridors had an average bedrock depth of less than 2 m. Of the eight study streams, Staunton River had the deepest average bedrock depth (3.4 m), the coldest summer temperature profiles and substantially higher summer baseflow indices compared to the other study steams. Staunton River also exhibited paired air and water annual temperature signals suggesting deeper groundwater influence, and the stream channel did not dewater in lower sections during any baseflow survey. In contrast, Paine Run and Piney River did show pronounced, patchy channel dewatering, with Paine Run having dozens of discrete dry channel sections ranging from 1 to greater than 300 m in length. Stream dewatering patterns were apparently influenced by a combination of discrete deep bedrock (20+ m) features and more subtle sediment thickness variation (1–4 m) depending on local stream valley hydrogeology. In combination, these unique datasets show the first large-scale empirical support for existing conceptual models of headwater stream disconnection based on spatially variable underflow capacity and shallow groundwater supply.

Virginia↗
USGS2021

Simulated effects of sea-level rise on the shallow, fresh groundwater system of Assateague Island, Maryland and Virginia

The U.S. Geological Survey, in cooperation with the National Park Service, developed a three-dimensional groundwater-flow model for Assateague Island in eastern Maryland and Virginia to assess the effects of sea-level rise on the groundwater system. Sea-level rise is expected to increase the altitude of the water table in barrier island aquifer systems, possibly leading to adverse effects to ecosystems on the barrier islands. The potential effects of sea-level rise were evaluated by simulating groundwater conditions under sea-level-rise scenarios of 20 centimeters (cm), 40 cm, and 60 cm. Results show that as sea level rises, low-lying areas of the island originally represented as receiving freshwater recharge in the baseline scenario are inundated by saltwater. This change from freshwater recharge to saltwater decreases the overall amount of freshwater recharging the system. As the water table rises in response to the higher sea levels, freshwater flow out of the system changes, with more freshwater leaving as submarine groundwater discharge and less freshwater leaving as seeps and evapotranspiration. At the current land-surface altitude, as much as 50 percent of the island may be inundated with a 60-cm rise in sea level, and the low-lying marshes may change from freshwater to saltwater. Groundwater levels at 32 wells were monitored for as long as 12 months between October 2014 and September 2015 on Assateague Island. Results from objective classification analysis of 14 shallow monitoring wells show two dominant processes affecting groundwater levels in two different settings on the island. On the western side of the island, between the primary dune and the inland bays, water levels clearly respond to precipitation events. This side of the island is more protected from ocean tides and typically is more vegetated than the eastern side. On the eastern side of the island, between the Atlantic Ocean and the primary dune, water levels clearly respond to tidal events. Specific conductance was measured at four wells, two on the western part of the island and two on the eastern part of the island. Specific conductance values in the two wells west of the primary dune show episodic decreases, coinciding with precipitation events. Specific conductance values in the two wells on the eastern side of the primary dune show episodic increases, coinciding with high-tide events. These high frequency monitoring data are intended to aid in designing a monitoring network that can document both short-term and long-term hydrologic processes on Assateague Island National Seashore. This study uses a modeling approach consistent with models developed for Gateway National Recreation Area, Sandy Hook Unit (New Jersey) and Fire Island National Seashore (New York). Combined, these models are meant to improve the regional capabilities for predicting climate-change effects on barrier islands and provide resource managers with a common set of tools for adaptation and mitigation of potentially adverse effects of sea-level rise.

Maryland, Virginia↗
USGS2021

Assessment of contaminant trends in plumes and wells and monitoring network optimization at the Badger Army Ammunition Plant, Sauk County, Wisconsin

