Peer Review Comments and Responses on National Nutrient Strategy

REVIEWERS

David Correll, Ph.D.
Smithsonian Environmental Research Center

William Keith
State of Arkansas, Department of Pollution Control and Ecology

Douglas Knauer
Chief
State of Wisconsin, Department of Natural Resources, Bureau of Integrated Science Services Research Center, Environmental Contaminants Research

Karol Erickson
Environmental Engineer
State of Washington, Department of Ecology, Environmental Investigations and Laboratory Services Program

Michael F. Hirshfield, Ph.D.
Vice President
Chesapeake Bay Foundation, Resource Protection

Tom Fontaine
Director
SFWMD, Everglades Systems Research Division

Jonathan Higgins
The Nature Conservancy

Robert G. Wetzel
Bishop Professor of Biological Sciences
The University of Alabama, Department of Biological Sciences, Aquatic Biology Program

Vladimir Novotny, Ph.D., P.E.
Professor of Environmental and Water Resources Engineering

Hassan Mirsajadi
State of Delaware, Department of Natural Resources and Environmental Control

GENERAL COMMENTS, CONCERNS, QUESTIONS

• Needs are urgent. (W. Keith)
• Imperative that subsequent actions be to recruit all states to be involved in the process so that the results will be a product the states are sold on and will adopt rather than be forced into. (W. Keith)
• What is the relationship between this strategy and TMDLs for waterbodies with excess nutrients? (K. Erickson)

Editorial concerns:
Spell 'phosphorus' correctly.
Edit for consistencies in format.

RESPONSES TO PROVIDED QUESTIONS

1) Is the regionalized/watershed approach described in the strategy:

a) likely to accomplish the desired goal of reducing overenrichment problems in the Nation's surface waters? If not, what would you suggest?

William Keith, State of Arkansas

• In the absence of any other formalized and directed approach on the national, regional and probably state scale, this proposal offers significant opportunity to adequately address nutrient overenrichment. The ultimate goal of controlling the problem will likely depend more on the political issues than the scientific ones; however, a well directed scientific approach, as is proposed, must be part of the process.

Douglas Knauer, State of Wisconsin

• No. The regionalized/watershed approach is correct but an approach does nothing to create a reduction in the overenrichment problems. It will however identify the overenrichment problems in the Nation's surface waters.

  For the approach to work the stepwise progression on p.5 should be modified. The first step should be Public Involvement, without it the process will fail. Step 9 should be Project Implementation. This draft strategy addresses only the problem identification but not the means to successfully implement the project.

EPA Response: These suggestions have been added to the 10-step management process presented in the Strategy

Karol Erickson, State of Washington

• The approach is on the right track. A regional/watershed-specific approach is necessary for nutrient criteria. However, implementation of control measures will be critical to accomplishing the goal and the Strategy does not discuss this in detail. It starts us on the path toward accomplishing the goal but the guidance documents will be more important to accomplishing the goal. Presumably, the strategy relies on existing regulatory authority for reducing nutrient sources. This approach will likely be as successful as other recent efforts to reduce nonpoint pollution, where progress has been very slow.

EPA Response: EPA has added more details on implementation and expectations on the part of States/Tribes in the Strategy.In addition, details on implementation will evolve, naturally, as Regional Nutrient Teams are formed, meet and debate these important issues.

Michael Hirshfield, Chesapeake Bay Foundation

• No. The Strategy is ambiguous on States' responsibilities in establishment of nutrient criteria (States may, can, should or must develop numeric criteria?). Implies that it is optional (strategy as well as memo from Tudor Davies and Bob Wayland). My reading of this strategy is that EPA intends simply to provide information to the states and hope that they use it to develop criteria for nutrients. This will not do. Part of the rationale behind EPA's limited role of consultant is that a 'single national number' for nutrients is not appropriate. However, it is not appropriate for metals either, and EPA has developed freshwater and marine numbers for many of them. This reason should not limited EPA's role to providing guidance to the states as they develop criteria. Responsibility of developing nutrient criteria should not be left up to states. Recommend that EPA establish them and states adopt them. EPA should not play the role of the consultant and merely provide guidance. The alternative of states establishing them and EPA stepping in if the states do not leaves too much opportunity for inconsistencies.

EPA Response: EPA has changed the Strategy to more clearly present EPA's, State and Tribes's responsibilities in developing and adopting nutrient criteria within a prescribed time period. EPA will take a large role in collecting nutrient data, developing waterbody-type guidance, and developing ecoregional nutrient criteria by 2000. It is expected that States will use both the waterbody-type guidance and ecoregional nutrient criteria developed by EPA to develop their own nutrient criteria and ultimately, water quality standards for nutrients by 2003, where sufficient data are available.

Tom Fontaine, SFWMD

• Yes. A regionalized watershed approach should, in concept, accomplish the goal of reducing overenrichment. It is certainly a superior approach to setting nationwide, non-watershed based criteria.

