![[logo] US EPA](http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Screen Attributes to Define Metrics
- Classify coral reefs
- Develop testable hypotheses
- Define data collection protocols
- Biological sampling
- Screen attributes to define metrics
- Determine appropriate sampling effort
- Define data protocols
- Validate decision processes
- Define biocriteria
- Implement monitoring programs
- Diagnose causes of degradation
- Evaluate management effectiveness
- Communicate results
Select metrics based on observed changes in value across a human influence gradient (i.e., dose-response curves).
Sampling reef areas across a gradient of human influence provides an immediate test of the sensitivity and responsiveness of the candidate coral metrics. To be selected as a metric or bioindicator, a measured attribute of the coral assemblage must demonstrate a reliable and consistent association with human disturbance.
For the initial development and testing of biological metrics, a small spatial scale can simplify the process of selecting metrics that are good indicators of biological condition. Metric values within a smaller spatial area should be more similar than those collected from a larger regional area. Consequently, differences in metric values may be more easily associated with disturbance observed at the sampling locations. When testing metrics with a larger regional data set, natural gradients must be considered to be sure that a change in metric values is due to human influence and not confounded with a natural gradient associated with latitude or salinity, for example.
Figure. The percentage of diseased coral is shown for coral reef sites located along the Florida Keys.More diseased coral was found near Key West and in the Upper Keys where the reefs are closer to urban development; however, human disturbance was not quantified for this study.
- Develop scoring rules for metrics based on expectations for reference condition
One benefit associated with sampling across a gradient of disturbance when testing metrics is that it simplifies setting expectations for metrics. In order to define biocriteria, we need to know what range of metric values represents the best possible condition as well as how far metric values can decline in response to degradation.