Model-based assessment and mapping of total phosphorus enrichment in rivers with sparse reference data

Peter C. Esselman; R Jan Stevenson

doi:10.1016/j.scitotenv.2023.163418

Home/Archive/USGS

USGS · 70243190

Model-based assessment and mapping of total phosphorus enrichment in rivers with sparse reference data

Peter C. Esselman·R Jan Stevenson

View original source ↗Export metadata ↓

Abstract

Water nutrient management efforts are frequently coordinated across thousands of water bodies, leading to a need for spatially extensive information to facilitate decision making. Here we explore potential applications of a machine learning model of river low-flow total phosphorus (TP) concentrations to support landscape nutrient management. The model was trained, validated, and then applied for all rivers of Michigan, USA to identify potential drivers of nutrient variation, predict alteration in nutrient concentrations from minimally disturbed conditions, and explore reach specific sensitivity to riparian agricultural change. A boosted regression tree model of low-flow TP concentrations trained on natural and anthropogenic landscape predictors accounted for 53 % of variation in cross-validation data, had good accuracy, little bias, and plausible relationships between predictors and response. Percent riparian agricultural cover accounted for the greatest root mean square error reduction in the modeled response (33.2 %), followed by riparian soil permeability (12.9 %), watershed slope (9.6 %), and percent urban cover (9.6 %). An apparent non-linear relationship between TP concentrations and percent riparian agricultural cover suggested steep positive increases in stream TP concentrations between 10 and 30 % upstream riparian agricultural cover. Predicted minimally disturbed TP concentrations were spatially variable and ranged from 7.0 to 48.5 μg l −1 , with the highest concentrations in watersheds draining low-permeability lake plain soils. Comparison of minimally disturbed predictions to those from the early 2000s suggested that much of northern Michigan existed close to the reference condition, while lower Michigan streams were often substantially enriched. Our predicted values of minimally disturbed condition generally agree with previous studies but offer greater geographic specificity. Expanded application of machine learning modeling with landscape predictor data have great potential to inform large scale strategy development in landscapes with sparse reference data.

Explore related subjects

Science of the Total Environment Article

Source-reported bounding extent: 41.696089° to 48.190975° latitude; -90.41562° to -82.412965° longitude. This indicates report coverage, not an exact sampling location. View area on OpenStreetMap.

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Peter C. Esselman, R Jan Stevenson. 2023. Model-based assessment and mapping of total phosphorus enrichment in rivers with sparse reference data. https://doi.org/10.1016/j.scitotenv.2023.163418

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

Mercury biomagnification across food webs with varying non-native fish presence: Implications for native fish conservation in the upper Colorado River Basin

Environmental contaminants and non-native species introductions negatively affect aquatic ecosystem conservation. Mercury (Hg) accumulates within food webs where it can biomagnify to toxic concentrations, which can be affected by altered trophic relationships from non-native species. This study examined Hg concentrations (n samples = 655) and stable carbon and nitrogen isotope ( n = 746) compositions in native ( n = 313) and non-native ( n = 342) species among four river subbasins (Gunnison, Lower Green, Colorado-Dolores, and White-Yampa), in the Upper Colorado River Basin (UCRB) during 2016 and 2017 to inform the potential combined effects of Hg exposure and non-native species interactions on native fishes of conservation concern. We assessed Hg biomagnification at the fish assemblage level, compared concentrations to risk thresholds for fish and human health, and evaluated niche overlap among trophic levels across subbasins. The generalist-invertivores feeding guild had the lowest Hg concentrations (0.001–0.191 μg g −1 wet weight), and no individuals exceeded the U.S. Environmental Protection Agency aquatic life criterion (0.225 μg g −1 ). Piscivores had the highest concentrations (0.008–1.840 μg g −1 ), and 74% of individuals exceeded the criterion. Biomagnification across trophic levels was observed throughout the UCRB, yet varied by subbasin, with higher magnification factors in the Lower Green and White-Yampa subbasins which also had the highest proportion of non-native species. Stable isotope compositions revealed niche overlap among native and non-native species. Additionally, trophic position varied within species among the subbasins. The proliferation of non-native species has likely reshaped food webs and may have intensified the potential threat Hg poses to native fish recovery and conservation. This study documented increased Hg concentrations relative to past assessments, suggesting that continued monitoring could assist in evaluating trends in Hg accumulation. Particularly studies that focus on sources entering the food web, vectors for accumulation, and competitive interactions among native and non-native species.

