The Proxies Project is a series of studies to improve our understanding of water contaminants and water quality hazards. These studies develop models and technical approaches for estimating concentrations and assessing risk associated with:
- harmful algal blooms (HABs)
- per- and polyfluoroalkyl substances (PFAS)
- 12 elements of concern (EoC)
Harmful Algal Blooms • Per- and Polyfluoroalkyl Substances • Elements of Concern
A ‘proxy’ (also known as a ‘surrogate’) is a measurement or model of one thing that helps us to understand some other thing that we’re interested in (like HABs, PFAS, or EoC). The proxy is typically simpler, cheaper, and (or) more rapidly measured than the direct measurement of the thing we’re interested in. In this project, the proxies we can readily measure or model are water constituents (like temperature, dissolved oxygen, turbidity, and salinity) and physical processes (like streamflow and runoff). The things we’re interested in are the water quality hazards and contaminants.
The three workgroups that make up the Project are taking a combination of approaches to develop proxy models for the three water contaminants (PFAS and EoC) and water quality hazards (HABs) noted above, although all workgroups are using:
- Geospatial data
- In-situ sensors (sensors that stay in the water)
In-situ sensors provide continuous, high-resolution data for a wide range of water quality metrics, which are necessary for proxy model development.
The specific studies are being conducted in hydrologic basins. A hydrologic basin is an area of land where all the precipitation drains into a specific river, like the Mississippi River Basin or the Columbia River Basin. The studies in the Proxies Project range from sub-basins (small drainage areas completely contained within larger basins) to multi-basins (independent large basins or groups of interconnected basins). This big range in spatial scale makes it easier to take the approaches and models developed by the Proxies Project and ultimately apply them to the whole country.
The three hydrologic basins for these studies include:
- Delaware River Basin
- Upper Colorado River Basin
- Illinois River Basin
Additional PFAS-focused studies are also being conducted in Cape Cod and the Potomac River Basin.
This webpage has links to all products resulting from the Proxies Project. The sections below provide more information on the goals, objectives, and approaches for each of the three workgroups, as addressed by this four-year effort (fiscal years 2021-2024).
Project Wide Product
Web Tool
Catalogue of Geospatial Datasets for the USGS Water Quality Program, Proxies Project
Harmful Algal Blooms (HABs)
Harmful algal blooms (HABs) happen when naturally occurring algae become very abundant and (or) releases toxins that can be harmful to people, pets, and wildlife. We generally understand the major drivers of HABs, like high water temperature, slow water, and sunlight. But we understand less about how these factors interact to start a HAB at a particular place and time, particularly when and where HABs occur in flowing water, like rivers and streams.
The main goals of the HABs part of the Proxies Project are to:
- Better understand the processes controlling HABs in rivers
- Develop better approaches to identify and forecast HABs at different spatial scales
Specific objectives of the HABs studies include:
- Assess existing metrics and thresholds used to define HABs
- Develop proxy approaches for identifying HABs in rivers, using continuous monitoring
(in-situ sensor) data at the basin or sub-basin scale - Map rivers vulnerable to HABs across the conterminous U.S. (lower 48 states)
- Use these new approaches to create a framework for forecasting HABs in river systems
The HABs workgroup is using existing data sources and a combination of geospatial analytics, machine learning, and modeling to meet these objectives.