Jonathan Higgins, The Nature Conservancy

• The regional approach will not sufficiently provide a framework to identify natural trophic conditions unless a finer scale classification is used to identify lake and stream types that have similar expected chemical and hydrological conditions with each region.

  Not clear whether overenrichment was with regards to the natural levels of nutrients for given waterbodies or whether it meant that the nutrient loadings exceeded levels that were already elevated for recreation uses (e.g. increasing fish crops for recreation or commercial fishing). Also many nutrient poor systems have become mesotrophic, and are not considered overenriched because there are not noticeable indications, but they are elevated above their natural state. This distinction needs to be made, and could have a large impact on the identification of overenriched lakes.

  The term "region" was never defined. Does this refer to watersheds at the USGS 8 digit HUC code or ecoregions as defined by Jim Omernik (EPA) or Robert Bailey (USFS). Regardless, if regions are delineated by geologic and environmental factors that strongly determine the general natural trophic status of surface waters, then the regional approach will make sense at a broad level. Omernik uses land use to delineate regions of similar water quality, which does not explain the expectations of "natural" trophic conditions, but rather what exists now. Bailey's ecoregionalization seems to better delineate natural conditions. (TNC is presently testing this ecoregionalization.)

  Even within an ecoregion or a watershed, there can be a large range of natural nutrient levels of lakes and streams depending on the geological matrix in which they are imbedded, the sources of their water, their size and drainage network position. It would be very useful to apply a classification of lakes or streams to the framework that is proposed. Lakes and rivers within a "region" that should have the same nutrient levels can be identified and the least disturbed of each type can be used as reference conditions to provide target levels of nutrients. (TNC has developed and is applying a freshwater aquatic classification system according to climatic, geologic, size, drainage network position, source of water, hydrologic regime, chemical characteristics and other variables that can be sued for the assessment of natural levels of nutrients. I can provide information about the system if requested.)

EPA Response: EPA has decided to use Jim Omernik's ecoregional delineation for nutrients as a starting point for data collection and analysis in the overall effort of nutrient criteria development. EPA is open to the use of other regional approaches, such as Bailey's ecoregionalization, if they fit a particular region more appropriately. In addition, EPA believes that if data variability is too great within a specific EPA nutrient ecoregion, it is suggested that States/Tribes look to ecoregional levels III and IV to assess data variability and reference conditions, and criteria development.

Vladimir Novotny

• The approach described is brief and general. The watershed approach is feasible and most likely to succeed; however, the document is too vague to provide any reasonable guidance. I presume that some documentation and guidelines will be developed that will provide more specific guidance.

EPA Response: EPA will be developing more detailed, water body-type guidance documents, scheduled to be completed by 2000

The document appears not to follow established procedures for the Use Attainability Analysis and TMDL. The follow-up guideline documents should be conforming to these processes and procedures.

EPA Response: Agreed. The attainment of designated uses if a major focus when conducting a nutrient assessment and setting nutrient criteria and standards. It is necessary to match up a target nutrient level with a designated use in order to establish nutrient standards and set TMDLs.

Hassan Mirsajadi

• The most effective avenue for addressing nutrient overenrichment problems is for EPA to provide a national guidance with sufficient built-in flexibility for regional and watershed approaches.

EPA Response: Agreed. EPA Waterbody-type guidance documents will provide flexibility for regional and/or watershed approaches.

b) rational, practical and scientifically responsible (e.g., is the approach implementable by States and Tribes given their resources)?

William Keith, State of Arkansas

• Is appropriate and idealistically implementable; but will depend on the proper people being involved in development of the regional strategies (need state and tribe people who will be directly involved in implementation of the strategy; this would likely require a several year project due to time constraints of the appropriate people).

EPA Response: The Strategy has been re-written to describe the people who are needed to set criteria, analyze data and implement criteria (management and evaluation).

Douglas Knauer, State of Wisconsin

• Identification of the degree of overenrichment problems within states is possible, but successfully addressing the problems will depend on the priority and funds given it by the individual states.

Karol Erickson, State of Washington

• The states will implement as far as resources will allow. The approach fits in well with Washington State's watershed approach.

Michael Hirshfield, Chesapeake Bay Foundation

• The science is adequate. It's time to get on with it.

Tom Fontaine, SFWMD

• The strategy is rational, but the extent to which it is implementable depends on the available funds. There are many parameters that can be measured to detect nutrient enrichment and its effects. We need to choose those that have the greatest signal to noise ratio, since natural systems are inherently variable.

  Would rank the state of science sufficient for writing technical guidance as follows:

  Lakes>Streams & Rivers>>Estuaries>>Wetlands

EPA Response: Agreed. EPA is striving to increase both funds and people (EPA HQ and Regional) devoted to the nutrient effort.

Jonathan Higgins, The Nature Conservancy

• Given the appropriate level of guidance from documents and technical support and training from EPA and other agencies, I think the approach is reasonable. The most critical steps are the identification of target levels of nutrients, and the technical aspects of sampling, preservation and analysis. The procedures chosen for the monitoring must be simple and inexpensive in order for them to be used by a wide range of groups. Further review of this question is addressed in the 'specific comments'.

