Quantifying and Documenting Protection of the Designated Use

Deriving NNC is a scientific synthesis process that moves from interpreting the results of a particular approach to quantitative estimates of the criteria. The two elements comprising the process—quantifying and documenting the level of protection that is necessary to protect the designated use—rely on the analytical results of a particular approach. The process employs commonly used approaches and best practices that constitute the traditional scientific process.

Quantifying

Each of the analytical approaches—the reference condition, stressor-response models, and mechanistic models—provides quantitative outputs that relate nutrient stressors to assessment endpoints. Rigorous examination and interpretation of these quantitative relationships provide information about the impacts of nutrient pollution on designated uses.Key considerations can be organized by data adequacy and quality, degree and type of uncertainty associated with each analytical line of evidence, and the relationship of each analysis to the conceptual model relating nutrients to assessment endpoints. Considerations listed are relevant to all three approaches unless otherwise noted.

Data Adequacy and Quality

Reporting the temporal and spatial scales, as well as any temporal and/or spatial aggregation, of the original data included in the analysis and the implications for interpretation of the data.
Quantifying the natural or least disturbed condition based on rigorous, quantitative selection factors (applicable only to the Reference Condition approach).

Degree and Type of Uncertainty Associated with Each Analytical Line of Evidence

Expressing, through quantitative classification, that the physical, chemical, and biological properties of the waters being analyzed are similar to the waters to which criteria will apply (applicable only to the Reference Condition and Stressor-Response approaches).
Reporting the descriptive statistics (e.g., measures of central tendency, percentiles, measures of variation) or estimated parameters of a statistical distribution or model, including the statistical confidence of those statistics and estimates.
Documenting how certain a priori analytical assumptions associated with each approach are satisfied.

Relationship of Each Analysis to the Conceptual Model Relating Nutrients to Assessment Endpoints

Reporting and justifying the a priori desired quantitative target for a given assessment endpoint.
Quantifying the magnitude of the criteria.
Quantifying the duration and frequency of exposure at a specific magnitude, which ensure protection of the assessment endpoint.
Explicit consideration that estimated criteria also will protect downstream designated uses.
Documenting how certain a priori analytical assumptions associated with each approach are satisfied.

When the results of each approach have been examined in the context of these considerations, relative interpretations can be derived for that approach and used in selecting an appropriate value. This is especially helpful when two or more approaches are used or when an approach is compared to other approaches and estimates reported in the peer-reviewed literature. For example, the same underlying data can be used in reference condition and mechanistic modeling approaches; however, the results could differ substantially with respect to temporal and spatial scales. A reference condition approach might yield annual average estimates for criteria for a small portion of a lake whereas a mechanistic model’s analytical output might yield monthly or seasonal estimates for criteria for the entire area of the same lake. Interpreting and reconciling these differing results are important aspects of the synthesis process of deriving criteria. Results can be expressed in terms of the strengths and dimensions (spatial and temporal) of each approach and how they compare to one another. Or a rationale can be applied that describes how the approaches offer independent agreement and clarity on the ways in which nutrient stressors relate to assessment endpoints and, in turn, designated use protection.

Documenting

Hallmarks of rigorous NNC development are clarity, transparency, reasonableness, and consistency. Thorough documentation of the data and analyses within these four elements provide confidence and rigor (USEPA 1998).

Clarity

Describing the process used in interpreting the analyses in a manner that is clearly replicable by an appropriately experienced technical audience.
Fully discussing and explaining any unusual issues.

Transparency

Documenting the scientific basis for the conceptual model used to illustrate the assessment endpoints of interest (and their connection to designated uses), as well as the measures of effect and exposure used in the analysis.
Documenting the data used in the analysis: environmental sample collection, data generation, data management, data screening (refer to Box 1 – Documenting Data).
Documenting how the analysis was performed, including what assumptions were made and why, so analytical results can be reproduced independently;
Documenting the stepwise inferences made between the results of the analysis and estimated criteria.

Reasonableness

Acknowledging uncertainties and assumptions in a forthright manner.
Explaining confidence that the models (reference, stressor-response, or mechanistic) developed reflect the a priori conceptual model of nutrient pollution and its effects.
Explaining confidence that the underlying data represent water quality conditions over space and time of the waters to which the criteria will apply.
Describing reasonable alternatives and conclusions that can be drawn (refer to Box 2 – Combined Criteria).

Consistency

Describing how estimated criteria compare to similarly protective values estimated from other efforts.

Box 1 – Documenting Data

Any analysis is only as rigorous as its underlying data. Data can be documented in the following ways:

A data quality objectives plan (sometimes referred to as a “quality assurance project plan” or QAPP) that documents how environmental samples will be collected, methods that describe how data will be generated, and independent checks on the quality of the data that are generated.
Reporting when secondary data were used and documenting their sources.
Reporting sampling locations, number of samples, and the date and time of collection.
Reporting limits of detection and the detection of outliers.
Describing how censored data were handled (e.g., removed, included).
Analyte reporting (e.g., was total nitrogen measured directly or estimated? Were nutrient concentrations reported as elemental concentrations or compound-specific concentrations?)
Reporting how data were aggregated (spatially and/or temporally) for analysis.

Box 2 - Combined Criteria

EPA recognizes that developing numeric values for both nitrogen and phosphorus may present challenges associated with temporal and spatial variability, particularly in flowing waters. For this reason, guidance has been developed for an alternative combined approach to nutrient criteria development and application that integrates both causal and response variables into a single water quality standard.

The following resources describe this approach in more detail: