Water Quality Criteria: Notice of Ambient Aquatic Life Water Quality Criteria for Dissolved Oxygen (Saltwater): Cape Cod to Cape Hatteras

AGENCY:

Environmental Protection Agency (EPA).

ACTION:

Notice of availability of ambient aquatic life water quality criteria for dissolved oxygen (saltwater): Cape Cod to Cape Hatteras.

SUMMARY:

Pursuant to section 304(a)(1) of the Clean Water Act (CWA), the Environmental Protection Agency announces the availability of the completed document titled, Ambient Aquatic Life Water Quality Criteria for Dissolved Oxygen (Saltwater): Cape Cod to Cape Hatteras. The document contains EPA's recommended national 304(a) criteria for dissolved oxygen in saltwater to protect aquatic life. These water quality criteria recommendations apply to coastal waters (waters within the territorial seas, defined as within three miles from shore under section 502(8) of the CWA) of the Virginian Province (Cape Cod to Cape Hatteras). However, with appropriate modifications they may be applicable to coastal waters (as defined under section 502(8) of the CWA) in other provinces of the United States. Under the CWA, States, Territories, and Tribes are to adopt water quality criteria to protect designated uses. EPA has promulgated regulations to implement this requirement (see 40 CFR part 141). EPA's recommended water quality criteria do not substitute for the Act or regulations, nor is it a regulation itself. Thus, EPA's recommended water quality criteria do not impose legally-binding requirements. States, Territories, and authorized Tribes retain the discretion to adopt, where appropriate, other scientifically defensible water quality standards that differ from these recommendations. EPA may change these section 304(a) criteria recommendations in the future.

Because these criteria were under development prior to the Agency's revision and implementation of its current processes for notice of data availability and criteria development (see Federal Register, December 10, 1998, 63 FR 68354 and in the EPA document titled, National Recommended Water Quality—Correction EPA 822-Z-99-001, April 1999), and because EPA believes it is important to invite and consider public input in development of draft criteria, we enabled the public to submit significant scientific information and views to EPA (see Federal Register, January 19, 2000, 65 FR 2954) that might not have otherwise been identified during development of these criteria. EPA has reviewed the scientific information and views submitted by the public and has made revisions to the criteria where appropriate. Even though we are not required to respond to specific issues submitted by the public, we have provided a brief summary of some of the issues that lead to a revision, along with our response, in the section titled Supplementary Information.

This document has been approved for publication by the Office of Science and Technology, Office of Water, U.S. Environmental Protection Agency. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.

ADDRESSES:

Copies of the complete document, titled: Ambient Aquatic Life Water Quality Criteria for Dissolved Oxygen (Saltwater): Cape Cod to Cape Hatteras can be obtained from EPA's National Service Center for Environmental Publications (NSCEP) 1-800-490-9198. Alternatively, the document and related fact sheet can be obtained from EPA's web site at http://www.epa.gov/waterscience/standards/dissolved/ on the Internet.

FOR FURTHER INFORMATION CONTACT:

For questions regarding the development of the criteria contact Erik L. Winchester, USEPA, Health and Ecological Criteria Division (4304), Office of Science and Technology, 1200 Pennsylvania Avenue, NW., Washington, DC 20460; or call (202) 260-6107; fax (202) 260-1036; or e-mail winchester.erik@epa.gov. For questions regarding implementation issues under State water quality standards programs contact Jim Keating, USEPA, Standards and Health Protection Division, (202) 260-3845; or email keating.jim@epa.gov.

SUPPLEMENTARY INFORMATION:

Introduction

Section 304(a)(2) of the CWA calls for information on the conditions necessary “to restore and maintain biological integrity of all * * * waters, for the protection and propagation of shellfish, fish and wildlife, to allow recreational activities in and on the water, and to measure and classify water quality.” EPA has not previously issued saltwater criteria for dissolved oxygen (DO) because, until recently, the available effects information was insufficient. This criteria document is the result of an extensive multi-year research effort to produce sufficient information to support the development of saltwater DO criteria. The water quality criteria presented in the document represent EPA's best estimates, based on the data available, of DO concentrations necessary to protect aquatic life and uses associated with aquatic life.