EPA response: These critical steps will be described in detail for each waterbody type in the waterbody-type guidance documents.

2) Is the state of the science sufficient to write technical guidance for nutrient assessment of each of the surface waters described in the Strategy (e.g., lakes and reservoirs; streams and rivers; estuarine and coastal marine waters; wetlands)?

William Keith, State of Arkansas

• Regardless, the need is to develop the necessary guidance using the best available science. Shortcomings in the technology will have to be addressed as effectively as possible in the short term until long term directions can be provided.

Douglas Knauer, State of Wisconsin

• A positive yes, for lakes and reservoirs (several EPA publications associated with the subject).

Jonathan Higgins, The Nature Conservancy

• I think that the science for lakes and rivers is pretty far along. I am not familiar enough with the level of knowledge of wetlands and estuaries.

Vladimir Novotny

• The state of science is adequately advanced to develop adequate technical guidance documents. To some degree the present application and interpretation of water quality criteria may be a hindrance. Based on the tradition of the point source waste load allocation, most WLAs have been done for low flow design conditions that represent hydrologic conditions of extreme and prolonged drought and exclude nonpoint sources. Several accepted water quality models used for calculating load capacity for nutrients are deterministic models that do require determination of some design period, usually which are unsuitable for design of controls for nonpoint, storm-driven sources. Using only deterministic/steady state models and focusing only on low flow design periods may be an obstacle to successful implementation of the nutrient control strategy.

EPA Response: EPA believes the development of both "causative" criteria such as nitrogen and phosphorus together with "biological indicator" criteria, such as algal biomass, Chlor a and secchi depth will help States and Tribes make a good first cut at controlling eutrophication. While the guidance documents do focus on the development of nutrient criteria, as described above, they will also present a number of parameters that can be used to assess and control eutrophication beyond the four just mentioned. This may be described as a "second generation" of nutrient criteria development, beyond the basic four described above. The setting of criteria is also looked upon as an amalgam of several factors which include: historical data, reference condition data, expert panel/peer review, empirical model analysis and an evaluation of downstream effects.

Hassan Mirsajadi

Although based on our current scientific knowledge, we do not have answers to all questions regarding nutrient overenrichment, this should not stop EPA from developing a technical guidance for nutrient overenrichment management. The guidance can take advantage of the existing wealth of knowledge for addressing well-understood problems. More complex issues regarding nutrient overenrichment can be addressed through future research.

2a) Are the assessment endpoints/indicators described in the Strategy reflective of the current science and the best available measurements?

William Keith, State of Arkansas

• They are appropriate. Most important guidance for these activities should be developed in the waterbody type and region specific guidance.

Douglas Knauer, State of Wisconsin

• Concerning "2. Chemical/Biomass Parameters"...
  • There is no discussion of depth profile. In stratified lakes/reservoirs, need to know the conditions in hypolimnion, and also would want an assessment of sediment phosphorus loadings to the hypolimnion. The sediments may also be a very important source of phosphorus loading in polymictic lakes that experience repeated anoxia of the bottom waters followed by complete mixing throughout the growing season.
    • Total dissolved Phosphorus should be measured; will give an assessment of sediment phosphorus loadings to water column.
    • Take DO measurements throughout water column, not just hypolimnion.
    • There is no mention of temperature. (Even though hypolimnetic DO was suggested as a parameter and determination of hypolimnetic boundaries are determined by temperature.)
  • There is no one method for measuring Total N (which is a combination of total Kjeldahl and the inorganic-N).

EPA Response: Many of these suggestions, which are considered important, will require further discussion during the process of developing the lakes and reservoirs technical guidance document. The issue of depth profile has been added to the Strategy.

Tom Fontaine, SFWMD

• Yes, except that sediment and phosphorus accretion rates in wetlands was left out.

EPA Response: EPA has added the language back in regarding accretion rates in wetlands

Jonathan Higgins, The Nature Conservancy

• The assessment of endpoints needs to be further developed. Application of classification systems and the identification and evaluation of reference sites need to be implemented within the different regions. There is still a need for further development of procedures for these. The identification of endpoints will be difficult. As I indicated under Question 1a, the endpoint selection needs to be clarified with regards to natural or recreational use. Many regions may not have enough information to develop endpoint ranges.

EPA Response: EPA believes the best place to make these determinations is in the technical guidance manuals, since as the commenter points out, endpoints will differ for each waterbody type. The determination of a what constitutes a reference site on a consistent basis is another issue that will need to be clarified on a national basis before regional teams proceed to data collection and analysis, and criteria development. When data are not present for a given region, EPA may choose not to develop a criterion for that region until acceptable data exist for that particular region. In most cases, a lack of data will trigger the initiation of data collection.

Vladimir Novotny

• The assessment/indicators reflect current science for most situations. THM precursors or THM formation potential could be added to the reservoirs and lakes used for water supply that receive minimal treatment.

