Clean Water Program, continued

March 4, 1998

Marilyn Fonseca
Northwest Region
Water Quality Division
Oregon Department of
Environmental Quality
2020 S.W. 4th Ave., Suite 400
Portland, Oregon 97201-4987

Re: Draft TMDLs for Dissolved Oxygen, Eutrophication (pH and Nutrients), Bacteria, and Toxics in the Columbia Slough Basin.

Dear Marilyn:

Northwest Environmental Advocates is pleased to be able to comment on the draft TMDLs for the Columbia Slough. It is certainly accurate to say that they have been a long time in coming. Unfortunately, the TMDLs illustrate that even the passage of time has not resulted in sufficient data upon which to base many decisions. While NWEA is pleased to see quantitative calculations for load allocations and waste load allocations within these TMDLs, most of these allocations are to categories of sources and the documents get mushy in showing how these allocations will be tied to allocations for specific sources and implemented. While NWEA greatly appreciates that these TMDLs contain information about implementation of allocations, including follow-up monitoring and responsibilities of parties, they stop short of what is sufficient to be an implementation plan, as recognized in the summary information which discusses the implementation of a "water quality management plan (WQMP)." There is no clear information provided about DEQ's preparation of a WQMP, such as when it will be produced and whether the public or EPA will be asked to review it and how it will or will not differ from the sum of Memoranda of Agreement (MOAs) and other regulatory and voluntary actions mentioned throughout the documents. Moreover, there are numerous references to monitoring plans and monitoring that will take place prior to the development of Phase II TMDLs, but there is no indication of whether this information will be available to the public and what assurances the public or EPA should have that the information will be used to meet the TMDLs' allocations prior to the development of phase II TMDLs. In short, much is left to the imagination.

I. GENERAL COMMENTS APPLICABLE TO ALL PARAMETERS

A. Missing Information

There are some other basic pieces of information that should be included in the TMDLs. For example, as the documents address multiple parameters, presumably they include more than one TMDL, yet the document refers to a single TMDL. It is also not clear what seasons are covered by the TMDLs, despite the clear statutory requirement that TMDLs establish loads to meet water quality standards "with seasonal variations." While there is no indication on the face of the documents that the TMDL for dissolved oxygen is limited to winter months (not defined by the TMDL), the text only establishes loading capacity and allocations for those months. Either the waterbody is not limited for that parameter in other seasons and no TMDL is required or this TMDL addresses the waterbody for a limited period. The Department cannot mention the other seasons, not establish a TMDL that will not lead to attainment in those seasons and still credit itself with having completed the job.

B. Phased TMDLs

The draft TMDLs make numerous references to their being Phase I TMDLs and that subsequent phases will be done. Additionally, the TMDLs make numerous references to data that will be collected and analysis that will be done, presumably in Phase II. However, nowhere in the draft TMDLs are there any reference to the timeframe in which Phase II will be done, from the development of a work plan to its completion and submittal to EPA, nor any analysis of the sum total of the data gathering elements scattered throughout the document will be a sufficient basis upon which to prepare the Phase II TMDLs in a timely fashion. Not only does the Department have a poor track record on conducting Phase II TMDLs, but an unacceptable period of time has elapsed for the production of these draft TMDLs particularly considering its serious impacts to beneficial uses. This record supports what would otherwise be sound policy of establishing Phase II elements, such as a clear plan and timeframe, as critical. Without these elements for Phase II, these draft TMDLs fail on their face to clearly establish that they will lead to attainment of water quality standards.

As one of very few TMDLs developed by Oregon, this TMDL sets the stage for those to come. Unfortunately, despite the years of effort, many aspects of these TMDLs are hampered by the lack of data and modelling, and a concurrent reluctance to unambiguously require conservative controls. The result is, in many instances, that the TMDLs do not demonstrate that they will lead to attainment of standards, that there is a long a varied set of references to Best Management Practices (BMPs) that may or may not be instituted in a set of unenforceable agreements, and many similar references to further collection of data. Monitoring information and vague commitments to vague controls do not add up to a TMDL. Despite the need for TMDLs for the Slough to be completed and implemented, if the TMDL is inadequate after this long period of development, it will serve no purpose to the environment and public health to finalize it.