Contact: Jennifer Murphy
HABs Products
Data Releases
- Compilation of state-level freshwater harmful algal bloom recreational and drinking water guidelines for the conterminous United States as of 2022
- Harmonized discrete and continuous water quality data in support of modeling harmful algal blooms in the Illinois River Basin, 2005 - 2020
- RiverMET: Workflow and scripts for river metabolism estimation including Illinois River Basin application, 2005 – 2020
- A national harmonized dataset of discrete chlorophyll from lakes and streams (2005-2022): U.S. Geological Survey data release
- Harmonized continuous water quality data in support of modeling harmful algal blooms in the United States, 2005 - 2022
- Daily estimates of chlorophyll concentration for 82 U.S. rivers: U.S. Geological Survey data release
- Modeled transport components of daily chlorophyll-a in the Illinois River, 2018 through 2020: U.S. Geological Survey data release
- Early warning indicators for harmful algal bloom assessments in the Illinois River: U.S. Geological Survey, Data Release
Papers
- River Metabolism Estimation Tools (RiverMET) with Demo in the Illinois River Basin
- The "H," "A," and "B" of a HAB: A definitional framework
- Predicting Daily River Chlorophyll Concentrations at a Continental Scale: Water Resources Research, v. 59, no. 11, p. e2022WR034215
- Spectral mixture analysis for surveillance of harmful algal blooms (SMASH): A field-, laboratory-, and satellite-based approach to identifying cyanobacteria genera from remotely sensed data: Remote Sensing of Environment, v. 279, p. 113089
- River control points for algal productivity revealed by transport analysis. Geophysical Research Letters
- Metabolism Regimes in Regulated Rivers of the Illinois River Basin, USA
- Chlorophyll a in lakes and streams of the United States (2005–2022)
- Evaluation of metrics and thresholds for use in national-scale river harmful algal bloom assessments. Ecological Indicators
- Software Application for Spectral Mixture Analysis for Surveillance of Harmful Algal Blooms (SMASH): A tool for identifying cyanobacteria genera from remotely sensed data
Software or code
- Source Code: Forecasting river chlorophyll-a using random forest algorithms (v0.1.0)
- Source code: A national harmonized dataset of discrete chlorophyll from lakes and streams (2005-2022) (v1.0).
- SAS - Software Application for Spectral Mixture Analysis for Surveillance of Harmful Algal Blooms (ver. 1.0.7, September, 2023): U.S. Geological software release
Per- and Polyfluoroalkyl Substances (PFAS)
Per- and polyfluoroalkyl substances (PFAS) are a diverse group of chemicals that:
- Last a long time in the environment (hence their nickname “forever chemicals”)
- Occur in many waterbodies (and in both surface water and groundwater)
- Are potentially harmful to human health and aquatic ecosystems
The current approach for measuring PFAS concentrations in natural waters, like rivers and lakes, is liquid chromatography-tandem mass spectrometry—an analytical method that is very good at detecting PFAS in water. However, this method is expensive and often involves long wait times for results. New proxy models and sensor-based technologies provide faster estimates of PFAS concentrations in water at a finer spatial scale. This information helps identify where we need to collect discrete water samples (samples of water collected at one point in time and taken back to a laboratory for analysis) and when we need to use traditional analytical approaches to quantify PFAS concentrations.
The main goals of the PFAS component of the Proxies Project are to:
- Model PFAS concentrations in surface water and groundwater
- Develop sensor-based approaches that provide proxy measurements for estimating PFAS concentrations in natural waters
Specific objectives of the PFAS studies include:
- Develop a data exploration and visualization tool that shows the locations of potential PFAS sources in each basin, which will help inform PFAS modeling efforts
- Develop surface-water models for PFAS concentrations at the basin spatial scale
- Conduct field studies using various sensor technologies to develop proxy approaches for estimating PFAS concentrations in natural waters
- Conduct field studies to better estimate PFAS concentrations in contaminated groundwater sites
To meet these objectives, the PFAS workgroup uses a combination of:
- Existing and new data collected by both discrete sampling and in-situ sensors
- Geospatial analytics focused on potential PFAS sources