EPA Response: This is an intriguing area with regard to nutrient control. While it is not clear what connection nitrogen and phosphorus directly play in THM production, it is obvious that highly productive waterbodies associated with a high level of organics in the water column would be prone to THM production. EPA will endeavor to determine linkages between nutrient levels and THM production. This issue may be discussed in more detail in the lakes and reservoirs technical document, depending on available information.

Hassan Mirsajadi

• The endpoints referred to in the draft strategy are a good starting point and can be expanded through a comprehensive literature search.

2b) Are the data storage and processing techniques (including models) described in the Strategy appropriate?

William Keith, State of Arkansas

• They are appropriate. Most important guidance for these activities should be developed in the waterbody type and region specific guidance.

Karol Erickson, State of Washington

• Agrees with need for improvements to existing models and agrees with the recommended improvements (streams section).

Tom Fontaine, SFWMD

• The general outline of data storage and processing techniques was adequate in concept, but lacked the details necessary to judge appropriateness. It is essential to document the quality of the data available to modelers and statisticians. (Comments on recent modeling developments for wetlands in Specific Comments section.)

Jonathan Higgins, The Nature Conservancy

• I am not familiar enough with the data storage and processing techniques described to five a meaningful answer. I would suggest applying the Index of Hydrologic Alteration (IHA) developed by Brian Richter, which measures the degree of alteration to natural hydrologic regimes of river systems, and can be used to identify the appropriate flow regimes that should be implemented in management plans for restoration of stream systems. (Can provide more information if requested.)

Hassan Mirsajadi

• The models referred to in the draft strategy are a good starting point and can be expanded through a comprehensive literature search.

2c) Are the management practices described in the Strategy appropriate?

William Keith, State of Arkansas

• They are appropriate. Most important guidance for these activities should be developed in the waterbody type and region specific guidance.

Douglas Knauer, State of Wisconsin

• "Trophic State Manipulation" should be eliminated--most of the options are attempts to alter the trophic state. The appropriate term that should be used is "Biomanipulation" and this could include "Fish stocking and removal".

EPA Response: This change has been made in the Strategy.

Karol Erickson, State of Washington

• This section could be greatly improved. There is a large body of information available on reducing nutrient contributions to streams. The methodologies should be targeted to the major nutrient sources identified in the monitoring and modeling elements of the project, which should identify whether point or nonpoint sources are significant contributors. For point sources, reductions in waste load allocations may be necessary. For nonpoint sources, there are a wide variety of management options, depending on the type of nonpoint source. Abatement methodologies specific to agriculture should be included here, and for urban/suburban areas. Education should be listed as one abatement methodology.

EPA Response: These suggestions will addressed in the waterbody type technical guidance documents

Jonathan Higgins, The Nature Conservancy

• I am not familiar enough with the management practices described to give a meaningful answer. I would suggest applying the Index of Hydrologic Alteration (IHA) developed by Brian Richter, which measures the degree of alteration to natural hydrologic regimes of river systems, and can be used to identify the appropriate flow regimes that should be implemented in management plans for restoration of stream systems. (Can provide more information if requested.)

Vladimir Novotny

• The management practices described in the document are not complete, the document itself specifies that it is not all inclusive. Practices not included are:

  Lakes and reservoirs

  • Relocation of sewage outfalls downstream
  • Restoration of upstream wetlands
  • Hypolimnetic aeration (for water supply reservoirs and lakes)
  • Point source nutrient removal (tertiary treatment, phosphorus precipitation)
  • Stormwater management (ponds, wetlands, infiltration)

  
  Streams and rivers

  • Removal of manmade impoundments that have lost their utility and are now places for excessive development of algae and water quality degradation
  • Restoration of riparian and flood plain wetlands
  • Point source nutrient removal (tertiary treatment, phosphorus precipitation)
  • Stormwater management (ponds, wetlands, infiltration)

EPA Response: EPA has added these suggestions to the Strategy.

• The strategy should consider load trading between point and nonpoint sources.

  I have some doubts whether nutrient management, to the same extent as for rivers, impoundments and estuaries, is needed for wetlands. Wetlands are usually nutrient deficient or can assimilate high loads of nutrients, especially nitrate-N. Many wetlands are actually naturally hypereutrophic. What should be the goal of nutrient management of wetlands? (Approaching the natural state of the wetland should be the management goal.)

EPA Response: EPA will address these issues when we focus our efforts on developing a wetlands technical guidance document, in 2000.

Hassan Mirsajadi

• The suggested BMPs referred to in the draft strategy are a good starting point and can be expanded through a comprehensive literature search.

ADDITIONAL COMMENTS AND CONCERNS

Dr. David Correll, Smithsonian Institute

• Attempt to subclassify streams, lakes and estuaries by dominant vegetation is a mistake--the dominant vegetation is usually the end result of how the system is reacting to human interventions.

EPA Response: Agreed. We are using this dichotomy only for the purposes of initiating comprehensive assessments of all factors - physical, chemical and biological.