C. Identification of Beneficial Use Support

The draft TMDLs should identify more specifically the beneficial uses that are affected both for the purpose of illustrating the benefits of implementing the TMDLs to the public as well as for accurate application of the narrative criteria and beneficial use support aspects of the water quality standards. For example, the TMDLs should cite to information about the use of the Slough by salmonid populations. The draft TMDLs fail to fully evaluate uses that may be more sensitive than those used to derive numeric criteria or guidance values. Specifically, the TMDLs do not identify the use of the Slough for subsistence level fishing that includes the consumption of fish/shellfish at higher than average levels, consumption of species that have higher than average levels of contamination due to lipid content, and consumption of fish in "non-traditional" fashion, i.e. full body rather than fillets. By failing to recognize existing information about the actual existing uses in the Slough, the draft TMDLs fail to adequately protect the most sensitive beneficial uses. Moreover, since the consumers of these fish are generally low income, recent immigrants and ethnic groups, this analysis is necessary to address the issue of environmental justice, which the draft TMDLs fail to do.

D. Source Information

Throughout the TMDLs, from the "WQ Concerns at a Glance" on the first page through the analysis, the draft TMDLs are a patchwork of information on sources. For example, "groundwater" is listed as a "known source" but groundwater is a conduit not a source. NWEA does not object to quantifying the load contributions related to this conduit, but the document obscures the actual anthropogenic sources that are causing impairment in the Slough. There is no consistency from this summary to the TMDLs, from tables to text, and from parameter to parameter, in clearly identifying all conduits and all sources. For example, under known sources there are many obvious sources missing, including direct runoff from lands adjacent to the Slough, remediated and contaminated hazardous waste sites, and air deposition. Not only would the TMDLs be dramatically improved if sources were fully discussed (e.g. there are multiple types of "stormwater" including those that drain from municipal and industrial [pipes as well as direct runoff from land adjacent the water), but a complete list should be used consistently throughout the TMDLs.

E. Ambiguous Findings

When finalized, the draft TMDLs should contain unambiguous findings of the Department, regardless of whether they are based on technical information or professional assumptions. For example, there are several places where DEQ notes that something "may" be true, e.g. the reasons for summer violations of DO criteria. Draft TMDL at 9.

F. Margins of Safety

Discussions of sources relate to margins of safety (MOS) in the draft TMDLs. It is impossible to evaluate the sufficiency of the chosen MOS without information about the unknown sources of the particular parameter, at the least, as well as information on the conservatism of underlying assumptions and the likelihood of sufficient controls. For example, the 20 percent of total BOD5 load that is allocated as an MOS.

Time Frames for Attainment of Loading Capacity and Allocations

The TMDLs should establish the timeframes for attainment..... There are several references to how the TMDLs will be implemented through water quality management plans developed as part of the Lower Willamette Subbasin plan. While there is some suggestion that these are agricultural management plans pursuant to Senate Bill 1010, it is not clear what the scope of these plans are, under what legal authority they will be prepared, whether they are enforceable, and in what time frame they will be prepared and implemented.

The first section of each TMDL should be standards not criteria....

II. PARAMETER-SPECIFIC TMDLS

A. Dissolved Oxygen TMDL

The TMDL discussion and analysis of allocations should include an analysis of the certainty of installment of sufficient pollution controls necessary to attain the allocations. Designing allocations solely on the basis of current relative contribution of loads is not relevant to ensuring that the TMDL is successfully in a timely fashion. There is, however, no discussion of what technical or regulatory difficulties might
be encountered in enforcing the allocations on the different sources. For example, while there is discussion of the Port of Portland's airport NPDES permit, there is no corresponding discussion related to the allocation to the Air National Guard (ANG). There is simply no information to even judge whether the ANG allocation will be implemented through an NPDES permit or other mechanism, casting serious doubt on whether the TMDL will lead to attainment.