- Hydrodynamic modeling to predict concentrations
Contact: Larry Barber, Doug Kent
PFAS Products
Data Releases
- Potomac River Watershed Accumulated Wastewater Ratios and Predicted Environmental Concentrations
- Uptake of per- and polyfluoroalkyl substances by fish, mussel, and passive samplers in mobile laboratory exposures using groundwater from a contamination plume at a historical fire training area, Cape Cod, Massachusetts - Chemical and biological data from August to September 2018
- Upper Colorado River Basin accumulated wastewater ratios
- Laboratory fluorescence and total dissolved nitrogen measurements for surface water samples collected from the Rio Grande during a 24-hr time period near Albuquerque, New Mexico
- Optical measurements for surface water samples collected within the Neshaminy Creek basin during November 2021: U.S. Geological Survey data release
- Laboratory Optical Measurements from Discrete Surface Water Samples Collected During Water Quality Mapping Campaigns on the Illinois Waterway and Chicago Area Waterway Systems: U.S. Geological Survey data release
- Longitudinal assessment of per- and polyfluoroalkyl substances (PFAS) and co-occurring inorganic and organic contaminants in Neshaminy Creek, Pennsylvania, November, 2021
- Laboratory Optical Measurements Collected for Surface Water Samples Collected within Fraser River in the Upper Colorado River Basin from September 2021 through August 2023
- Environmental Sampling and Modeling Results to Characterize Surface-Water Quality at 32 Sites Across the Potomac River Watershed, 2022 (ver. 2.0, published July 30, 2024). U.S. Geological Survey data release, 2024
Papers
- Uptake of per- and polyfluoroalkyl substances by fish, mussel, and passive samplers in mobile laboratory exposures using groundwater from a contamination plume at a historical fire training area, Cape Cod, Massachusetts
- Wastewater reuse and predicted ecological risk posed by contaminant mixtures in Potomac River Watershed streams
- Temporal variability and sources of PFAS in the Rio Grande, New Mexico through an arid urban area using multiple tracers and high-frequency sampling. Emerging Contaminants, v.10, 100314
- Bioconcentration of per- and polyfluoroalkyl substances and precursors in fathead minnow tissues environmentally exposed to aqueous film-forming foam–contaminated waters: Environmental Toxicology and Chemistry
- Using a Time-of-Travel Sampling Approach to Quantify Per- and Polyfluoroalkyl Substances (PFAS) Stream Loading and Source Inputs in a Mixed-Source, Urban Catchment
- Factors contributing to pesticide contamination in riverine systems: The role of wastewater and landscape sources
Web tool
Elements of Concern (EoC)
Elements of concern (EoC) are a group of elements that are monitored at the local, regional, or national level because of their known toxicity in aquatic systems. Traditional, discrete water sampling approaches for monitoring EoC concentrations require a lot of time and money, and often involve prolonged wait times for obtaining results. Continuous proxy measurement of EoCs in rivers allows for more rapid and (near) real-time estimates of EoC concentrations, which may reveal unhealthy levels of EoC sooner. Proxy approaches for estimating EoC concentrations across large spatial scales (sub-basins and basins) can help identify sources of EoC and show how far EoC travel.
The main goals of the EoC part of the Proxies Project are to:
- Explore modeling approaches that provide high-frequency (rapid and near real-time), site-specific estimates of EoC concentrations in streams
- Explore modeling approaches that provide sub-basin- to basin-scale spatial estimates of EoC concentrations in streams based upon specific landscape and geographic factors
Specific objectives of the EoC studies include:
- Develop a data exploration and visualization tool for mapping concentrations of the 12 EoC in the three study basins (Delaware River Basin, Illinois River Basin and Upper Colorado River Basin) using existing data from the Water Quality Portal
- Develop spatially explicit models (models where the specific locations of things matter) using a combination of machine learning and a long list of attributes for each basin (for example, geology and soil characteristics, mining areas, population centers, and infrastructure). These models show us which attributes affect how EoC vary across the landscape at different spatial scales, from small drainage areas to large basins.