William Keith, State of Arkansas

• Formatting: Rearrange sections V, VI, VII and VIII to organize them by waterbody type (Lakes and Reservoirs; Streams and Rivers; Estuaries and Coastal Marine Waters; Wetlands) with the major topics as subsections under each waterbody (Nutrient Measurement and Assessment; Data Storage and Processing; Basic Management Options; Research Needs).

  The strategy properly separates, discusses, and chooses to address the regional strategy by waterbody type. However, it needs to recognize and address the fact that conditions exist where one waterbody type affects another and the objectives met by one type may not be appropriate for the different type (or different segment) downstream. For example, management goals and activities suitable for rivers and streams may not be sufficient to protect downstream lakes and reservoirs. (Similar situations between streams and estuaries, coastal waters, wetlands and even between segments of the same stream that may be periphyton-dominated changing to plankton-dominated.)

EPA Response: Agreed. EPA ideally would like to see watersheds examined as a whole when it comes time to develop criteria and set regulatory measures. However, in our effort to develop technical guidance, we will start where we know the most information exists, namely lakes and streams, and move forward in developing guidance for estuaries, coastal waters and wetlands. By 2000, EPA expects all guidance, except for wetlands, to be completed. When guidance for wetlands is completed in 2001, EPA expects States and Tribes to use the collective guidance to develop criteria (unless they choose to use EPA defaults) on an ecoregional basis and then to apply these criteria on a watershed basis. EPA has also modified the strategy to discuss the issue of downstream effects and how it should be factored into criteria development.

Tom Fontaine, SFWMD

• There is a lack of standard entries between the waterbody types. In some cases, "early indicators" are noted, and in others there are none mentioned. In some cases, a characterization of the system (e.g. seagrass dominated) is noted, but in other cases, no such characterization is made. I would suggest that a standard approach be adopted to eliminate confusion.

SPECIFIC COMMENTS

EPA Response: Many of the specific comments were added to the Strategy or were considered so specific to a waterbody type that they were added to one of the waterbody technical guidance documents.

I. INTRODUCTION AND BACKGROUND

• Michael Hirshfield, Chesapeake Bay Foundation

  • Page 1, sentence 1--This sentence makes no sense. It seems to be trying to excuse nutrients as somehow different from toxics; if the point is that nutrients at some level are necessary, fine; but many heavy metals that are toxic at high levels are fine at some level (and in some cases, necessary). Anthropogenic releases of nutrients are not essential to the ecosystems. The sentence could be rewritten to read "In contrast to man-made compounds never found in nature unless we discharge them, having some nutrients in the water is essential..." But, I'm not sure of the point.

  
  Tom Fontaine, SFWMD

  • Page 1, paragraph 2, sentence 1--Is there a uniform definition of "assimilative capacity" for all of the systems? Is there agreement among scientists that a system becomes "hypereutrophic" immediately following exceedence of the assimilative capacity? Shouldn't exceedence of assimilative capacity lead to a progression towards hypereutrophy?

  
  Robert Wetzel, University of Alabama

  • Page 1, paragraph 2, line 5--Should read "...nutrients phosphorus and nitrogen." On a global basis and in the U.S., over 70%, likely over 85%, of surface waters are limited first by P, secondarily by N. (More detailed explanation in comments received.)

    Page 1, paragraph 4--The discussion of N:P ratios needs modification. A relative constancy in the molar ratio of C:N:P of 106:16:1 among marine plankton, termed the Redfield ratio, is generally supported by numerous studies. Despite some variance (usually less than 20%), the constancy has been attributed to relatively nutrient-deficient growth conditions of marine plankton (J.C. Goldman et al., 1979).

    In contrast, marked deviations in the C:N:P proportions of the Redfield ratio occur in the seston of lakes, coupled to physiological indicators such as rates of growth and productivity (Kilham 1990; Sommer 1990). The C:P and N:P ratios of lake seston are generally higher than the Redfield ration for marine waters. P limitation tends to be much greater than N limitation to algal productivity in small lakes. Streams, shallow lakes, and reservoirs with short residence times have C:P ratios <350 and N:P ratios<26.

    Spatial and temporal differences in cellular stoichiometry are large. Epilimnetic ratios, particularly of N:P, are higher, indicative of high P demands in proportion to total inputs, than those of the hypolimnion, and N:P ratios decrease with depth. (Dr. Wetzel attached to his comments a table titled "Stoichiometric ratios of phytoplankton-dominated seston of lakes as indicators of relative nutrient limitations".)

  Tom Fontaine, SFWMD

  • Page 1, paragraph 4, sentence 2--My understanding is that the Redfield ratio was established in saltwater systems, so you should point out that your example is illustrative, and not necessarily applicable to all systems.

  
  Karol Erickson, State of Washington

  • Page 1, paragraph 4, last sentence--Suggest ratio be expressed in terms of grams (e.g., 10:1 on mass scale).

  
  A. KEY FACTORS

  • Karol Erickson, State of Washington

    • Page 1, paragraph 5--Not sure what this paragraph tells us. Seems to be a mish-mash of terms.