How the allocation for future growth., which appears to be derived from the 150 percent increase in storm water pollutant loading predicted by the City of Portland, will be applied is unclear. The TMDL simply states that new discharges "will either have to demonstrate that adequate reserve capacity exists in the TMDL or trade effluent with [another source]." Draft TMDL at 17. The TMDL should clearly state on its face that until allocations are met by other sources, no new or increased loads will be allowed and establish how it will be determined that allocations have been met. It is not clear how a party would make this demonstration but it is implied that the reserve capacity would be found in the TMDL rather than in the waterbody itself, suggesting that the TMDL is merely guidance rather than binding. It is inappropriate for the TMDL to suggest that effluent trading is an acceptable approach when, at the same time, the draft TMDL stops short of establishing effluent limits in stormwater permits and only recommends the establishment of BMPs.

As with other of the draft TMDLs, the TMDL for DO suggests that much of what should be control strategies are actually monitoring for another phase of the TMDL process. For example, the draft document states that the Port "will be required to conduct additional instream and source load monitoring and modeling to refine estimates of the rates and loads used to generate the LC and identify what allocations they can achieve." Draft TMDL at 15. More specifically, item four (4) in the list of TMDL "requirements" calls for the Port to "identify the final TMDL BOD allocations to be achieved with implementation of controls." Draft TMDL at 15. This TMDL must establish what the allocations are for implementation or it is not a TMDL. If the Department does not believe that these allocations are attainable, it must make allocations that are. Absent that belief, and implying that they are not attainable, the TMDL does not on its face lead to attainment of standards and therefore will not be approvable by EPA. If the TMDL is little more than a series of future monitoring actions, it will not lead to attainment and also falls short of the definition of a TMDL. Otherwise, it is not a TMDL; it is a work plan to develop a TMDL. Everything in this section suggests that not only are the alleged allocations not fully enforceable but that they are not real and will not be real until the "winter of 1998/1999" when the Port, not the Department, makes new allocations. While nothing in this or any other TMDL prevents the Department from promulgating a new TMDL at some future time that supersedes any TMDL, stating in advance that the allocations are entirely subject to change and that they will not be determined by the Department prevents this from being a TMDL.

There is no basis for delay in installing the pollution controls necessary to implement a TMDL. This TMDL, however, proposes that non structural BMPs (for an unknown allocation) need not be implemented until "winter 2000/2001" with the "identification and implementation" of pilot structural BMPs not required until May of 2002. (This sentence leaves considerable ambiguity about when implementation of both pilot and final structural BMPs would have to be in place.) Moreover, nowhere in this discussion of implementation strategy is there a statement that these elements will be contained in a fully enforceable schedule within an NPDES permit. If this is a TMDL, implementation of necessary controls to meet allocations established by the Department in this TMDL must begin immediately.

The implementation strategy for the draft TMDL states that WLA for the DMAs will "not be included as an effluent limit" in MS4 permits. Draft TMDL at 16. DMAs will be required through non-enforceable MOAs, rather than their MS4 permits, to describe and implement BMPs and monitoring. In other words, the Department is forgoing all opportunities to make the elements of this TMDL binding upon the regulated sources, thereby undermining the likelihood that the TMDL will lead to attainment of water quality standards.

There is nothing in the description of the monitoring and BMP process that even suggests a timeframe in which monitoring and evaluation of BMP effectiveness will be completed nor a timeframe and process in which BMPs will be improved based upon that analysis.

The draft TMDL's reference to the development of a basin-specific general storm water permit similarly suggests unacceptable delays. The TMDL states that this will be done "[w]hen storm water permits are renewed." Draft TMDL at 16. This could easily be at least another five years. And, when it is completed, the draft TMDL states that it will not incorporate the WLA. Id.

B. Eutrophication (pH and Nutrients) TMDL

For the sake of clarity, the Department should state where it is interpreting and applying narrative criteria, such as where it uses "interim targets" for total phosphorus. Draft TMDL at 17. It wold also help the reader to explain what it means for a criterion to be an "action level," as opposed to normal numeric criteria.