- Develop models that provide high-frequency estimates of EoC concentrations at USGS continuous water monitoring sites using commonly deployed in-situ sensors (pH, specific conductance, dissolved oxygen, turbidity, temperature, and streamflow)
- Develop models that estimate EoC concentrations through time and across the landscape using new sensor technology and newly collected field data
To achieve these objectives, the EoC workgroup is using:
- Existing data from the Water Quality Portal
- New data collected in the field (including EoC concentration data and paired sensor data)
- Geospatial approaches with machine learning and other forms of statistical modeling
Contact: Mark Marvin-DiPasquale, Blaine McCleskey
EoC Products
Data Release
Software or code
Web Tool
The Proxies Project is a series of studies to improve our understanding of water contaminants and water quality hazards. These studies develop models and technical approaches for estimating concentrations and assessing risk associated with:
- harmful algal blooms (HABs)
- per- and polyfluoroalkyl substances (PFAS)
- 12 elements of concern (EoC)
Harmful Algal Blooms • Per- and Polyfluoroalkyl Substances • Elements of Concern
A ‘proxy’ (also known as a ‘surrogate’) is a measurement or model of one thing that helps us to understand some other thing that we’re interested in (like HABs, PFAS, or EoC). The proxy is typically simpler, cheaper, and (or) more rapidly measured than the direct measurement of the thing we’re interested in. In this project, the proxies we can readily measure or model are water constituents (like temperature, dissolved oxygen, turbidity, and salinity) and physical processes (like streamflow and runoff). The things we’re interested in are the water quality hazards and contaminants.
The three workgroups that make up the Project are taking a combination of approaches to develop proxy models for the three water contaminants (PFAS and EoC) and water quality hazards (HABs) noted above, although all workgroups are using:
- Geospatial data
- In-situ sensors (sensors that stay in the water)
In-situ sensors provide continuous, high-resolution data for a wide range of water quality metrics, which are necessary for proxy model development.
The specific studies are being conducted in hydrologic basins. A hydrologic basin is an area of land where all the precipitation drains into a specific river, like the Mississippi River Basin or the Columbia River Basin. The studies in the Proxies Project range from sub-basins (small drainage areas completely contained within larger basins) to multi-basins (independent large basins or groups of interconnected basins). This big range in spatial scale makes it easier to take the approaches and models developed by the Proxies Project and ultimately apply them to the whole country.
The three hydrologic basins for these studies include:
- Delaware River Basin
- Upper Colorado River Basin
- Illinois River Basin
Additional PFAS-focused studies are also being conducted in Cape Cod and the Potomac River Basin.
This webpage has links to all products resulting from the Proxies Project. The sections below provide more information on the goals, objectives, and approaches for each of the three workgroups, as addressed by this four-year effort (fiscal years 2021-2024).
Project Wide Product
Web Tool
Catalogue of Geospatial Datasets for the USGS Water Quality Program, Proxies Project
Harmful Algal Blooms (HABs)
Harmful algal blooms (HABs) happen when naturally occurring algae become very abundant and (or) releases toxins that can be harmful to people, pets, and wildlife. We generally understand the major drivers of HABs, like high water temperature, slow water, and sunlight. But we understand less about how these factors interact to start a HAB at a particular place and time, particularly when and where HABs occur in flowing water, like rivers and streams.
The main goals of the HABs part of the Proxies Project are to:
- Better understand the processes controlling HABs in rivers
- Develop better approaches to identify and forecast HABs at different spatial scales
Specific objectives of the HABs studies include:
- Assess existing metrics and thresholds used to define HABs
- Develop proxy approaches for identifying HABs in rivers, using continuous monitoring
(in-situ sensor) data at the basin or sub-basin scale - Map rivers vulnerable to HABs across the conterminous U.S. (lower 48 states)
- Use these new approaches to create a framework for forecasting HABs in river systems
The HABs workgroup is using existing data sources and a combination of geospatial analytics, machine learning, and modeling to meet these objectives.
