United States Department of the Interior
620 South Walker Street
Bloomington, IN 47403-2121
(812) 3344261 FAX (812) 334-4273

Mr. Vince Epps
Office of Land Quality
Indiana Department of Environmental Management
100 N Senate Ave.
Indianapolis, Indiana 46206-6015

Dear Mr. Epps:

October 31, 2000

This regards your June 21, 2000 letter requesting our review of CBS's May 2000 letter and data transmitted on fish tissue collected from Clear Creek downstream of the Lemon Lane Landfill site in Monroe County, Indiana in 1996 and 1999.

This letter has been prepared pursuant to our role as providers of biological and technical assistance to EPA under the Fish and Wildlife Coordination Act (16 USC 661 et seq.) and is consistent with the intent of the National Environmental Policy Act of 1969, the Endangered Species Act of 1973, and the U.S. Fish and Wildlife Service's Mitigation Policy.

We have reviewed both the 1996 and the 1999 datasets, and all other available Clear Creek analytical chemistry data from 1983 to present There are 215 fish tissue analytical chemistry samples in this dataset which we have tabulated and provide to you for your use (see tables 1-10, enclosed) We provide the following comments relative to the existing body of knowledge for Clear Creek and provide our opinion regarding trends of PCB contamination in Clear Creek

There are essentially 9 locations along Clear Creek where fish tissue samples have been collected over the years: 1) lst & Walker Streets (~0.7 miles downstream of Lemon Lane groundwater source) 2) Allen Street (~1 miles downstrearn), 3) behind Wee Willies restaurant (located at 1724 S. Walnut Street) (~2 miles downstream), 4) Country Club Road (~3 miles downstream), 5) Gordon Pike ( ~4 miles downstream), 6) Fluckmill Rd (~9 miles downstream), 7) Ketcham Road (~12 miles downstream), 8) Gore Road (~14 miles downstream), and 9) Strain Ridge Road (~15 miles downstream)

As we have learned from the original risk assessment planning and implementation, there are several generalizations that can be used to describe what is occurring in Clear Creek. First, proximity to the source of contamination (Lemon Lane's Groundwater discharge) and size (which is directly related to age) of fish present play a predominant role in the bioaccumulation of PCBs. In the upper reaches of Clear Creek the fish community consists primarily of minnows such as creek chubs and bluntnose minnows In the middle reaches of Clear Creek several species of pan fish appear (e.g. longear sunfish, green sunfish and rock bass). In the downstream reaches minnows and pan fish are found in proportionately lower numbers, but suckers, redhorse, catfish, large and smallmouth bass are more commonly captured. PCB bioaccumulation is somewhat related to duration of exposure (i.e. older resident fish of Clear Creek are likely to have accumulated higher concentrations of PCBs). There also appear to be specific differences in PCB bioaccumulation potential between species apparently having to do with dietary and/or physiological differences. For instance, this Clear Creek dataset indicates that stonerollers were almost always more contaminated than similar sized white suckers from the same location and green sunfish were almost always more contaminated than a similar sized longear sunfish sample from the same location.

There are other distinct trends in the tissue data as well. For each of the following tables of analytical results, we present for your consideration our observations on PCB contaminant trends and the interrelationships between the various datasets.

The farther downstream from the Lemon Lane groundwater discharge that the creek chub samples are collected, the lower the PCB contamination has been in the samples. Within each sample location, there appears to be a significant correlation between length (as a surrogate for age) and whole body PCB concentration. The concentrations of PCBs in the creek chub fillet samples collected in 1999 by CBS do not represent a real downward trend in PCB concentrations when compared to whole body creek chub samples from previous years because one would expect PCB concentrations to be reduced in filleted versus whole body samples if for no other reason than a reduction in the lipids remaining in the sample. We would like to believe that Indiana would not base fish consumption advisories on fillet samples collected from minnows.

This species should be a target for any and all future sampling efforts that influence fish consumption advisories. It is clear from this dataset that green sunfish bioaccumulate PCBs more significantly than many other co-located fish species. It is also a desirable species of panfish from an angler's point of view. Collectively, the green sunfish data clearly point out that samples collected at Fluckmill Road have much lower concentrations of PCBs than either upstream or downstream samples. In fact, even much smaller green sunfish samples from Ketcham Road are more contaminated than samples from the Fluckmill Road site. This may reflect the fact that the habitat in the Fluckmill Road reach of Clear Creek consists primarily of a riffle/run microhabitat than some of the other sampling locations.

Except for the 1 sample in 1999, there really is not any other analytical chemistry data on blugills for Clear Creek. Longear sunfish is fairly common throughout the middle and lower reaches of Clear Creek and have been shown to bioaccumulate significantly less PCBs than green sunfish. Although there can be a lot of variability between longear sunfish samples at the same locations, the longear sunfish samples from Strain Ridge Road in 1999 and from Ketcham Road in 1996 indicate that bioaccumulation is still significant. We recommend that future comparisons of PCB bioaccumulation in longear sunfish also be done at Ketcham Road, Gore Road and/or Strain Ridge Road.