Overview of the Problem

Hypoxia is defined in this document as the reduction of DO concentrations in water below air saturation. Oxygen is essential in aerobic organisms for the proper functioning of cellular processes. When hypoxia exists, organisms may get an insufficient amount of oxygen into their system which results in reduction in cellular energy and a subsequent loss of ion balance in cellular and circulatory fluids. If oxygen insufficiency persists, death will ultimately occur, although some aerobic animals also possess anaerobic metabolic pathways, which can delay lethality for short time periods (minutes to days). The animals most sensitive to hypoxia are those inhabiting well oxygenated environments which are not normally exposed to low DO levels. EPA's Environmental Monitoring and Assessment Program (EMAP) for the estuaries in the Virginian Province (defined as Cape Cod to Cape Hatteras) has shown that 25% of the area of the Province is exposed to some degree to DO concentrations less than 5 mg/L. Persistent DO levels below 5 mg/L can have an adverse effect on various life stages of aquatic organism. EMAP also has generated field observations that correlate many of the biologically degraded benthic areas with low DO in the lower water column. These two reports serve to emphasize that low DO (hypoxia) is a major concern within the Virginian Province. Even though hypoxia is a major concern for many waters, a strong technical basis for developing benchmarks for low DO effects has been lacking until recently.

In the Virginian Province, hypoxia is essentially a warm water phenomenon. In the southern portions of the Province, such as the Chesapeake Bay and its tributaries, reduced DO may occur any time between May and October; in the more northern coastal and estuarine waters, it may occur at any time from late June into September. Hypoxic events can occur on seasonal or diel (daily) time scales. Seasonal hypoxia often develops as a consequence of water column stratification, which prevents mixing of well oxygenated surface water with deeper water. Diel cycles of hypoxia often occur in non-stratified shallow habitats where nighttime respiration temporarily depletes DO levels. Hypoxia may also persist more or less continuously over a season (with or without a cyclic component) or be episodic (i.e., of irregular occurrence and indefinite duration). The fauna most at risk from hypoxic exposure in the Virginian Province are primarily summer inhabitants of subpycnocline (i.e., bottom) waters.

Overview of the Protection Approach

The approach to determine DO criteria to protect saltwater animals within the Virginian Province takes into account both continuous (i.e., persistent) and cyclic (e.g., diel, tidal, or episodic) exposures to low levels of DO. The continuous situation considers exposure durations of 24 hours or greater. Criteria for cyclic situations cover hypoxic exposures of less than 24 hours, but which may be repeated over a series of days. Both scenarios cover three areas of protection that are summarized here, and explained in more detail in the criteria document: (1) Protection for juvenile and adult survival; (2) Protection for chronic (growth) effects; and (3) Protection for larval recruitment effects (estimated with a generic recruitment model).

The approach to derive these DO water quality criteria combines features of traditional water quality criteria with a new biological framework that uses a mathematical model to integrate time (replacing the concept of an averaging period) and establish protection limits for different life stages (i.e., larvae versus juveniles and adults). Where practical, data were selected and analyzed in manners consistent with the Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of Aquatic Organisms and Their Uses (hereafter referred to as the Guidelines).

The saltwater DO criteria segregate effects on juveniles and adults from those on larvae. The survival data on the sensitivity of the juveniles and adults are handled in a traditional Guidelines manner. To address cumulative effects of low DO on larval recruitment to the juvenile life stage (i.e., larval survival as a function of time), the new biological approach to deriving criteria uses a mathematical model that evaluates the effect of DO conditions on larvae by tracking intensity and duration of effects across the larval recruitment season. Protection for larvae of all species is provided by using toxicological data on the larval stages of nine sensitive aquatic organisms.