Contact: Jennifer Murphy
HABs Products
Data Releases
- Compilation of state-level freshwater harmful algal bloom recreational and drinking water guidelines for the conterminous United States as of 2022
- Harmonized discrete and continuous water quality data in support of modeling harmful algal blooms in the Illinois River Basin, 2005 - 2020
- RiverMET: Workflow and scripts for river metabolism estimation including Illinois River Basin application, 2005 – 2020
- A national harmonized dataset of discrete chlorophyll from lakes and streams (2005-2022): U.S. Geological Survey data release
- Harmonized continuous water quality data in support of modeling harmful algal blooms in the United States, 2005 - 2022
- Daily estimates of chlorophyll concentration for 82 U.S. rivers: U.S. Geological Survey data release
- Modeled transport components of daily chlorophyll-a in the Illinois River, 2018 through 2020: U.S. Geological Survey data release
- Early warning indicators for harmful algal bloom assessments in the Illinois River: U.S. Geological Survey, Data Release
Papers
- River Metabolism Estimation Tools (RiverMET) with Demo in the Illinois River Basin
- The "H," "A," and "B" of a HAB: A definitional framework
- Predicting Daily River Chlorophyll Concentrations at a Continental Scale: Water Resources Research, v. 59, no. 11, p. e2022WR034215
- Spectral mixture analysis for surveillance of harmful algal blooms (SMASH): A field-, laboratory-, and satellite-based approach to identifying cyanobacteria genera from remotely sensed data: Remote Sensing of Environment, v. 279, p. 113089
- River control points for algal productivity revealed by transport analysis. Geophysical Research Letters
- Metabolism Regimes in Regulated Rivers of the Illinois River Basin, USA
- Chlorophyll a in lakes and streams of the United States (2005–2022)
- Evaluation of metrics and thresholds for use in national-scale river harmful algal bloom assessments. Ecological Indicators
- Software Application for Spectral Mixture Analysis for Surveillance of Harmful Algal Blooms (SMASH): A tool for identifying cyanobacteria genera from remotely sensed data
Software or code
- Source Code: Forecasting river chlorophyll-a using random forest algorithms (v0.1.0)
- Source code: A national harmonized dataset of discrete chlorophyll from lakes and streams (2005-2022) (v1.0).
- SAS - Software Application for Spectral Mixture Analysis for Surveillance of Harmful Algal Blooms (ver. 1.0.7, September, 2023): U.S. Geological software release
Per- and Polyfluoroalkyl Substances (PFAS)
Per- and polyfluoroalkyl substances (PFAS) are a diverse group of chemicals that:
- Last a long time in the environment (hence their nickname “forever chemicals”)
- Occur in many waterbodies (and in both surface water and groundwater)
- Are potentially harmful to human health and aquatic ecosystems
The current approach for measuring PFAS concentrations in natural waters, like rivers and lakes, is liquid chromatography-tandem mass spectrometry—an analytical method that is very good at detecting PFAS in water. However, this method is expensive and often involves long wait times for results. New proxy models and sensor-based technologies provide faster estimates of PFAS concentrations in water at a finer spatial scale. This information helps identify where we need to collect discrete water samples (samples of water collected at one point in time and taken back to a laboratory for analysis) and when we need to use traditional analytical approaches to quantify PFAS concentrations.
The main goals of the PFAS component of the Proxies Project are to:
- Model PFAS concentrations in surface water and groundwater
- Develop sensor-based approaches that provide proxy measurements for estimating PFAS concentrations in natural waters
Specific objectives of the PFAS studies include:
- Develop a data exploration and visualization tool that shows the locations of potential PFAS sources in each basin, which will help inform PFAS modeling efforts
- Develop surface-water models for PFAS concentrations at the basin spatial scale
- Conduct field studies using various sensor technologies to develop proxy approaches for estimating PFAS concentrations in natural waters
- Conduct field studies to better estimate PFAS concentrations in contaminated groundwater sites
To meet these objectives, the PFAS workgroup uses a combination of:
- Existing and new data collected by both discrete sampling and in-situ sensors
- Geospatial analytics focused on potential PFAS sources
- Hydrodynamic modeling to predict concentrations
Contact: Larry Barber, Doug Kent
PFAS Products
Data Releases
- Potomac River Watershed Accumulated Wastewater Ratios and Predicted Environmental Concentrations
- Uptake of per- and polyfluoroalkyl substances by fish, mussel, and passive samplers in mobile laboratory exposures using groundwater from a contamination plume at a historical fire training area, Cape Cod, Massachusetts - Chemical and biological data from August to September 2018
