LAW OFFICE OF DAVID A. LUDDER
November 30, 2012
Mr. Eric Shaw
Standards and Assessments Section
Florida Department of Environmental Protection 2600 Blair Stone Road, MS 6511
Tallahassee, FL 32399-2400
Re: Health-based Water Quality Criteria
Dear Mr. Shaw:
This comment is submitted in response to the Florida Department of Environmental Protection’s (FDEP’s) request for comments on proposed revisions to health-based water quality criteria in Rule 62-302.530.
I. Omission of Water Quality Criteria Parameters
Completely omitted from the proposed water quality criteria are criteria for Asbestos, Methylmercury (in fish tissue), and Dioxin. If FDEP fails to promulgate criteria for these parameters, the U.S. Environmental Protection Agency (EPA) may have to promulgate new or revised criteria for these parameters.
EPA currently recommends a water quality criterion for Asbestos of 7 million fibers per liter for water and fish consumption uses based on the maximum contaminant level for drinking water. NATIONAL RECOMMENDED WATER QUALITY CRITERIA – HUMAN HEALTH CRITERIA TABLE (http://1.usa.gov/PMPLKz). See LIST OF CONTAMINANTS AND THEIR MCLS-INORGANIC (http://1.usa.gov/PMQ0FB).
FDEP should revise the Table in Rule 62-302.530 to include the EPA recommended criterion for Asbestos in Class I waters.
In the proposed revisions to the Table in Rule 62-302.530, the criteria for Mercury remains unchanged. The values have their origin in EPA’s AMBIENT WATER QUALITY CRITERIA FOR MERCURY – 1984 (http://1.usa.gov/OBouv4). Those values are based on a Food and Drug Administration (FDA) “action level,” rather than a reference dose (RfD) or cancer potency factor. The methods used by EPA to derive the criteria for Mercury in 1984 have been abandoned and replaced by a new method.
In 2001, EPA published WATER QUALITY CRITERION FOR THE PROTECTION OF HUMAN HEALTH: METHYLMERCURY (http://1.usa.gov/P4AJMo). The recommended criterion for methylmercury is based on a RfD of 0.1 μg/kg/day. The fish tissue criterion of 0.3 mg methylmercury/kg fish “is the concentration in fish tissue that should not be exceeded based on a total fish and shellfish consumption-weighted rate of 0.0175 kg fish/day.” GUIDANCE FOR IMPLEMENTING THE JANUARY 2001 METHYLMERCURY WATER QUALITY CRITERION (2010) (http://1.usa.gov/P4D4XL) at xvi.
“With the publication of the 2001 fish tissue criterion, EPA withdrew the previous human health water quality criterion for mercury as the recommended section 304(a) water quality criterion for states and authorized tribes to use as guidance in adopting water quality standards . . .. These water column criteria, however, may be temporarily part of revised mercury criteria until the triennial review that follows the criterion adoption to help the transition in implementing the fish tissue criterion.” Id. at 20 (emphasis added).
EPA recommends that when states and authorized tribes adopt new or revised methylmercury water quality criteria, they adopt the criteria in the form of a fish tissue methylmercury concentration. This is the preferred form for the following reasons:
- ! A criterion expressed as a fish tissue concentration is closely tied to the “fishable” designated use goal applied to nearly all waterbodies in the United States.
- ! A fish tissue concentration value is expressed in the same form (fish tissue) through which humans are exposed to methylmercury.
- ! A fish tissue concentration value is more consistent with how fish advisories are issued.
- ! At environmentally relevant concentrations, methylmercury is currently easier to detect in fish tissue than in water samples.
FDEP has not proposed to adopt a fish tissue criterion for Methylmercury. FDEP claims to have “addressed the mercury issue by recently completing a Statewide TMDL for mercury.” TECHNICAL SUPPORT DOCUMENT: DERIVATION OF HUMAN HEALTH-BASED CRITERIA AND RISK ASSESSMENT (Nov. 2012) at 53. According to FDEP’s website (http://www.dep.state.fl.us/water/tmdl/draft_tmdl.htm#mercury) however, the mercury TMDL remains draft. In any case, the draft MERCURY TMDL FOR THE STATE OF FLORIDA (Oct. 29, 2012) (http://bit.ly/SpqVRm) relies on narrative “free-from” water quality criteria, id. at 5, not numeric fish tissue Methylmercury concentrations as recommended by EPA. Moreover, a mercury TMDL is not a substitute for a water quality criterion.