    
    Tom Fontaine, SFWMD

    • Page 1, paragraph 5, sentence 1--I think you should speak of causes of eutrophication and consequences, instead of elements. As stated, sources and reactions to those sources are mixed. Also, don't forget to include internal sediment loading sources, which actually reflect prior loading from external sources.

      Page 2, paragraph 1, last sentence--The 1994 report referenced surveyed rivers and streams, yet mention is made in this sentence of lakes and estuaries. Were these systems also surveyed? What about wetlands?

      Page 2, paragraph 2, sentence 2--Is 503(d) supposed to be 303(d)? (also Karol Erickson, State of Washington)

      Page 2, paragraph 2, sentence 3--My understanding of the Clean Water Act is that agricultural runoff is excluded from consideration. Isn't that ironic when "Agriculture is the most widespread source..." While the strategies for prevention of nutrient enrichment will identify the major sources, it would be desirable if EPA could play a role in the regulation of this most widespread source.

      Page 2, last sentence of section--change ";" to "," and hyphenate chemical-specific.

    Michael Hirshfield, Chesapeake Bay Foundation

    • Page 2, last sentence of section--What is meant by "...address a specific element of pollution often manifested in the existing criteria"? (Also, Tom Fontaine, SFWMD)

II. THE EMERGING NUTRIENT STRATEGY

• Karol Erickson, State of Washington

  • Page 3, paragraph 1, sentence 2--Reductions in DO result also from algal respiration; and pH can be increased due to algal photosynthesis.

III. KEY OBJECTIVES OF THE NUTRIENT STRATEGY

IV. REGIONAL AND WATERBODY SPECIFIC APPROACH

• Karol Erickson, State of Washington

  • Page 5, paragraph 1, sentence 1--Take out the word "Essentially".

  
  Jonathan Higgins, The Nature Conservancy

  • Page 5.--Need a description of the effects of overenrichment on elements of biodiversity. Changes in nutrient levels will change community structure. More natural levels may bring about more appropriate habitat for some uncommon or threatened species, conservation status for the bodies of water, and conservation tourism for an area.

  
  Karol Erickson, State of Washington

  • Page 5, Step 1--Sentence should read "Make sure problem exists and is clearly defined in terms..."

  
  Douglas Knauer, State of Wisconsin

  • Page 5, Step 2--Suggest employing paleolimnology. Most often the changes in water quality parameters can be directly associated with the background investigation of landuse changes.

  Karol Erickson, State of Washington

  • Page 5, Step 3, sentence 1--Change "loading sources" to "parameters". Parameters should include, for example, algal measurements, streamflow, light/substrate, depth, and velocity.

    Page 5, Step 3, sentence 2--Should read "....to accommodate seasonal and year-to-year variation."

  Jonathan Higgins, The Nature Conservancy

  • Page 6, Step 6. Economic Evaluation.--Also must evaluate the impact to present uses of waterbodies, such as commercial and recreational fishing.

  
  Karol Erickson, State of Washington

  • Page 6, Steps 7 and 8-- "Implementation of Controls" should be inserted as a step between steps 7 and 8.

V. TECHNICAL GUIDANCE MANUALS FOR NUTRIENT MEASUREMENT AND ASSESSMENT

Jonathan Higgins, The Nature Conservancy

• Page 7, paragraph 1, line 3.--Change sentence to read "This is an important point because overenrichment and natural levels and the impacts of overenrichment to aquatic ecosystems differ from one geographic area to another."

  Note: surveys that include variables that include diel variations should be conducted at the same general time of day. The objective is not to measure diel variability.

• A. LAKES AND RESERVOIRS

  Douglas Knauer, State of Wisconsin

  • Need to incorporate a landscape approach to lake assessments. Need actual demonstration projects to investigate new in-lake management tools where water quality is set by ecoregion parameters.

  
  Jonathan Higgins, The Nature Conservancy

  • Page 7, paragraph 4 (paragraph 1 of Lakes and Reservoirs section), line 3.--Change sentence to read " An element of this guidance will be to equate lake nutrient environmental impacts and public health, recreational uses and biodiversity issues and concerns . . ."

  
  Tom Fontaine, SFWMD

  • Page 7, paragraph 5, line 2--There are many locations in the US where year round sampling is possible. Therefore, I'm not sure we need to specify that data gathering will be indexed (what does indexed mean?) to summer surveys. As nutrient criteria will be developed regionally or on a site-specific basis, it seems that we should not bias the reader towards summer surveys.

  
  Douglas Knauer, State of Wisconsin

  • Page 8, #1--"Early Warning Indicators" should be re-titled "Watershed Indicators".

  
  Jonathan Higgins, The Nature Conservancy

  • Page 8, #1--should be "Changes in hydrologic regimes in watershed"

  
  Tom Fontaine, SFWMD

  • Page 8, #1--"Early warning indicators" of over-enrichment needs to be defined on the basis of something measured in the waterbody. By the time we can map the changes in the watershed or do the land use/loading assessments, we may have completely missed the early warning indicator. This is not to say that we shouldn't do the load assessments, etc., however. They are important in establishing a record against which future loads are judged.