- Upper Colorado River Basin accumulated wastewater ratios
- Laboratory fluorescence and total dissolved nitrogen measurements for surface water samples collected from the Rio Grande during a 24-hr time period near Albuquerque, New Mexico
- Optical measurements for surface water samples collected within the Neshaminy Creek basin during November 2021: U.S. Geological Survey data release
- Laboratory Optical Measurements from Discrete Surface Water Samples Collected During Water Quality Mapping Campaigns on the Illinois Waterway and Chicago Area Waterway Systems: U.S. Geological Survey data release
- Longitudinal assessment of per- and polyfluoroalkyl substances (PFAS) and co-occurring inorganic and organic contaminants in Neshaminy Creek, Pennsylvania, November, 2021
- Laboratory Optical Measurements Collected for Surface Water Samples Collected within Fraser River in the Upper Colorado River Basin from September 2021 through August 2023
- Environmental Sampling and Modeling Results to Characterize Surface-Water Quality at 32 Sites Across the Potomac River Watershed, 2022 (ver. 2.0, published July 30, 2024). U.S. Geological Survey data release, 2024
Papers
- Uptake of per- and polyfluoroalkyl substances by fish, mussel, and passive samplers in mobile laboratory exposures using groundwater from a contamination plume at a historical fire training area, Cape Cod, Massachusetts
- Wastewater reuse and predicted ecological risk posed by contaminant mixtures in Potomac River Watershed streams
- Temporal variability and sources of PFAS in the Rio Grande, New Mexico through an arid urban area using multiple tracers and high-frequency sampling. Emerging Contaminants, v.10, 100314
- Bioconcentration of per- and polyfluoroalkyl substances and precursors in fathead minnow tissues environmentally exposed to aqueous film-forming foam–contaminated waters: Environmental Toxicology and Chemistry
- Using a Time-of-Travel Sampling Approach to Quantify Per- and Polyfluoroalkyl Substances (PFAS) Stream Loading and Source Inputs in a Mixed-Source, Urban Catchment
- Factors contributing to pesticide contamination in riverine systems: The role of wastewater and landscape sources
Web tool
Elements of Concern (EoC)
Elements of concern (EoC) are a group of elements that are monitored at the local, regional, or national level because of their known toxicity in aquatic systems. Traditional, discrete water sampling approaches for monitoring EoC concentrations require a lot of time and money, and often involve prolonged wait times for obtaining results. Continuous proxy measurement of EoCs in rivers allows for more rapid and (near) real-time estimates of EoC concentrations, which may reveal unhealthy levels of EoC sooner. Proxy approaches for estimating EoC concentrations across large spatial scales (sub-basins and basins) can help identify sources of EoC and show how far EoC travel.
The main goals of the EoC part of the Proxies Project are to:
- Explore modeling approaches that provide high-frequency (rapid and near real-time), site-specific estimates of EoC concentrations in streams
- Explore modeling approaches that provide sub-basin- to basin-scale spatial estimates of EoC concentrations in streams based upon specific landscape and geographic factors
Specific objectives of the EoC studies include:
- Develop a data exploration and visualization tool for mapping concentrations of the 12 EoC in the three study basins (Delaware River Basin, Illinois River Basin and Upper Colorado River Basin) using existing data from the Water Quality Portal
- Develop spatially explicit models (models where the specific locations of things matter) using a combination of machine learning and a long list of attributes for each basin (for example, geology and soil characteristics, mining areas, population centers, and infrastructure). These models show us which attributes affect how EoC vary across the landscape at different spatial scales, from small drainage areas to large basins.
- Develop models that provide high-frequency estimates of EoC concentrations at USGS continuous water monitoring sites using commonly deployed in-situ sensors (pH, specific conductance, dissolved oxygen, turbidity, temperature, and streamflow)
- Develop models that estimate EoC concentrations through time and across the landscape using new sensor technology and newly collected field data
To achieve these objectives, the EoC workgroup is using:
- Existing data from the Water Quality Portal
- New data collected in the field (including EoC concentration data and paired sensor data)
- Geospatial approaches with machine learning and other forms of statistical modeling
Contact: Mark Marvin-DiPasquale, Blaine McCleskey
EoC Products
Data Release
Software or code
Web Tool