Accordingly, FDEP should revise the Table in Rule 62-302.530 to include the EPA recommended criterion for Methylmercury in fish tissue adjusted for Florida’s higher fish consumption rate.
Florida currently has no water quality criteria for Dioxin and does not propose any. The EPA-criteria for Dioxin in Florida promulgated in 1992 (40 C.F.R. § 131.36(d)(6)) is based on a fish consumption rate of only 6.5 grams per day. EPA’s current recommended water quality criteria for Dioxin (2,3,7,8-TCDD) is 5.0E-09 ug/L for water and fish consumption uses and 5.1E-09 ug/L for water consumption uses alone. NATIONAL RECOMMENDED WATER QUALITY CRITERIA – HUMAN HEALTH CRITERIA TABLE (http://1.usa.gov/PMPLKz). These recommended criteria are based on a cancer slope factor (q1*) equal to 1.56E+05 (mg/kg-day)-1 published in AMBIENT WATER QUALITY CRITERIA FOR 2,3,7,8-TETRACHLORODIBENZO-P-DIOXIN (1984) (http://1.usa.gov/TZIY0Q) and a national default fish consumption rate of 17.5 grams per day adopted by EPA in METHODOLOGY FOR DERIVING AMBIENT WATER QUALITY CRITERIA FOR THE PROTECTION OF HUMAN HEALTH (2000) (http://1.usa.gov/TZI8kQ).
FDEP should revise the Table in Rule 62-302.530 to include the EPA recommended criteria for Dioxin adjusted for Florida’s higher fish consumption rate (including protection of high-end consumers, e.g., 99th percentile, at 10-4 cancer risk level).
The excuse that FDEP should defer promulgation of Dioxin criteria because EPA is conducting a reassessment of Dioxin might be acceptable if we had confidence that the EPA reassessment might be concluded in a reasonably short time-frame. However, EPA has been reassessingthetoxicityofDioxinsince1991. HISTORYOFDIOXINREGULATION&REASSESSMENT (http://bit.ly/TZILuT). See DIOXIN (http://1.usa.gov/TZJSKQ). There is no basis to assume that the reassessment will be concluded within one or even two years. In the meantime, FDEP has no criteria for Dioxin and the EPA criteria promulgated for Florid are under-protective.
Accordingly, FDEP should revise the Table in Rule 62-302.530 to include the EPA recommended criterion for Dioxin in fish tissue adjusted for Florida’s higher fish consumption rate.
II. Exclusion of Chemicals Based on Safe Drinking Water Act
FDEP acknowledges that several chemicals pose a threat of dermal absorption from showering. TECHNICALSUPPORTDOCUMENT:DERIVATIONOFHUMANHEALTH-BASEDCRITERIA AND RISK ASSESSMENT (Nov. 2012) at 11 (“the permeability through the epidermis (skin) of some compounds is sufficient to add a third significant exposure route via dermal absorption during showering/bathing and swimming”). These include hexachlorobutadiene, pentachlorophenol, total (carcinogenic) PAHs, acenapthene, anthracene, flouranthene, flourene, and pyrene. However, FDEP decided not to consider showering exposures for the following reason:
The HHPRC advised the Department that there was little value in including exposure from showering/bathing because drinking water is treated and must meet drinking water standards; therefore, the exposures can safely be assumed to minimal. DEP agrees with the committee recommendation and has therefore not pursued inclusion of showing exposures as a component of the criteria derivation or risk assessments.
TECHNICAL SUPPORT DOCUMENT: DERIVATION OF HUMAN HEALTH-BASED CRITERIA AND RISK ASSESSMENT (Nov. 2012) at 34.