While it is practical to consider alternation of flows as a solution to the algal growth impairment of the Slough, the draft TMDL fails to both make appropriate allocations to pollution loads and to establish flow targets or allocations with a mechanism to ensure that those flows will, in fact, be met. Instead, the TMDL merely discusses the benefits of controlling residence time in the Slough. In fact, the TMDL states that increased macrophyte growth caused by the increased flow will present flood control management problems for the Multnomah County Drainage District (MCDD), suggesting that flow management may not be a successful approach to attainment of standards. Id. at 23. Instead, the TMDL must establish what flows are required to lead to attainment, given the pollution loads evaluated and allocated and establish that there is a mechanism to ensure that those flows are maintained. NWEA believes that establishing mechanisms to ensure flows are met, when they comprise the basis of the TMDL calculations, is a necessary part of a TMDL, here and elsewhere. Where load reductions are not even contemplated, there is no other alternative or the TMDL will not clearly lead to attainment of water quality standards.

However, this is not a TMDL to address flow issues; there are also pollution loads to take into account. If the Department believes that loads should be controlled for some sources, i.e. new point source discharges and Oregon Fresh Foods NPDES, there is little rationale for not considering controls on existing loads. Instead the draft TMDL establishes all current loadings as allocations. Regardless of the finding that flow management is the most effective way to reduce algal growth and its influence on pH, the TMDL should also result in relevant pollutant load reductions where applicable.

The draft TMDL discusses the possibility that summer and winter storm water "may contribute phosphate in particulate form" which

Similar to the seasonal problem with the draft DO TMDL, this TMDL fails to model the impact of summer storm water loads of ortho-phosphate. Draft TMDL at 20. Therefore, the TMDL should clearly state that it only applies to winter until such time as this modeling takes place. While the alternative appears to be to presume that the summer allocations to storm water presented in Table 5 are accurate, the text of the TMDL clearly indicates that the Department does not necessarily believe that this TMDL will lead to attainment in the summer months. If, after years of effort that lead to development of a TMDL that is not sufficient, the Department should not prepare to submit it to EPA but rather should gather the necessary data and do it correctly. The "phased approach" to TMDL development is not carte blanche to submit what happens to be on hand.

Tables 5 and 6 should include a listing of existing loads and existing flows with the allocation of flows necessary for attainment. At the very least, it would provide clarity to the reader. While we clearly read this draft TMDL to not call for reductions in summer loadings, there are mixed messages in the text with regard to winter loadings. Currently there are no estimates of necessary load reductions but merely a small section stating that BMPs to control phosphate loadings will be contained in MOAs with DMAs. Draft TMDL at 23. How can the quantified analysis presented in this TMDL be used to appropriately set the BMPs required to lead to attainment, without any sense, even of a percent load reduction needed, of the goal of those BMPs? Why will it be sufficient for DMAs to identify BMPs "which may reduce contributions of phosphates via storm water" without a requirement that they "will" reduce loadings or that they must be implemented rather than just identified? It is unclear whether this TMDL is making a finding that loads need to be reduced because the text states that the TMDL "establishes the existing load as the LC." Draft TMDL at 22. Read broadly, this TMDL seems to contemplate no reductions in the status quo and therefore will not lead to attainment of water quality standards.
Where it is expected that the time frames for attainment of necessary load reductions will vary, they should be clearly set out, preferably in table form. For example, a load reduction of 40 percent of nitrogen in groundwater is expected over a 30 year period. Other allocations should have time frames attached to them.

reiterated permit problem page 23

There are difficulties in correlating the text and the tables found elsewhere in the documents. Oregon Fresh Foods is given an allocation of 0.56 mg/L of total phosphate in the text and a "point source" is given 73 kg as a winter ortho-phosphate waste load allocation in Table 6. While NWEA is pleased to read that a waste load allocation is actually going to be placed into an NPDES permit, it is odd that the evaluation of what the WLA should be falls under the implementation strategy section rather than that for allocations. More important, it appears that the allocation is a standard localized dilution analysis rather than an analysis of an appropriate portion of the total loading capacity for this specific source. In other words, this analysis could be done in the absence of a TMDL the purpose of which is to go beyond dilution analyses and consider the cumulative impact of multiple sources. There is no technical justification for setting an allocation that allows a pollutant to be discharged at a level that is both the current level and a level at which the water quality standards are violated.