There is no significant difference between the rock bass fillet samples collected from Ketcham Road in 1996 and 1999. Filleting rock bass reduces PCB concentrations by as much as 40-80 percent We are assuming for this data review that all fillet samples presented here follow IDEM filleting standard methodology (scaled, with skin remaining). Fluckmill Road samples are again significantly lower than other downstream reaches

Table 5-Largemouth and Smallmouth Bass

There are no 1999 1argemouth or smallmouth bass samples with which to compare with the existing data. Previous samples from the downstream most sampling points, although variable, potentially had extremely high fillet PCB concentrations. These are the areas were angling pressure would be highest, and these species are likely to be the most highly sought after. Given the stream morphology of Clear Creek we assume that the stretch from Gore Road to Strain Ridge represents one of the more highly depositional areas of Clear Creek.

Only three catfish samples have been collected for chemical analysis during the 17 years that fish tissue samples have been collected from Clear Creek. In 1992, a 1 pound flathead catfish was captured at Strain Ridge Road with nearly 6% lipids and a PCB concentration of 3.74 ppm, whole body wet weight. In contrast, a slightly smaller brown bullhead was also captured at Ketcham Road that was only 0.5% lipid and 0.34 ppm. This data indicates that PCBs in catfish are somewhat related to the % lipids in a fish and the higher PCB concentrations will likely be found in the older, long term residents of Clear Creek

The stoneroller is a bottom-feeding minnow that is typically found in riffle habitats of streams. It feeds on "diatomaceous scum, mud with its insect larvae, small mollusks and filamentous algae" (McClane 1978). According to McClane (1978) stonerollers "are highly regarded as a food fish in Tennessee," however, any of the stonerollers collected in Clear Creek are probably too small for human consumption. Direct comparisons between the 1996 and 1999 analytical data are difficult because they were not captured from the same locations . These fish have high lipid concentrations (1.0-7.5%) and substantial concentrations of PCBs.

Northern hog suckers are also a bottom feeding fish that is uniquely adapted to feeding in the swift currents of riffles (McClane 1978). The 1999 dataset for hog suckers at Strain Ridge Road consists of whole body samples and fillet samples. The 1999 whole body samples were much smaller than the 1996 Strain Ridge Road hog sucker samples so they are not directly comparable. However, the 1999 fillet samples came from hog suckers that are similar in size to the 1996 whole body samples so based on this it could be concluded that filleting the largest fish does significantly reduce PCB concentrations in the samples

We combined the analytical chemistry data from white suckers, spotted suckers, and redhorsesuckers into one table for these comparisons. White suckers are more common than other sucker species in the upper half of Clear Creek having been captured as far north as Allen Street during the spring spawning run. The historic average length of captured white suckers at Ketcham Road was 326.7 mm and the average length of other sucker species in the lower reaches of Clear Creek is 353.6 mm. In 1996, 1997 and 1999 at least one sucker fillet sample exceeded the 1.9 ppm PCB benchmark for level 5 consumption advisory. This indicates that no significant changes are warranted in Indiana's Fish Consumption Advisory for Clear Creek. White suckers at Fluckmill Road follow the trend that many other fish in Clear Creek do in that Fluckmill Road fish samples are not as highly contaminated as other sampling locations of Clear Creek.

This summary table summarizes all the fish tissue samples collected from Clear Creek from 1983 to 1999. It is broken out by sampling event, location, and by species sampled and sample type. Also reported are the 6 samples that have been analyzed for dioxins and furans. These samples verify that PCBs are the only significant contaminant of concern in Clear Creek fish. Overall, this data indicates that the PCB contamination has not significantly changed in Clear Creek in the past 8 to 10 years. Any and all future analytical chemistries on Clear Creek fish will continue to add to an understanding of the PCB contamination trends of Clear Creek however, we offer the following suggestions to accurately quantify risks to humans from consuming Clear Creek fish.

We recommend that emphasis be placed on samples from Ketcharn Road, Gore Road and/or Strain Ridge Road over samples collected at Fluckmill Road. Ketcham, Gore and Strain Ridge most often contain fish of a size that anglers would pursue whereas the Fluckmill Road site is more reflective of riffle habitat conditions. If a more upstream sampling location is required on Clear Creek, then sampling at Country Club Road would be highly recommended. Green sunfish samples should be of primary importance in the "panfish" category over bluegill and longear because they have tended to be significantly more contaminated and yet are equally sought after by anglers. Although data on stonerollers and hog suckers is valuable ecologically, it is probably less so from a human health perspective. More emphasis should be placed on analytical data from the larger (older) fish, which are more sought after by anglers. We recommend spotted suckers, redhorse, white suckers, flathead catfish, smallmouth and largemouth bass samples be collected when possible. Filleting techniques should follow standard scaleless, skin on protocols.

If you have any questions regarding this information, please contact Dan Sparks of my staff at (812) 334-4261, extension 219.

Sincerely Yours,
Scott E. Pruitt
Field Supervisor