This conclusion is arbitrary for several reasons. First, drinking water standards are based on ingestion of contaminants, not dermal absorption. A standard based on ingestion may not be protective for dermal absorption. Second, drinking water standards are not always based solely on protection of human health. Health-based standards may be modified to account for costs of treatment. Third, there are no drinking water standards for hexachlorobutadiene, acenapthene, anthracene, flouranthene, flourene, and pyrene. Thus, drinking water standards provide no protection for dermal absorption of these contaminants. And finally, FDEP has identified no universal minimum treatment technology employed by public water supplies or the capability of such treatment technology to remove hexachlorobutadiene, acenapthene, anthracene, flouranthene, flourene, and pyrene from a Class I waterbody. Without such an analysis, human exposure to these contaminants during showering must be assumed to be at the maximum criteria level.
Accordingly, FDEP needs to reconsider dermal absorption of contaminants during showering.
III. Fish Consumption Estimates
A. Commercial Fish Consumption
FDEP developed the commercial fish consumption distribution “based on the variability described in the NHANES distribution adjusted to the commercial landings supported mean (“landings mean”). TECHNICAL SUPPORT DOCUMENT: DERIVATION OF HUMAN HEALTH-BASED CRITERIA AND RISK ASSESSMENT (Nov. 2012) at 18. “This approach assumes that national NHANES data are representative of the variability among Florida consumers. National NHANES data currently represent the best estimate of consumption patterns in the absence of more contemporary Florida or regional specific information.” Id. (emphasis added).
A fish consumption distribution derived from the national NHANES distribution is likely to be unrepresentative of consumption in Florida and other southeastern coastal states. This is so for at least two very obvious reasons. Florida offers more fishing opportunities than many other states included in the national NHANES data because of its extensive coastline. And, Florida has a warmer climate than many other states included in the national NHANES data allowing for greater fishing frequencies. A regional consumption distribution can probably by developed from NHANES (see e.g., Mahaffey, KR; Clickner, RP; Jeffries, RA, 2009, Adult women’s blood mercury concentrations vary regionally in the United States: Association with patterns of fish consumption (NHANES 1999-2004), Environmental Health Perspectives 117:47-53 cited in EXPOSURE FACTORS HANDBOOK, EPA/600/R-09/052F (Sep. 2011) at 10-21 ).
At the November 15, 2012 workshop on draft revised human health criteria, FDEP explained that it did not pursue a regional NHANES consumption distribution because it was in a hurry to promulgate revised criteria. Administrative convenience is not a justification for promulgating under-protective criteria that are not based on the best available science. Accordingly, FDEP needs to further explore a regional distribution from NHANES.
B. Recreational Fish Consumption
In the development of the recreational fish consumption distribution, FDEP notes:
A realistic analysis of long-term average consumption rates must also take into consideration the frequency (number of weeks per year) at which individuals actually consume fish and the variation in frequencies among individuals. The long-term average daily consumption for an individual is the product of their average consumption (for weeks when they choose to consume fish) and the frequency of consumption divided by number of days. The 7-day recreational consumption distribution was used to define the 7-day average consumption rate, and a binomial distribution of 52 trials (weeks) with success probability of 42.3% was used to represent the frequency of consumption (Figure 2-8). As previously explained, the success probability was set based on average fishing days per year.
TECHNICAL SUPPORT DOCUMENT: DERIVATION OF HUMAN HEALTH-BASED CRITERIA AND RISK ASSESSMENT (Nov. 2012) at 29 (emphasis added). The consumption frequency (probability) of 42.3% “assumes that the decision to eat seafood in any one week is a random process and that there is a 42.3% chance that a fish consumer will choose to eat recreationally caught fish during any week. The 42.3% chance was based on information from a U.S. Fish and Wildlife Service survey of fishing, hunting, and wildlife-associated recreation (USFW 2008). This survey estimated that on average, Florida resident anglers fished 22 days per year. It was assumed that for the average angler this represented fishing during 22 weeks of the year, which translated to having fish available for consumption on average 22 out of 52 weeks (42.3%).” Id. at 28-29. In other words, fishing frequency among anglers was used as a surrogate for recreationally-caught fish consumption frequency among the general population.