Colorado, Utah, Wyoming↗

USGS2026

Evaluating groundwater quality influences from oil field operations and other anthropogenic activities in an urban setting, Santa Fe Springs, California

Groundwater quality is often affected by anthropogenic activities in urban settings. This study examines groundwater quality in and around the Santa Fe Springs Oil Field in Los Angeles County, California, where oil and gas production commonly intersects with high density industrial, commercial and residential land uses. Utilizing a combination of new and historical data, we evaluated potential pathways that would allow for oil field formation fluids to migrate into groundwater and whether mixing may have occurred based on the distribution of groundwater and oil field formation fluid tracers in samples. Samples were analyzed for a wide array of constituents including volatile organic compounds, light hydrocarbons, major ions, and various isotopic compositions. Despite evidence of oil field infrastructure providing potential pathways of migration via uncemented annular spaces, casing breaches and historical disposal of oil field formation water in surface ponds, the distribution and occurrence of stable isotopes of water, chloride, boron, and total dissolved solids do not indicate mixing of oil field formation water and groundwater. However, methane isotopic signatures and the presence of heavier alkanes suggest gas from oil-bearing formations have migrated from depth via oil field well infrastructure. Volatile organic compound detections were mainly from manufactured compounds unrelated to oil and gas production, with a relatively limited number of petroleum hydrocarbons also detected. Volatile organic compounds were generally found in wells tapping shallow, modern aged groundwater, indicating anthropogenic activities occurring at or near land surface as the source. Study results suggest that while oil field infrastructure provides migration pathways for oil field formation fluids to be introduced into groundwater, urban land uses not related to oil and gas production are the primary drivers of groundwater quality degradation.

California↗

USGS2026

Hydrologic variability drives environmental and geospatial relationships in Smallmouth Bass (Micropterus dolomieu) distribution

Hydrologic variation is a primary driver of stream ecosystems. Changing hydrology can lead to assemblage shifts and alterations in suitable habitat for freshwater species. As climate change is predicted to alter flow patterns in addition to increasing water temperatures, insight into relationships between species occupancy, hydrology, and temperature is critical for understanding current and future distributions. We examined how hydrologic variability, temperature, and other environmental variables interact to influence Micropterus dolomieu (Smallmouth Bass) occurrence. We used Spatial Stream Network models, allowing for the incorporation of spatial autocorrelation along streams' unique dendritic network, to examine Smallmouth Bass occupancy across a range of hydrologic variation in the Ozark-Ouachita Interior Highlands, USA. Hydrologic variation was the main driver of Smallmouth Bass occurrence, with occurrence more likely in groundwater streams with low hydrologic variation and high flow permanence. For groundwater streams, occurrence was positively associated with summer stream temperature and negatively associated with annual stream temperature. As variation increased, more variables showed significant relationships with occurrence. Distance metrics were important for all models, however as hydrologic disturbance increased, flow connected distance played a lesser role and stream distance played a greater role. Hydrologic variability was the overarching determinant of Smallmouth Bass occurrence and strongly influenced the predictive importance of environmental variables and geospatial relationships. Greater hydrologic variability resulted in stronger statistical relationships between occurrence and environmental variables and an increased importance of system connectivity. As climate change alters hydrologic processes and streams become more variable, understanding and accounting for these shifting relationships is essential.

Arkansas, Kansas, Missouri, Oklahoma↗