Segmenting Large Water Bodies to Develop Protective Criteria for Distinct Regions

In Nutrient Criteria Technical Guidance Manual for Lakes and Reservoirs, EPA recognized that establishing a single, national nutrient criteria for lakes is not a realistic goal because of the significant variability of water bodies that exist across the country in a variety of climates, geographic locations, and ecosystems. In addition, within a region or a single water body, variability can exist due to differences in microtopography and tributary inputs. As a result, the nutrient criteria development process uses a classification system to clarify those differences. Classifying rivers, lakes, and estuaries reduces the variability within classes and maximizes the variability between classes. The goal is to develop criteria that protect water body uses in each class. Similarly, the goal in segmenting large water bodies into smaller, distinct regions is to ensure that each important region within a water body is equally protected.

Purpose for water body segments

Segmenting large water bodies is a way to develop protective criteria for each distinct region of a water body. The result of protective criteria for distinct water body classes is similar for classification in both the reference condition and the stressor-response approaches. Mechanistic models are usually applied to individual water systems, thus classification may not be necessary to develop a model. However, some differences can exist within larger water bodies that would require model parameterization or model inputs to be handled differently in various areas of the model (e.g., nutrient fluxes greater near large source of nutrient loads). In addition, criteria developers, might choose to segment a water body to recognize water quality differences along gradients within a water body and natural/man-made geographical dividing features (e.g., a causeway or narrowing of the water body) to ensure that the criteria are protective. When possible, it is best to align the classification segments for model parameterization and water quality criteria.

Water body attributes for segmenting

Water bodies might be segmented based on a variety of attributes depending on water body type. Characteristics to consider when segmenting include:

  • In estuaries, habitat, salinity, depth (including channels), embayments, open water, and tidal tributaries;
  • In rivers, slope, velocity (e.g., steep fast-flowing versus flat slow-moving), tributary inflow, water clarity (e.g., clear versus turbid), and shading; and
  • In lakes, embayments, open water, forebay, and depth.

For example, a long river might transition from clear to turbid, or from fast-flowing, turbulent water to slow, tranquil water. Each river segment could have different algal response to the light availability and current velocity. A water system also could have major chemical changes where tributaries from different watersheds flow into the river, lake, or estuary. Large lakes and estuaries can have embayments with drastically different characteristics from the main channel. In estuaries, salinity, physical features such as bridges and causeways, SAV coverage, and depth distribution might be considered to account for differences in hydrology and ecosystem dynamics. For example, the Chesapeake Bay has been divided into segments based on five principal habitats and salinity regimes, and criteria have been developed to protect the specific aquatic life uses in each segment.

Model considerations for segmentation

If you plan to segment a large water body, you should consider how each segment relates to the model. For example, you might need to define the model grid with enough detail for each segment and align model grid cell boundaries with segment boundaries. Also, you might consider whether adequate calibration data exist for each segment and whether you have targets for each segment. Targets for every segment are not necessary if adjacent or nearby segments have targets.

Case Studies

St. Louis Bay, MS

  • Five spatial strata were identified based on classification process
  • Strata were used as the spatial framework for the analysis
This website is in beta. Information on this website is not final and is subject to change