It is common knowledge that many anglers share their catch with non-anglers who were not captured in the U.S. Fish and Wildlife survey. In addition, some portion of recreational catch is often frozen for later consumption. Thus, the assumption that recreationally-caught fish consumption frequency among the general population is equal to the fishing frequency of anglers (22 weeks per year) is arbitrary and likely to underestimate recreationally-caught fish consumption frequency among the general population. Accordingly, FDEP needs to revisit and revise the consumption frequency for recreational fish consumers.
C. Combined Commercial and Recreational Consumption
FDEP also notes:
USFWS (2008) estimated that there were 1,881,000 resident recreational anglers (ages 16 and up) in the state of Florida, which represents 13.5% of the 2010 Census adult fish consumer population (13,911,266). It was assumed that 13.5% represents the portion of the population that are recreational consumers. The portion that are general consumers can therefore be estimated by subtracting the percentage of non-consumers and recreational consumers from 100% , which results in 80.5% of the population falling into the general consumer category.
TECHNICAL SUPPORT DOCUMENT: DERIVATION OF HUMAN HEALTH-BASED CRITERIA AND RISK ASSESSMENT (Nov. 2012) at 30. In other words, the validity of the estimated percentage of the population that falls into the general consumer category (80.5%) is only as good as the validity of the estimated percentage of the population that falls into the recreational consumer category (13.5%). These estimated percentages play an important role in determining the combined fish consumption rate.
The two above consumption distributions, in conjunction with a custom distribution describing consumer pattern, were used in the Monte-Carlo risk analyses. The consumer pattern distribution indicated whether an iteration represented a non-consumer, general consumer, or recreational consumer. Probabilities for non-consumer, general consumer, and recreational consumer were set to 0.06, 0.805, and 0.135, respectively. The consumption rate for non-consumers was set to 0 g/day. Consumption rates for general consumers were established based on the landings adjusted commercial distribution, while the rates for recreational consumers were set as the sum of the landings adjusted commercial and recreational consumption distributions.
The validity of the conclusion that 13.5% of the Florida are recreational fish consumers is questionable for two reasons. First, as discussed above, the population of recreational fish consumers is likely to be larger than the population of recreational anglers. This is so because many anglers share their catch with non-anglers. Thus, the assumption that the number of recreational fish consumersisequaltothenumberofrecreationalanglersisnotvalid. Second,theTechnicalSupport Document fails to undertake a critical analysis of the validity and accuracy of 1,881,000 (13.5%) estimate of resident anglers. The origin of the estimate is U.S. Fish & Wildlife, 2006 NATIONAL SURVEY OF FISHING, HUNTING, AND WILDLIFE-ASSOCIATED RECREATION (http://www.census.gov/prod/2008pubs/fhw06-fl.pdf). That survey states:
The Survey is a snapshot of one year. The information it collected tells us how many people participated and how much they spent on their activities in the State in 2006. It does not tell us how many anglers, hunters, and wildlife watchers there were because many do not participate every year. For example, based on information collected by the Survey’s household screen and detailed phase, we can estimate that about 33 percent more anglers and hunters participated nationally in at least 1 of the 4 years prior to the survey year 2006. Id. at 2 (emphasis added). The survey methodology indicates that “[t]he total screening sample in Florida consisted of 3,800 households. * * * Of all housing units in sample, about 3,166 were determined to be eligible for interview. Interviewers obtained interviews at 2,958 of these units . . .. Local field representatives conducted interviews by telephone when possible, otherwise through a personal visit. ” Id. at 66. These are fewer households than were surveyed in Degner, D.L., C.M. Adams, S.D. Moss, and S.K. Mack. 1994, PER CAPITA FISH AND SHELLFISH CONSUMPTION IN FLORIDA. Florida Agricultural Market Research Center, University of Florida, Gainesville, FL.
Accordingly, FDEP needs to revisit and revise its estimates of the percentage of the population that falls into the recreational consumer category and general consumer category.
For the foregoing reasons, FDEP has additional work to do before it can propose water quality criteria for the protection of human health that are not arbitrary.
David A. Ludder