C. Bacteria TMDL
We read the text of this draft TMDL to say that although CSO loads will be virtually removed in a few years, insufficient information exists to determine whether that action will be sufficient to results in attainment of the bacteria standard. In other words, this draft TMDL is not a TMDL but one may be needed in the future. This general perspective is underscored by the TMDL's notation that insufficient data exist to establish a TMDL that covers the entire waterbody. E.g., Id. at 26, 28. It is also made clear by that fact that, while CSOs and other sources of raw sewage are in essence given allocations of zero, there are no allocations to any other potential sources of bacteria. Id at 29. The draft TMDL actually gives CSOs an allocation of their "current load" until the year 2002. Id. at 28. This is absurd on its face. A TMDL that makes an allocation of a huge volume of raw sewage that causes immediate violations of water quality standards on a regular basis cannot be read to lead to attainment. The proper allocation is the control level that has been approved by the Environmental Quality Commission (three in ten year discharge). The fact that there is a compliance schedule to achieve this allocation does not alter the appropriate allocation.

References to removal of Willamette River CSOs in this draft TMDL should be removed as there is no qualitative or quantitative analysis that connects these statements with attainment of allocations or standards in the Columbia Slough. If the intent of these statements is to imply that allocations for these loads to the Slough will be attained in a longer time frame than other allocations, a table with allocations and relative time frames should be included.

The draft TMDL also cites to the bacteria management plans required by the water quality standards. Id. at 28. It must be made clear in this TMDL that a TMDL is more than a bacteria management plan.

The draft bacteria TMDL appears to set aside whatever discussion of allocations it contains and substitute the requirements of the narrative criteria contained within the standard where the TMDL discusses the requirement for WLAs to be zero and BMPs necessary for urban storm water. Id. at 29. This is evidence that the Department is not seriously considering the merit of this TMDL. A TMDL is not sufficient to merely point to the standard and say, "yes, we'll do that." A TMDL is supposed to be a quantitative analysis of the relative contributions and the load reductions of the various sources necessary to attain standards. Moreover, the monitoring proposed does not purport to evaluate the effectiveness of pollution controls, attainment of allocations, or attainment of the standard. It merely involves monitoring that should have been done prior to and in support of the development of a TMDL. As such, it neither satisfies the definition of a TMDL nor proposes to gather sufficient evidence to determine if the alleged TMDL is successful. It is more like a work plan to develop a TMDL. The claim set out in the implementation strategy for this TMDL that industrial storm water permittees will be "required to detect and eliminate discharges of human waste to the Slough" and "required to implement BMPs to the MEP to control bacteria loads" is unsubstantiated. There is no reasonable assurance presented here upon which to base such findings. It is merely a statement of the wishful thinking of the agency, without an indication of the regulatory vehicle that will be used, the timeframe in which the "requirements" will be met, the follow-up monitoring to assess the effectiveness of the "requirements," etc.. This is neither a TMDL nor an implementation plan.

D. Toxics TMDL(s)

One of the fundamental failings of the TMDL(s) for "toxics" is the Department's failure to apply the narrative criterion (OAR 340-41-445(2)(p)(A)) that it cites at the outset of the toxics TMDL. Id. at 30. The criterion specifically states that toxic substances shall not be introduced in "amounts, concentrations, or combinations" which "may be harmful, may chemically change to harmful forms in the environment, or may accumulate in sediments or bioaccumulated in aquatic life or wildlife" to "levels that adversely affect public health, safety, or welfare; aquatic life, wildlife." The TMDL fails to consider the majority of these elements of the criterion. MORE HERE>>>>

The TMDL makes specific the Department's failure to apply the narrative criteria in its "five conditions to interpret and apply the water quality standard." (The write-up demonstrates the Department's continuing failure to understand the fundamental definition of a water quality standard as containing beneficial use support, numeric and narrative criteria, and an antidegradation policy. The narrative criteria are repeatedly referred to as "standards," and impairment of beneficial use support is listed as separate from "Water Quality Standards Violations." Id. at 30.) Condition 2 limits the consideration of sediment contamination to "levels which analytical models demonstrate that water quality standards are violated." Id. at 31. The confusion between standards and criteria muddies this odd sentence, making the argument circular, however putting aside this problem it appears the document is saying that levels of chemicals in sediments that violate numeric criteria that apply to the water column are a violation. This is the only interpretation of the narrative criteria that applies to sediments so the Department has failed to implement the aspects of its narrative criterion that pertain to "amounts," "concentrations," and "combinations" which "may accumulate in sediments" to "levels that adversely affect public health" or "aquatic life" or "wildlife."

