Water: Recovery Potential
Step 2: Design the Approach
Screening Methodology Tutorial
Well-organized data are the foundation of any assessment. Key organizing elements take shape in the screening design step. These include:
Establish IDs. In the first step, you selected the targeted units for your screening assessment - usually one or more types of water bodies or watersheds. You now need a unique identifier for every individual unit you plan to assess. Organized systems of IDs probably already exist and should be used if possible. For example, state-reported impaired waters under Clean Water Act Section 303(d) each have a List ID available in state datasets or in EPA's ATTAINS data system. The National Hydrography Dataset (NHD) also represents a standard source for identifiers of water bodies or segments. If your focus is on watersheds, the Watershed Boundary Dataset (WBD) is a widely-used national source of drainage-based hydrologic units at several scales, each with a common ID system. The value of using existing ID schemes is not only to use terms familiar to others, but also because many of the measurements needed for recovery potential indicators, such as water quality monitoring or land cover data, may have already been compiled and referenced to these IDs.
Select candidate indicators. Note that substantial information on recovery potential indicators is available through this website in the form of a recovery literature database, indicators lists, and indicator-specific fact sheets. Although many indicators that may influence recovery are summarized here for your consideration, the objective of this step is to use your own group's expert judgment to identify a number of factors that appear to be most relevant to the set of waters being assessed. This is best done in an informal group discussion involving those most familiar with the area's water bodies, impairment types, and restoration track record to date. The indicator selection worksheet (PDF) (2 pp, 131K, About PDF), which contains example indicators, may be helpful to jump-start the process of discussing and identifying the factors most relevant to restorability. Note that there is no need to use all indicators; also note that adding new indicators is appropriate where the existing list doesn't capture factors that are highly relevant to your waters or describe them in sufficient detail. The workgroup should feel free to identify, modify or add key factors that are not listed. Consider this stage of indicator selection more inclusive than exclusive - capture all the candidate ideas now and reduce to manageable numbers later.
The candidate indicators are organized in three classes - ecological, stressor, and social context - based on the three main mechanisms by which natural and human-made driving factors influence recovery potential. Ecological indicators measure those properties that are related to the capacity to maintain or reestablish natural structure and processes. Stressor indicators are associated with reduced natural function due to the negative impacts of pollutants and other stressors. Social context indicators address a broad array of community, regulatory, economic, and behavioral measures that often have a profound influence on restoration success independent of the environmental factors. Although there is much flexibility in the choice of individual indicators, it is crucial to measure some indicators in each of these three classes in order to carry out Recovery Potential Screening.
Selecting indicators in each of the three classes is not only influenced by what can be measured with the available data, but also by the need to assemble a collection of recovery-relevant factors that each provide a different 'piece of the puzzle'. It may not be possible to obtain data on all of the factors you'd like to measure, but every effort should be made to measure indicators that are not all related to the same component of their indicator class (e.g., ecological indicator selection should include more than just different measurements of watershed land cover). Ideally, the selection should attempt to address as many of the key components below as possible:
Key components for ecological indicator selection include:
- Ecological history;
- Watershed natural land cover;
- Flow dynamics;
- Channel, corridor and near-shore stability;
- Biotic community integrity; and
- Aquatic connectivity.
Key components for stressor indicator selection include:
- Watershed-level disturbance;
- Corridor or near-shore disturbance;
- Biotic or climatic risks;
- Severity of pollutant loading;
- Hydrologic alteration; and
- Legacy of past, trajectory of future land use.
Key components for social context indicator selection include:
- Watershed-level leadership, organization and involvement;
- Protective ownership or regulation;
- Level of information and planning;
- Restoration cost, difficulty, or complexity;
- Socio-economic considerations; and
- Relevance to human health and beneficial uses.
Identify and compile the needed datasets. Once the candidate indicators have been identified, available data sources should be considered. In developing this website's list of example indicators, factors measurable from commonly available data were emphasized. Two types of data sources yield most of the indicators: geospatial data and water quality monitoring data. Geospatial data (also called landscape, mapped, or Geographic Information Systems (GIS) datasets) represent a wide variety of water or watershed characteristics that are often uniformly available (and thus comparable) across large areas and can be directly measured, counted or otherwise quantified to obtain specific recovery potential indicator values. Monitoring data, which may also be available in geospatial form or referenced to specific locations, refers mainly to consistent measurements of aquatic conditions gathered through statewide ambient water quality monitoring programs under the Clean Water Act and consolidated by EPA into national datasets. Any dataset used must be capable of producing consistent, comparable values for all of the waters being screened. The indicator reference sheets each contain notes on commonly available national data sources for their measurement, but state-level or other sources may also exist. Once the candidate indicators and their data sources are selected, compiling the datasets can begin.
Define how each indicator is measured. Having verified that appropriate data are available, an exact process for measuring each candidate indicator now needs to be defined and verified. More than one measurement method is often possible; see the ways to measure an indicator on the scoring guidance and tools page.
Although options for measurement are provided in general terms in the indicator fact sheets, determining the final measurement method is a project-specific decision based on the available data, data quality considered acceptable for the project, and the underlying reasons for the indicator's selection. Multiple measurement methods for the same indicator are often possible, and testing and comparing their results is advisable. In order to keep good records of all indicators you evaluate for possible use, give a unique name to each version of each indicator you will measure and document the way it is measured, including units. All candidate indicator measurement methods should be well documented, transparent and reproducible. You can download, rename and use the indicator tracking record (MS Excel xlsx, 16K) for this purpose.
Review and refine preferred indicators. Indicator selection, the search for data, measurement method development, and testing may significantly modify your initial assumptions about indicators. Several candidate indicators may be abandoned due to insufficient data sources, quality or consistency. This final stage of step two is a reexamination of all the candidate indicators and measurement alternatives that you would like to include in the screening. Have your workgroup review and discuss these indicators, add new metrics or remove less relevant or redundant metrics, and also consider data quality in this process. Summarize the outcome in the indicator tracking record for each metric the group discusses; you may choose to mark the preferred indicator selections in the "Usage1" field. An initial list of five to ten preferred indicators in each of the three classes is a desirable target, but if resources permit, having many more at this stage is not problematic and will enable you to explore more alternative combinations of indicators to address more screening purposes.