    Page 8, #2--For chemical/biomass parameters, sampling needs to be done either in spring before thermal stratification, or the sampling regime needs to include the epilimnion and the hypolimnion, since the concentration of P & N are dependent on stratification levels. For Nitrogen, Total N or Nitrate/Nitrite are fine. For Chlorophyll, use Chlorophyll a; if different measurements are used, they are incomparable.

    Page 8, #3--For community structure parameters, I would include biomass for all.

    Page 8, #4--For the secondary parameters, sample different strata.

  B. STREAMS AND RIVERS

  Tom Fontaine, SFWMD

  • Page 8, paragraph 1 (of this section), sentence 2--Does "nutrients will saturate the biomass" mean that the periphyton has reached its saturation point (maximum uptake rate), with respect to nutrients?

  
  Robert Wetzel, University of Alabama

  • Page 8, Column 2, bullet 2-- "Black disk" is not an acceptable technique.

  
  Tom Fontaine, SFWMD

  • Page 9, Column 2, bullet 3-- "Sediment Composition"---with regard to what? P, N, invertebrates, etc?

  
  Robert Wetzel, University of Alabama

  • Page 9, Column 1, bullet 3--The ratio of AFDW/Chlor a is not an "autotrophic index" (whatever that means.)

  
  Tom Fontaine, SFWMD

  • Page 9, Column 1, bullets 9 & 10--duplicate "Biointegrity" entries.

  
  Karol Erickson, State of Washington

  • Page 9, Endpoint list appropriate for either--Is diel DO (bullet 2 in column 1) the same thing as DO amplitude (bullet 6 in column 2). Add "Hydrology: streamflow, velocity, depth" to this list.

  
  Jonathan Higgins, The Nature Conservancy

  • Page 8 and 9, list of endpoints--For plankton dominated systems, SRP will be an important nutrient for plankton. For periphyton dominated systems, how will benthic community metabolism be measured, and why is it an important parameter? For both types of systems, biointegrity should appear as an indicator. It is duplicated under the listed appropriate for either system.

  
  C. ESTUARIES AND COASTAL MARINE WATERS

  Robert Wetzel, University of Alabama

  • Page 10, bullet 5--Is this supposed to be "quantum irradiance levels"?

  
  D. WETLANDS

  Tom Fontaine, SFWMD

  • The importance of establishing reference wetlands for comparison purposes cannot be overemphasized. Other systems are way ahead in this regard. We must also establish internal references for individual wetlands so we can assess changes over periods of time. Certain flora, such as periphyton, serves this purpose well. But in addition, we need to establish a cheap, easy way to determine whether individual wetlands are accreting sediments and associated nutrients as a function of nutrient loading. We had extensive discussions on the need for measuring sediment and nutrient accretion in wetlands at the 1995 workshop, and I am very surprised to find no mention of it in this document.

  
  Robert Wetzel, University of Alabama

  • This section on wetlands is very weak. The key actors in the ecosystems are the attached microbial community, which reflects the nutrient absorptive and recycling capacities, and the macrophytes for long-term storage and interment of nutrients. Planktonic components (but not floating periphyton) are nearly always trivial in both mass and productivity.

  
  Jonathan Higgins, The Nature Conservancy

  • Page 11, #3b.--Does "macrophytes" include emergent vegetation?

    Page 11, discussion section--Why are reference wetlands discussed, but reference lakes and river sections are not mentioned in their section. I think this is very important for all ecosystem types.

  Tom Fontaine, SFWMD

  • Page 11, last paragraph--I am surprised that the work done by Kadlec for EPA on the North American Wetland Database is not mentioned here. He has attempted a number of classifications based on retention time and nutrient loads.

VI. DATA STORAGE AND PROCESSING

Tom Fontaine, SFWMD

• Page 12, paragraph 2--It is essential that the quality of the data entered in these databases be carefully documented. This would include information on methods used, minimum detection limits, comparison with standards, etc. Modelers should use due caution if QA aspects of the data are not available.

• A. LAKES AND RESERVOIRS

  Robert Wetzel, University of Alabama

  • Modeling is a tool, not a panacea. The only thing that modeling does is consolidate and narrow potential probability of relationships. With only mass (chemical or biotic) data from these surveys, one can only hope for very general relationships because these ecosystems function as composites of highly dynamic, non-linear, metabolic rate functions. Something like the Vollenweider-Reckhow models for phytoplankton densities vs. nutrients are of course useful, but phytoplankton do not constitute a lake or reservoir.

  
  Tom Fontaine, SFWMD

  • Page 12, paragraph 3, sentence 2--References should be given for the BATHTUB and Reckhow-Simpson technique.

  
  B. STREAMS AND RIVERS

  C. ESTUARIES AND COASTAL MARINE WATERS

  D. WETLANDS

  Robert Wetzel, University of Alabama

  • Mitsch and Gosselink (1993) and Howard-Williams (1985) provided only conceptual diagrams of potential relationships. These discussions are not models.