The TMDL fails utterly to consider the additive and/or synergistic effects of multiple toxic pollutants, thereby failing to show that the TMDL will lead to attainment of the standard. The TMDL fails to consider the amounts of toxic pollutants that may have and continue to accumulate in the sediment, instead restricting itself to a consideration of sediment partitioning. The TMDL fails to consider the existing uses of that portion of the public that consumes higher than average levels of higher than average contaminated fish, as discussed above. The TMDL fails to consider the other designated and existing uses, such as pisceverous birds and mammals as well as fish, that may be affected by the toxic contamination levels in sediments, bioaccumulated throughout the food chain, and/or in the water column. In short, the draft TMDL fails to show that the narrative criteria will be met by the TMDL. There are numerous studies upon which the Department can rely to apply the "gap filling" attributes of the narrative criteria, including but not limited to the narrative criterion OAR 340-41-445(2)(p)(C) which specifically calls for the use of "other published scientific literature" to "set guidance values." NWEA has provided the Department with studies related to the impact of toxic contaminants on salmonids. NWEA has had numerous conversations with Department staff over the years about its possession of a range of other published studies with regard to the impacts of toxic contaminants on wildlife. The Department is a signatory to a Memorandum of Agreement involving a Sediment Remediation Project that has generated substantial data and analysis of the risks posed by sediments in portions of the Slough. None of these are referenced in the text of this draft TMDL.

1. Lead (Pb)

The draft TMDL for lead proposes to substitute EPA's recommendation for the use of the dissolved fraction of the metal in lieu of the numeric criterion that applies to the total recoverable metal. Presumably this is through interpretation of the narrative criteria, although the document does not state the legal basis for this approach. NWEA does not believe that narrative criteria can be used to substitute for existing numeric criteria; they are stand-alone gap-fillers, not substitutes. Until such time as the numeric criterion has been changed and that change has been approved by EPA, pursuant to Clean Water Act section 303(c), the numeric criterion must be applied as the starting point for protection of the beneficial uses. Moreover, recent case law in the Ninth Circuit has held that until standards are approved by EPA, they may not be used by the states. CITE ALASKA CASE.

It is difficult to determine what the Department is considering doing in the next phase of this TMDL. The document refers to its failure to conduct "detailed modelling for toxics with CE-QUAL-W2." Id. at 33. And, it states that the "TMDL will focus on refining estimates of loads from various land uses, including estimates of the dissolved Pb load." Id. at 34. It is not clear whether the Department believes the modeling will be useful and whether it should have been done or will be done, and why it was not done. Likewise, it is not clear if the pages that follow the statement about what the TMDL "will" do are that activity or a statement about what should be done in the future. In either case, it appears that this document is not a TMDL because it goes on to state:

In Phase I of the TMDL, the allocations are based on dissolved Pb and requirements to quantify the dissolved Pb load from the sources are outlined. Estimates of dissolved Pb loads will be refined in the next phase of the TMDL, and will allow for allocations based on these estimates.

Id. at 34. In other words, the loading capacity and allocation numbers in this TMDL are make believe and they will not be real until the next phase of this action. Phase I of this TMDL is no more than an information gathering activity; it is not a quantitative plan to restore the water quality of the Slough to standards.

The draft TMDL purports to establish allocations for the entire Slough while noting that the higher concentrations in the upper end of the Lower Slough may come from an uncontrolled source (i.e., the Willamette River). Id. at 36. This is not conservative. The margin of safety established for this TMDL must specifically take this source into account in order to ensure that the TMDL will lead to attainment of standards. It is not acceptable to merely ignore an area of the waterbody where there is information that exceedence of criteria are greater.

The draft TMDL fails to consider the deposition of lead to sediments by establishing the loading capacity and related allocations on the basis of instream concentrations alone. Id. at 34-36. This is contrary to the clear language in the narrative criterion OAR 340-41-445(2)(p)(A) which requires the Department to evaluate "amounts" that will "accumulate in sediments." By establishing allocations that vary with flow, the draft TMDL does not limit the accumulations of lead that will violate state water quality standards.

