Informing Freshwater Management Strategies in the Chesapeake Bay Watershed by Using Observational Data and Expert Knowledge to Identify Influential Stressors
Human activities in the Chesapeake Bay watershed can negatively affect the abundance and diversity of macroinvertebrate communities in freshwater streams, which is a core measure of stream health. For example, urban development and agricultural intensification can degrade habitat and water-quality conditions in streams through sedimentation, nutrient runoff, and changes to instream habitat. A major goal of the Chesapeake Bay Program restoration effort is to improve stream health conditions in 10% of streams throughout the watershed. Given limited resources, identifying what stressors are most influential and where management is most likely to result in a positive change in stream health can help resource managers better plan and direct restoration efforts.

Study
The USGS team used expert knowledge in combination with biomonitoring data collected at 2,243 sites across the Chesapeake Bay Watershed to model how landscape-scale drivers, instream habitat stressors, and water quality stressors related to benthic macroinvertebrate communities in the watershed’s different regions (figure 1). The team used these models to predict where changes in stream-health measures across the watershed were likely occurring under different stressor conditions. They then used a resist-accept-direct (RAD) framework to identify regional management options based on predicted changes in benthic macroinvertebrate communities under differing stressor conditions.
Major Findings
- Water quality and habitat related to stream health: Various water-quality and habitat stressors appeared to influence macroinvertebrate populations. The most prevalent stressors included temperature, specific conductivity, physical habitat alterations, and habitat heterogeneity.
- Regional variation highlights complexity in stressor relationships: Prevalent stressors linked to macroinvertebrate populations varied by region within the watershed. For example, in the Northern Appalachian region, water temperature and specific conductivity were the primary stressors, while in the Southern Appalachian region, physical habitat stressors, like embeddedness, were most influential (figure 2). In mixed-land use areas like the Piedmont region, both habitat and specific conductivity were related to measures of stream health, whereas in the Coastal Plains region, multiple measures of habitat drove patterns in stream health metrics.
- Widespread observational data and modeling reveal new insights: The team, using a novel modeling approach, paired an extensive observational biomonitoring dataset with key physical habitat and water quality stressor data to identify which stressors most influenced benthic macroinvertebrates. In using their findings to then predict anticipated changes in macroinvertebrates, the team was able to highlight potential management actions that consider the complexity of stressors facing aquatic habitats in the watershed.

Management Applications
The findings from this study suggest that targeted management strategies that consider the unique, interconnected, and regionally specific stressors within the Chesapeake Bay watershed may be warranted. The use of extensive biomonitoring data collected throughout the watershed in combination with a modeling approach that enabled the identification of regional differences in influential stressors generated management-relevant information. Specifically, the team’s approach could be used by managers in a resist-accept-direct (RAD) framework to make more informed decisions by utilizing measured field data rather than findings from small-scale experimental studies, which may not capture the complexity of natural ecosystems across large landscapes. For example, some streams in the Southern Appalachian region have high stream-health scores because of fewer water-quality and habitat stressors. As a result, managers could choose to protect these healthy streams by trying to prevent (resisting) the introduction of new stressors. In contrast, managers may need to consider simply accepting the degraded condition of streams facing multiple stressors, as it may be determined that available resources could be better used to improve the conditions of streams facing fewer stressors elsewhere.
For more information
Stressor-driven changes in freshwater biological indicators inform spatial management strategies using expert knowledge, observational data, and hierarchical models Stressor-driven changes in freshwater biological indicators inform spatial management strategies using expert knowledge, observational data, and hierarchical models
Human activities in the Chesapeake Bay watershed can negatively affect the abundance and diversity of macroinvertebrate communities in freshwater streams, which is a core measure of stream health. For example, urban development and agricultural intensification can degrade habitat and water-quality conditions in streams through sedimentation, nutrient runoff, and changes to instream habitat. A major goal of the Chesapeake Bay Program restoration effort is to improve stream health conditions in 10% of streams throughout the watershed. Given limited resources, identifying what stressors are most influential and where management is most likely to result in a positive change in stream health can help resource managers better plan and direct restoration efforts.

Study
The USGS team used expert knowledge in combination with biomonitoring data collected at 2,243 sites across the Chesapeake Bay Watershed to model how landscape-scale drivers, instream habitat stressors, and water quality stressors related to benthic macroinvertebrate communities in the watershed’s different regions (figure 1). The team used these models to predict where changes in stream-health measures across the watershed were likely occurring under different stressor conditions. They then used a resist-accept-direct (RAD) framework to identify regional management options based on predicted changes in benthic macroinvertebrate communities under differing stressor conditions.
Major Findings
- Water quality and habitat related to stream health: Various water-quality and habitat stressors appeared to influence macroinvertebrate populations. The most prevalent stressors included temperature, specific conductivity, physical habitat alterations, and habitat heterogeneity.
- Regional variation highlights complexity in stressor relationships: Prevalent stressors linked to macroinvertebrate populations varied by region within the watershed. For example, in the Northern Appalachian region, water temperature and specific conductivity were the primary stressors, while in the Southern Appalachian region, physical habitat stressors, like embeddedness, were most influential (figure 2). In mixed-land use areas like the Piedmont region, both habitat and specific conductivity were related to measures of stream health, whereas in the Coastal Plains region, multiple measures of habitat drove patterns in stream health metrics.
- Widespread observational data and modeling reveal new insights: The team, using a novel modeling approach, paired an extensive observational biomonitoring dataset with key physical habitat and water quality stressor data to identify which stressors most influenced benthic macroinvertebrates. In using their findings to then predict anticipated changes in macroinvertebrates, the team was able to highlight potential management actions that consider the complexity of stressors facing aquatic habitats in the watershed.

Management Applications
The findings from this study suggest that targeted management strategies that consider the unique, interconnected, and regionally specific stressors within the Chesapeake Bay watershed may be warranted. The use of extensive biomonitoring data collected throughout the watershed in combination with a modeling approach that enabled the identification of regional differences in influential stressors generated management-relevant information. Specifically, the team’s approach could be used by managers in a resist-accept-direct (RAD) framework to make more informed decisions by utilizing measured field data rather than findings from small-scale experimental studies, which may not capture the complexity of natural ecosystems across large landscapes. For example, some streams in the Southern Appalachian region have high stream-health scores because of fewer water-quality and habitat stressors. As a result, managers could choose to protect these healthy streams by trying to prevent (resisting) the introduction of new stressors. In contrast, managers may need to consider simply accepting the degraded condition of streams facing multiple stressors, as it may be determined that available resources could be better used to improve the conditions of streams facing fewer stressors elsewhere.