The approach used to derive the new DO criteria deviates somewhat from EPA's traditional approach for toxic chemicals outlined in the Guidelines. However, where practical, data selection and analysis procedures are consistent with the Guidelines. Most of the terminology and the calculation procedures are the same, but one should consult the Guidelines for a more complete understanding of how these DO criteria were derived.

The juvenile/adult survival and the growth criteria provide boundaries within which to judge the DO status of a given site. If the DO conditions are above the chronic growth criterion (4.8 mg/L), then this site would meet objectives for protection. If the DO conditions are below the juvenile/adult survival criterion (2.3 mg/L), then this site would not meet objectives for protection. When the DO conditions are between these two values, then the site would require evaluation using the larval recruitment model that integrates duration and intensity of hypoxia to determine suitability of habitat for the larval recruitment objective.

The DO criteria are based entirely on laboratory findings. Field observations on the impact of low DO levels support the findings of laboratory studies. Field acute effects occurred in juvenile and adult animals at <2.0 mg/L, which would be predicted based on the 2.3 mg/L juvenile/adult criterion. In the field, behavioral effects generally occurred within the range where many of the laboratory sublethal effects occurred.

Revisions to the Draft Document

Approximately half of the views and information submitted by the public on the draft DO criteria addressed science or technical issues, and the other half addressed implementation issues. EPA considered only the science issues when making revisions to the criteria. EPA will review the implementation issues when developing future implementation guidance. The more significant revisions due to science issues are summarized here.

First, some commentors indicated that the larval recruitment model should not be based on the mud crab (Dyspanopeus sayi) alone. Based on further review of the toxicity information for other species, we have revised the dose-response curve in Figure 5 by using a final acute value (FAV) approach (see the Guidelines) to generate a new final larval survival curve that reflects responses of all nine species tested. Figures 5a and 5b have been replaced by a “Final Larval Survival Curve”, and Figure 5c has been removed. These changes to the larval recruitment approach necessitated that changes also be made to Figures 6, 7, 12, 14 and 17. Overall, these changes had minimal effect on the criteria. The point (4.64 mg/L) at which the larval recruitment curve levels off in the revised criteria is only slightly greater than the point (4.45 mg/L) in the draft document.

Second, some commentors raised issues about the effect that differences in larval life history requirements among species in the Virginian Province might have on the applicability of the larval recruitment model across species and regions in the Province. The consideration of all nine species in development of the larval recruitment model addresses this issue. Also, in an appendix we added an assessment of sensitivity that might be expected with the life history model parameters D (duration of larval development) and R (length of larval recruitment season). The sensitivity analysis was performed using the individual larval recruitment curve for the mud crab. The sensitivity of the model to these two parameters was evaluated by increasing or decreasing D while holding R constant, by holding L constant and increasing R, and varying both D and R at the same time. A range of values were chosen for this analysis that we believe encompass a reasonable range in species-specific larval life history requirements in the Virginian Province, and because the upper and lower ranges are relatively extreme values that can test the overall assumptions and sensitivity of the model. The results indicate that the DO curve associated with no greater than 5 percent cumulative impairment of seasonal larval recruitment is most sensitive to a simultaneous decrease in D and increase in R. Under these conditions, the protective DO value at 44 days (the length of mud crab larval development season) decreases. This evaluation shows that the model can easily be adjusted to account for latitudinal variations in life larval life history requirements, or even seasonal variations in timing of hypoxia events concurrent with larval development periods. The results also indicate the Virginia Province criteria are protective of most species under most conditions, but that in some site-specific situations they may be overprotective. In the absence of site-specific data that would suggest a lower level of DO may be acceptable, EPA believes that in order to ensure that most organisms and their uses are protected it is appropriate to derive Province-wide criteria that may be overprotective in some cases.