The draft TMDL does not give gross or specific allocations to all controllable sources. Id. at 36, 37. Instead, the text discusses the Willamette River, CSOs, St. John's Landfill, groundwater, sediment partitioning, NuWay Oil Site (three conduits), storm water, and a MOS. Id. at 36. The implementation strategy discusses storm water, industrial permittees, and Environmental Cleanup Sites, and Designated Management Agencies (urban stormwater). Table 14, which establishes the allocations, omits any reference to the Willamette River, industrial permittees, Environmental Cleanup Sites, and differentiation between types of stormwater, i.e., it merely makes a gross allocation to the category. Despite the extensive unknowns -- sources not known or quantified, lack of modelling, source controls with unknown effectiveness, sources that cannot be controlled -- the Margin of Safety is set at 10 percent "because this percentage was assigned to unknown sources in the Waterbody Assessment (CH2MHill, Part 2, 1995). Nowhere in the TMDL document is this alleged MOS -- prepared by a private consultant presumably paid by one or more of the sources -- evaluated or rationalized by the Department. Nowhere is there a discussion of future increased sources of lead that might or might not be able to be controlled. For example, previous documents related that the Oregon Steel company, a source of air deposition, was planning to dramatically increase its air emissions of lead. In the draft TMDL for dissolved oxygen, the Department cites the City of Portland's belief that stormwater pollutant loading will rise 150 percent. Id. at 13. However, these are neither factored into the MOS nor in allocations to future growth (whether zero or something more than zero). Nowhere in the discussion is a reference to the impact of the Sediment Remediation Project which presumably is restoring sediment contaminant levels to that required to implement water quality standards. Nowhere in this discussion of the MOS is there recognition the lack of sufficient information about the effectiveness of and time frame for controls to reduce stormwater loads of lead, controls that are essential to attainment of standards because storm water is the largest contributor of lead and because the TMDL establishes allocations for the following sources at the same level as current loadings: groundwater, Willamette River, St. John's Landfill, sediment partitioning, NuWay Oil Site, Environmental Cleanup Sites (unclear allocation but discussion of BMPs), and industrial permittees (unclear allocation but discussion of BMPs).

Further evidence that the TMDL is not sufficiently quantified to ensure that allocations to reduce storm water loadings of lead is provided in the TMDL's implementation strategy. While the strategy is full of references to BMPs by this and that source, the TMDL states that "BMPs should be developed on land uses, but further delineation of sources based on land use must be done." Id. at 37. The TMDL goes on to say that in Phase I, the DMAs will be required to obtain more data and to measure the effectiveness of BMPs. Id. at 38. A monitoring plan and some BMPs that may or may not be effective at meeting allocations do not add up to a TMDL that on its face demonstrates that it will lead to attainment of standards. Therefor, this is not a TMDL.

The treatment of Environmental Cleanup Sites has improved since earlier drafts; at least they are included now. However, these sites are not given an allocation, are not factored into the MOS, and are not identified in the body of the TMDL. Instead, the draft provides vague assurances that sites "have been moved to the cleanup program," that DEQ "will develop a monitoring protocol" for them, and that "individual site managers will select BMPs for surface runoff control and metals control and require BMP implementation." Id. at 38. None of these activities are necessarily any different from those that can and may occur in the absence of a TMDL. There is no assurance of when site-specific information will be developed and that that information will be used to make the sites come into compliance with specific allocations. In other words, there is no indication that this TMDL adds any value to existing efforts through a quantitative and professional analysis of what is required to restore water quality to standards.

There is no evaluation of whether the loading capacity developed by the draft TMDL will result in fish tissue levels that meet water quality standards.

2. Organic (DDT, PCB, Dieldrin)

We disagree with the position taken in the draft TMDL that it is appropriate to use bioconcentration factors are more chemical-specific than species-specific. Data gathered in the Lower Columbia River Bi-State Water Quality Study certainly indicated otherwise, among many other studies, a point recognized by the draft TMDL. Id. at 39. Moreover, it makes no sense at all to not normalize the data -- on different species and possibly different types of tissue -- for lipid content. The TMDL certainly cannot even pretend to protect for human health considerations without an evaluation of species consumed, lipid content, and consumption rates, as discussed generally above. The Department did not need to invent this wheel; a considerable effort to do this type of evaluation is contained in EPA's Columbia River Basin Dioxin TMDL, certainly enough to be the foundation for a more thoughtful and therefore accurate analysis of the data. It is not even clear from the TMDL if the samples were homogenized samples of multiple species or the data were simply thrown together.