Soil and groundwater at the Badger Army Ammunition Plant (BAAP), Sauk County, Wisconsin, were affected by several contaminants as a result of production and waste disposal practices common during its operation from 1942 to 1975. Three distinct plumes of contaminated groundwater originate on BAAP property and extend off-site, as identified by previous studies. Routine sampling of groundwater quality from a network of monitoring wells and off-site private wells has been performed since 1990, although the number of wells monitored and the monitoring frequency have varied as the approved monitoring plan was modified. During the period of monitoring from 1990 to 2018, numerous site investigations and remedial actions were conducted to address the sources of contamination, contaminated soils, and groundwater. Concentrations of contaminants reportedly decreased between 2000 and 2012 within all three plumes. Five or six contaminants of concern (COCs) were identified for each of the three plumes. An independent assessment of the contaminant plumes and of the monitoring network was conducted using groundwater-quality data collected from more than 600 wells between 2000 and 2018. In a study conducted by the U.S. Geological Survey (USGS), in cooperation with the Army Environmental Command, a consistent data aggregation and interpolation scheme was applied to derive the likely maximum groundwater plume extents in four 3-year time periods between 2000 and 2018. The plume extent was defined by the Enforcement Standard for each COC and represents the maximum concentration observed in each 3-year time period. The plume boundary analysis shows that the spatial extent of groundwater contamination decreased for most COCs during the study period. Some plume boundaries are not well delineated by the existing monitoring network, particularly the downgradient edge of the Propellant Burning Ground plume. Maps identify the plume boundary in each time period, the sampling well network used to delineate the plume, and wells that were sampled in the 2010–12 period but not sampled in the 2015–18 period. A series of statistical analyses using the Monitoring and Remediation Optimization System, version 3.0, program were applied to the available COC concentration data for two distinct periods, 2000 to 2012 and 2013 to 2018, with the break between periods coinciding with changes to the monitoring network in 2013. Trends in the concentration of COCs in individual wells varied, although generally more wells had decreasing than had increasing concentrations for most COCs in both time periods. The exceptions were ethyl ether in the 2004–12 period and 2,6-dinitrotoluene in the 2013–18 period, for which more wells had an increasing trend. Spatial moment analysis of concentration data from the well network was used to assess the stability of each plume for the COCs. During the 2000–12 period, most of the contaminant plumes for which data were sufficient to complete the analysis were either decreasing or stable in mass and size. The exceptions were carbon tetrachloride (associated solely with the Propellant Burning Ground plume) and 2,4-dinitrotoluene and 2,6-dinitrotoluene (in the Deterrent Burning Ground plume), which showed an increasing trend in mass. No COCs showed an increasing trend in plume mass in the 2013–18 period. Some wells with increasing trends in concentration or with concentrations greater than the enforcement standard are near the tail of a plume, where increased monitoring may be of value to better define future plume boundaries. A spatial optimization analysis covering the 2013–18 period identified six wells that provided information redundant to that from other wells. A temporal optimization analysis identified optimal sampling frequencies for 125 wells. Remedial actions directed at the Propellant Burning Ground plume coincided with a general decrease in plume mass and size, although in specific areas and depths, the plume size for specific contaminants may still be increasing.

Wisconsin↗
USGS2020

Making ‘chemical cocktails’ – Evolution of urban geochemical processes across the periodic table of elements

Urbanization contributes to the formation of novel elemental combinations and signatures in terrestrial and aquatic watersheds, also known as ‘chemical cocktails.’ The composition of chemical cocktails evolves across space and time due to: (1) elevated concentrations from anthropogenic sources, (2) accelerated weathering and corrosion of the built environment, (3) increased drainage density and intensification of urban water conveyance systems, and (4) enhanced rates of geochemical transformations due to changes in temperature, ionic strength, pH, and redox potentials. Characterizing chemical cocktails and underlying geochemical processes is necessary for: (1) tracking pollution sources using complex chemical mixtures instead of individual elements or compounds; (2) developing new strategies for co-managing groups of contaminants; (3) identifying proxies for predicting transport of chemical mixtures using continuous sensor data; and (4) determining whether interactive effects of chemical cocktails produce ecosystem-scale impacts greater than the sum of individual chemical stressors. First, we discuss some unique urban geochemical processes which form chemical cocktails, such as urban soil formation, human-accelerated weathering, urban acidification-alkalinization, and Freshwater Salinization Syndrome. Second, we review and synthesize global patterns in concentrations of major ions, carbon and nutrients, and trace elements in urban streams across different world regions and make comparisons with reference conditions. In addition to our global analysis, we highlight examples from watersheds in the Baltimore-Washington DC area, USA, which show increased transport of major ions, trace metals, and nutrients across streams draining a well-defined land-use gradient. Urbanization increased the concentrations of multiple major and trace elements in streams draining human-dominated watersheds compared to reference conditions. Chemical cocktails of major and trace elements were formed over diurnal cycles coinciding with changes in streamflow, dissolved oxygen, pH, and other variables measured by high-frequency sensors. Some chemical cocktails of major and trace elements were also significantly related to specific conductance (p < 0.05), which can be measured by sensors. Concentrations of major and trace elements increased, peaked, or decreased longitudinally along streams as watershed urbanization increased, which is consistent with distinct shifts in chemical mixtures upstream and downstream of other major cities in the world. Our global analysis of urban streams shows that concentrations of multiple elements along the periodic table significantly increase when compared with reference conditions. Furthermore, similar biogeochemical patterns and processes can be grouped among distinct mixtures of elements of major ions, dissolved organic matter, nutrients, and trace elements as chemical cocktails. Chemical cocktails form in urban waters over diurnal cycles, decades, and throughout drainage basins. We conclude our global review and synthesis by proposing strategies for monitoring and managing chemical cocktails using source control, ecosystem restoration, and green infrastructure. We discuss future research directions applying the watershed chemical cocktail approach to diagnose and manage environmental problems. Ultimately, a chemical cocktail approach targeting sources, transport, and transformations of different and distinct elemental combinations is beneficial to more holistically monitor and manage the emerging impacts of chemical mixtures in the world's fresh waters.

Applied Geochemistry↗