Third, some commentors suggested that the 5 percent cumulative reduction in larval seasonal recruitment may be too low a protection goal. EPA disagrees. Larval life stages are important and this protection goal is meant to protect them at a critical point in their development and transition to the juvenile life stage, which for many species corresponds to times of the year when hypoxia conditions occur. We selected this Province-wide protection goal because it is consistent with the approach outlined in the 1985 Guidelines for deriving ambient aquatic life water quality criteria, because 5 percent is also consistent with the level of protection afforded to juvenile and adult life stages, and because, in absence of data that suggests otherwise, this level of reduced larval recruitment from DO alone is believed to be protective of most species. EPA recognizes that large losses of larval life stages occur naturally, and that many species may be able to withstand a greater than 5 percent loss of larvae, from low DO or otherwise, without an appreciable effect on juvenile recruitment. On the other hand, this may not be the case for certain highly sensitive species or populations that are already highly stressed, for example an endangered species. This may also not be the case where there are other important natural or anthropogenic stressors that contribute to a loss of the larval life stage. In such situations, it may be that a 5 percent loss in larval recruitment from DO alone is not protective enough, and environmental risk managers may need to evaluate the Province-wide 5 percent protection goal in light of their site-specific factors that may contribute to a cumulative loss in seasonal larval recruitment. Also in response to this issue, an appendix was added to the document that shows, by using the mud crab as an example, how the larval recruitment criterion would change if the acceptable percentage impairment was increased. This example demonstrates the flexibility in the criteria approach and how one might change the protection goals on a site-specific basis should States and authorized Tribes choose to do so and have the data to support such a change, while still protecting designated uses. EPA believes the 5 percent cumulative reduction level in seasonal larval recruitment is appropriate and protective of populations in the Virginian Province in absence of data that suggest otherwise.

Implementation Overview

Implementation of DO criteria may be slightly different from that of chemical toxicants, but not for reasons associated with either biological effects or exposure. The primary reason that DO might be implemented differently from toxic compounds is because controlling the effects of low DO is not accomplished by directly regulating DO. Rather, hypoxia is a symptom of a problem, not the direct problem. Thus DO would be regulated primarily through the control of nutrients (e.g., nitrogen and phosphorus) and oxygen demanding wastes. As a stressor, DO also differs from most toxic compounds in that there can be a large natural component to the cause of hypoxic conditions in any given water body.

The DO criteria may also be appropriately used in a risk assessment framework. The criteria and management approach presented in the document could be used to compare DO conditions among areas and determine if DO conditions would be adequate to support aquatic life. Using the criteria, environmental managers could determine which sites need the most attention and what are the spatial and temporal extent of hypoxic problems from one year to the next. Environmental planners could also use the criteria in a risk assessment framework to evaluate how conditions would improve under different management scenarios, helping them make better management decisions.

EPA recommends that States and authorized Tribes within the Virginian Province adopt numeric DO criteria for saltwater applicable at all times of the year for all marine waters designated for the protection of aquatic life or for waters whose existing uses include aquatic life. States and Tribes may adopt numeric criteria based on EPA's ambient water quality criteria for DO, such criteria modified to reflect site-specific conditions, or other scientifically defensible methods, 40 CFR 131.11(b)(1). States and Tribes should adopt narrative criteria where numeric criteria cannot be established or to supplement numeric criteria, 40 CFR 131.11(b)(2). Because EPA has issued recommended section 304(a) criteria for DO, numeric criteria for DO can be established. Numeric criteria for DO can be implemented in NPDES permits by determining the need for and calculating specific limits for oxygen demanding wastes and nutrients that spur excess algal growth and subsequent decay of aquatic plants. Such criteria also serve as a definitive benchmark for determining impairment of waters for Clean Water Act Section 303(d) listing purposes and then as a starting point for establishing TMDL's, wasteload allocations for point sources, and load allocations for nonpoint sources.