The text discusses the fact that the screening level for dioxin in fish tissue is .007 ppb. That screening level is only appropriate for us if the evaluations discussed above (e.g., normalizing for lipid content, factoring in consumption patterns) have been done. The data presented in Table 18 report a detection level for all nondetects higher than the screening level, demonstrating that insufficient data exist upon which to base this TMDL. Similarly, of the storm water samples evaluated for non-dioxin organic, all had detection limits above the Table 20 criteria. Id. at 40. This is not useful information. Moreover, whether any of these pollutants was detected in stormwater is not necessarily helpful to establish a TMDL because the fact that such contaminants are not found in the water column does not mean that they are not contributing to the sediment contamination of the Slough and the tissue levels found in the fish.

The draft TMDL states that "storm water sediment is the likely source of contamination of fish tissue." Id. at 42. This is echoed in comments specific to findings in the Buffalo Slough. Id. at 46. The analysis of the contribution of existing sediment contamination (from a variety of sources over the years), however, is limited to the examination of sediment partitioning to the water column. Id. at 43. The conclusion of this analysis is that "a low instream concentration causes fish tissue concentrations to exceed consumption advisory levels." Id. at 44. The TMDL fails, however, to evaluate the contribution of the sediments to the fish tissue levels by direct consumption by bottom feeders such as carp (the fish most consumed by subsistence fishers at the Slough) and lower level organisms in the food chain. Therefore, the TMDL underestimates the loading to beneficial uses and overestimates the allocations allowed to sources. Since the allocation to the sediment-derived load is fixed with the remainder divided equally between the remaining sources, this failure is likely to result in the TMDL failing to lead to attainment of standards. This is even more likely with dioxin, where the contribution of loads from sediments is a complete guess, without any professional judgment presented as a basis. The greater the lack of information, the larger the MOS must be.

The draft TMDL states that EPA recommends the use of the harmonic mean flow but uses the annual average flow due to lack of data. Id. at 45. The TMDL does not state how this variance is considered in the chosen margin of safety.

The draft TMDL provides no assurance that it will lead to attainment of standards. The WLAs for storm water and ECSI sites are merely gross allocations to categories. There is no indication that allocations to specific sources will be made in the near future. The implementation strategy discusses the need for BMPs to control erosion in order to reduce new loadings of organic to the Slough. Other than stating that a new general storm water permit for the basin will be developed, there is no information upon which to believe that this will take place. The TMDL does not state when this permit will be developed, how erosion targets will be developed that will result in meeting the gross allocation for storm water sources, how the Department will establish BMPs sufficient to meet those targets, when they will be developed, when they will be implemented, what regulatory assurance exists that they will be implemented, and what will happen in what time frame based on the monitoring of BMP effectiveness. In short, there is no assurance let alone indication that the quantitative technical analysis in this TMDL, deficient as it is, will be carried over into these actions.

Our comments on the treatment of Environmental Cleanup Sites is the same as under the draft TMDL for lead above.

The draft TMDL should explain why there is no reference to the role that the Sediment Remediation Project will play in implementing the TMDLs for toxic contaminants to result in standards attainment. If the Department believes that the Project will have no impact, it should say so. If the Department believes it will have an impact, it should state what that impact will be. The TMDL, Remediation Project, and clean-up of contaminated sites should work together, not separately.

Conclusion

Thank you for this opportunity to comment. If you have any questions about our comments, please feel free to contact us for more information.

Sincerely,

Nina Bell
Executive Director

cc: EPA Region X

News/Media | Action Alerts | Programs | About Us | Getting Involved
Resources | Contact Us | Kids Page | Northwest Places | Home

133 SW 2^nd Ave., Portland, OR 97204-3526 (503) 29rols90 FAX 295-6634