Evaluating Flow Metric‐Based Stream Classification Systems to Support the Determination of Ecological Flows in North Carolina

Abstract

Hydroecological classification systems are typically based on an assemblage of streamflow metrics and seek to divide streams into ecologically relevant classes. Assignment of streams to classes is suggested as an initial step in the process of establishing ecological flow standards. We used two distinct hydroecological river classification systems available within North Carolina to evaluate the ability of a hydrologic model to assign the same classes as those determined by observed streamflows and to assess the transferability of such systems to ungaged streams. Class assignments were examined by rate of overall matches, rate of class matches, spatial variability in matches, and time period used in class assignment. The findings of this study indicate assignments of stream class: (1) are inconsistent among different classification systems; (2) differ between observed and modeled data; and (3) are sensitive to the period of record within observed data. One clear source of inconsistency/sensitivity in class assignments lies with the use of threshold values for metrics that distinguish stream classes, such that even small changes in metric values can result in different class assignments. Because these two hydroecological classification systems are representative of other classification systems that rely on quantitative decision thresholds, it can be surmised that the use of such systems based on stream flow metrics is not a reliable approach for guiding ecological flow determinations.

Fish and Invertebrate Flow‐Biology Relationships to Support the Determination of Ecological Flows for North Carolina

“…Combined with the findings of Eddy et al . (), that hydrological stream classification systems based on flow metrics do not support the determination of ecological flows in all rivers and streams in North Carolina, these collective results strongly support and defend the development of flow‐biology relationships at the statewide level. Although not explicitly tested, representing biological response through classification systems related to flow ( i.e ., flow‐based habitat guilds and EPTr), and not individual species, may have enabled classification at the state level.…”

“…See Eddy et al . () for a description of the locations of these gages. At the 28 gages, a five‐year calibration period (water years of 2002 through 2006) followed by a five‐year validation period (water years 1997 through 2001) were used to set and validate model parameters.…”

“…Similarly, past work has suggested, and the EFSAB therefore hypothesized, that determining significant flow‐biology relationships requires some form of restriction to river types/classes ( i.e ., Ecological Limits of Hydrologic Alteration [ELOHA]—Poff et al ., ), specific geographical regions, or other comparable ecological organization so that flow‐biology relationships are internally consistent. Initial EFSAB activities assessed flow‐based stream classification systems using statewide data and showed that existing stream classification systems based on hydrologic metrics could not be reliably reproduced and/or extrapolated to all streams within the state (Eddy et al ., ). Therefore, grouping flow‐biology relationships by hydrologic stream class was ruled out for determining relationships for North Carolina.…”

Featured Collection Introduction: North Carolina Ecological Flows

“…Eddy et al . () describes the Team's evaluation of different approaches to stream classification using two systems developed for North Carolina and the southeastern United States. The two hydroecological classification systems did not produce similar results using either observed or modeled daily time series of streamflow data.…”

Statistical Models to Predict and Assess Spatial and Temporal Low‐Flow Variability in New England Rivers and Streams

“…; MAEEA ), but have not yet been widely used in evaluating aquatic life use class designations, reference condition, or differential risks to aquatic populations over space or time (Eddy et al. ; Patterson et al. ; Phelan et al.…”