To take full advantage of the flexibility allowed in the DO criteria methodology for determining specific protective DO levels, it is necessary to characterize both the diurnal and season patterns of DO concentrations in response to natural and anthropogenic pollutant loadings for the location where the criteria are applied. Simplified approaches to establishing protective criteria that ensure a level of protection consistent with the detailed approach outlined in the DO criteria document are acceptable. Any approach a State or Tribe chooses to use to implement the DO criteria must be reflected in the State's or Tribe's water quality standards and submitted to EPA for review and approval. To determine the scientific defensibility of a State's or Tribe's approach as part of the Clean Water Act section 303(c) review and approval/disapproval process, EPA will review information concerning the characterization of diurnal and seasonal patterns of DO concentration in relation to the geographic areas and the times of the year the criteria applies, and would want the State or Tribe to provide all of the data and information the State or Tribe relied on for its rationale.

Limitations of the Criteria

These water quality criteria recommendations apply to coastal waters (waters within the territorial seas, defined as within three miles from shore under section 502(8) of the CWA) of the Virginian Province (southern Cape Cod, MA to Cape Hatteras, NC) of the Atlantic coast of the United States. The document provides the necessary information for environmental planners and regulators within the Virginian Province to address the question: are the DO conditions at a given site sufficient to protect coastal or estuarine aquatic life? The approach outlined in the document could be used to evaluate existing localized DO standards or management goals or establish new ones. The criteria do not address direct behavioral responses (i.e., avoidance) or the ecological consequences of behavioral responses, such as increased or decreased predation rates or altered community structure, nor do they address the issue of spatial significance of a DO problem. In addition, as with all criteria, these criteria do not account for changes in sensitivity to low DO that accompany other stresses, such as high temperature, extremes of salinity, or toxicants. Chief among these concerns would be high temperature because high temperature and low DO often appear together. Generally, low DO would be more lethal at water temperatures approaching the upper thermal limit for species. EPA believes the DO limits provided in the document are sufficiently protective under most conditions where aquatic organisms are not otherwise unduly stressed.

Although the DO criteria for the Virginian Province may be over- or underprotective of aquatic life in other regions, the approach used to develop the criteria is considered to be applicable to other regions with appropriate regional modifications. Organism adaptations to lower oxygen requirements may have occurred in locations where oxygen concentrations have historically been reduced due to high temperatures, or in systems with non-anthropogenic high oxygen demand. Conversely, organisms in another region could be adapted to colder temperature and higher DO regimes than those covered in the document, and thus may have different sensitivity to DO concentrations. In addition, effects of hypoxia may vary latitudinally, or site-specifically, particularly as reproductive seasons determine exposure risks for sensitive early life stages. For these reasons, an environmental risk manager would need to carefully evaluate water quality and biological conditions within the specific location and decide if the Virginian Province criteria would apply or if region-or site-specific considerations would need to be made.

Endangered or Threatened Species Policy Recommendations

When a threatened or endangered species occurs at a site and sufficient data are available to indicate that it is sensitive at concentrations above the recommended criteria, it would be appropriate to consider deriving site-specific DO criteria.

Future Implementation Information and Applications

In the future additional information will be provided that will specifically address implementation issues. In the current document, implementation issues are discussed in a more general manner, summarizing important issues that environmental managers should consider in adopting and implementation of DO water quality standards. The future implementation information will provide more detailed discussion of implementation issues by using real world example data sets where possible, or hypothetical data sets that show users how to integrate their data and management goals. Application of this guidance to marine waters outside the Virginian Province may also be discussed. As a component of the implementation guidance, EPA originally envisioned publishing a visual basic-based computer program that would allow States and other users to derive DO criteria to meet the larval recruitment protection goal for coastal and estuarine animals. However, the recent revisions in the criteria (i.e., the use of multiple species in the larval recruitment model) has precluded use of the visual basic model in its current format. Therefore, the model when available will likely be provided as a spreadsheet application compatible with commonly used software packages. EPA anticipates providing the additional implementation guidance in late 2001.

EPA believes the approach used to develop the criteria can be applied, with minor modifications and regional specific data, to derive DO criteria for other coastal and estuarine regions of the United States. Therefore, in the future EPA plans to prepare similar DO criteria for other provinces based on this approach. At such time, EPA intends to publish a Notice of Data Availability and formally request submission of data from parties interested in the development of DO criteria for other provinces.