  
  Tom Fontaine, SFWMD

  • Page 14, last paragraph--Recent work by Walker, Walker and Kadlec, Moustafa, etc. should now be referenced. (Email me for citations as I am out of the office and don't have them with me.)

  
  Robert Wetzel, University of Alabama

  • Page 15, paragraph 1, line 1--Why only for wetlands? The same statements apply to lakes and rivers.

    Use of "models" here is incorrect and is a marked violation of what models are. The relationships imply functional interactions, and one is really evaluating quantitative fluxes among dynamic biotic and non-living compartments.

    If the objective of this entire exercise is to be able to give "managers" predictive models that will let them "manage" environmental units with a few data points from their system, it is certainly destined to fail. One needs diagnostic information to evaluate the physiology of these ecosystems to determine (predict) probable changes. The question is really how little information is needed to predict tolerable changes. It is critical that EPA does not promise predictive (managerial) capacities from token data of simple parameters that cannot possibly provide reasonable probabilities of being any more accurate than a limnologist standing on the shore and guessing.

  Robert Wetzel, University of Alabama

  • Page 15, paragraph 1, last sentence--Concerned with establishment of management criteria and 'models' based on simplistic data that are non-verified from a spectrum of reference ecosystems. The literature data are not satisfactory for many ecosystems.

VII. BASIC MANAGEMENT OPTIONS

• Michael Hirshfield, Chesapeake Bay Foundation

  • What is this section meant to accomplish? It doesn't add anything a manager wouldn't already know.

  
  Jonathan Higgins, The Nature Conservancy

  • What about second home development and septic tank field regulations and improvements?

  
  Robert Wetzel, University of Alabama

  • The management options are generally good, assuming that eventually specific recommendations will be made.

  
  A. LAKES AND RESERVOIRS

  Robert Wetzel, University of Alabama

  • Page 16, last bullet--Trophic state manipulation (=biomanipulation) is, at best, a cosmetic, short-term method that redistributes problems, but does not solve the problems. Rarely, if ever, does predator manipulation "control algae".

  
  B. STREAMS AND RIVERS

  Jonathan Higgins, The Nature Conservancy

  • Page 17, 8th bullet--For hydrology, flow regime, etc, apply Index of Hydrologic Alteration t identify natural hydrologic regimes, and use this information to work with Dam operators to better replicate natural conditions as best they can without losing electric power generation or reservoir levels.

  C. ESTUARIES AND COASTAL MARINE WATERS

  D. WETLANDS

VIII. RESEARCH NEEDS

• A. LAKES AND RESERVOIRS

  Douglas Knauer, State of Washington

  • This set of research needs is ridiculous. We need to incorporate a landscape approach to lake assessments (e.g., document how forest management practices impact lake water quality) and conduct demonstration projects, such as the TROLS project in northern Alberta (Dr. Ellie Prepas). We need actual demonstration projects to investigate new in-lake management tools where water quality is set by ecoregion parameters (e.g., elevated phosphorus in the groundwater as a result of geological conditions).

  
  Jonathan Higgins, The Nature Conservancy

  • We need to have a better understanding of cascading trophic interactions; the effects of manipulations or nutrient changes on one level of the food chain, and how these changes then affect the rest of the community through the network of interactions that exist.

  
  Robert Wetzel, University of Alabama

  • Page 19, bullet 4--Trophic state manipulation (=biomanipulation) is, at best, a cosmetic, short-term method that redistributes problems, but does not solve the problems.

  
  B. STREAMS AND RIVERS

  Jonathan Higgins, The Nature Conservancy

  • We need to have a better understanding of cascading trophic interactions; the effects of manipulations or nutrient changes on one level of the food chain, and how these changes then affect the rest of the community through the network of interactions that exist.

    Page 19, 1st bullet--for Chlorophyll sampling methods see USGS NAWQA methods.

  C. ESTUARIES AND COASTAL MARINE WATERS

  D. WETLANDS

IX. NATIONAL COORDINATION OF REGIONAL NUTRIENT INITIATIVES

X. SUMMARY STATEMENT

APPENDIX A: SUMMARY OF WATER QUALITY CRITERIA AND STANDARDS FOR NUTRIENT ENRICHMENT

Michael Hirshfield, Chesapeake Bay Foundation

• This could be extremely useful with a more detailed review of how states are currently dealing with N and P standards (including actual numbers).

  Page 26, paragraph 2, last sentence--A laughable concern about "overcontrol" of nutrients. I defy anyone to find a location where overcontrol of nutrients is even remotely likely, especially given that the biologically desirable outcome is no additional anthropogenic nutrients.

APPENDIX B. DRAFT OUTLINE FOR THE DEVELOPMENT OF NUTRIENT ENDPOINT CRITERIA FOR RIVERS, LAKES, RESERVOIRS, ESTUARINE AND COASTAL SYSTEMS