Assessment Comments
Assessment is based on results of (1) ambient monitoring by Illinois EPA near Lock and Dam 15 from 2010-2011, (2) fish contaminant monitoring conducted by EPA/IDNR in 2001 and 2006, and (3) fish contaminant monitoring conducted by ALCOA in 2012.
Basis for Assessment
SUMMARY: The Class A1 (primary contact recreation) uses are considered “not assessed” (IR 3a) due to the lack of monitoring information upon which to base an assessment. The Class B(WW1) aquatic life uses are assessed (monitored) as "not supported" (IR Category 5a) due to violations of state water quality criteria for aluminum. The existing aquatic life impairment due to cadmium is proposed for de-listing due to an error in assessment. The assessment of support of the Class C (drinking water) uses is changed from "not supported" (IR 5a) to “fully supporting” (IR 2a) due to an error in assessment. This change represents a proposed de-listing of the Class C impairment due to arsenic. Fish consumption uses are assessed (evaluated) as "partially supported” (IR Category 3b) due to violations of state Human Health criteria (fish + water) for dieldrin. Results of fish contaminant monitoring in 2001, 2006, and 2012, however, suggest “full support” of fish consumption uses and also suggest that levels of dieldrin are below levels of detection in bottom feeding fish in this river segment. Sources of data for this assessment include (1) results of ambient water quality monitoring conducted from January 20, 2010 to December 6, 2011 by the Illinois Environmental Protection Agency at station M-02 at Lock and Dam 15 at Arsenal Island at river mile 482.9, (2) results of fish tissue monitoring in Pool 15 conducted for the U.S. EPA/IDNR (RAFT) fish tissue monitoring program in 2001 and 2006, and (3) results of fish contaminant monitoring conducted by ALCOA in 2012.
EXPLANATION: The Class A1 (primary contact recreation) uses are considered “not assessed” due to the lack of information upon which to base an assessment. Although ambient monitoring is conducted in this river segment by the Illinois EPA, the resulting data are for fecal coliform bacteria. These data are not applicable to Iowa’s water quality standard for E. coli to protect the designated Class A1 primary contact recreation uses.
Similar to the previous assessments, the Class B(WW1) aquatic life uses are assessed (monitored) for the current (2014) cycle as “not supported” (IR Category 5a) based on results of Illinois EPA monitoring at station M-02 near Rock Island, IL. The data from this station from 2010 through 2011 show repeated violations of Iowa’s aquatic life criteria for aluminum. Two of the 16 samples analyzed for dissolved aluminum exceeded the chronic aquatic life criterion (87 ug/l) for aluminum (maximum = 896 ug/l), and 14 of the 16 samples analyzed for total aluminum (88%) exceeded this criterion (minimum = 48.9 ug/l; maximum = 2,890 ug/l total aluminum) [Iowa’s aquatic life criteria for metals are expressed as “total recoverable” in the Iowa Water Quality Standards]. In addition, seven of the 16 samples analyzed for total aluminum (44%) exceeded the acute aquatic life criterion (750 ug/l). Based on Iowa DNR’s assessment/listing methodology, more than one violation of a chronic or acute water quality criterion for a toxic parameter over a three-year period suggests impairment of aquatic life uses. Thus, the frequency of violations of the aquatic life criteria for aluminum (14 violations) suggests that the Class B(WW1) aquatic life uses of this assessment segment should be assessed as “not supported” and thus this assessment segment is placed in IR Category 5a.
The existing impairment of the Class B(WW1) aquatic life uses due to cadmium, however, is proposed for de-listing. A review of the methods used by Iowa DNR to identify the existing impairment for cadmium showed that the impairment was based entirely on estimated data (i.e., data reported as between the method detection level and the reporting limit). As described in the U.S. Geological Survey’s Open File Report 99—193 (Oblinger Childress et al. 1999), estimated data may not be appropriate for addressing regulatory issues:
"[Laboratories] providing providing estimates of low concentrations gives the user the flexibility to decide when it is appropriate to censor the reporting level and when it is appropriate to use estimated values below the reporting level for interpretations that are based on particular objectives of a study. Some studies require a high degree of certainty for an individual analysis, such as those that address regulatory issues. Others do not require the same high degree of certainty for an individual analysis but rather base interpretation on results from large numbers of samples, such as studies that assess the occurrence of a constituent in the environment."
Given that (1) estimated values may not be appropriate for addressing regulatory issues (such as Section 303(d) listing), (2) all data used to identify violations of Iowa’s chronic aquatic life criteria for cadmium were estimated values, and (3) the Illinois method detection level for cadmium (0.18 ug/l) is below Iowa’s chronic criterion for cadmium (0.45 ug/l), the existing aquatic life impairment based on cadmium is proposed for de-listing for the current (2014) Integrated Reporting Cycle.
For chromium, copper, lead, nickel, and zinc, none of the approximately 16 samples analyzed for these metals exceed the respective chronic aquatic life criteria during the 2010-2012 period. Similarly, none of the approximately 16 samples exceeded Iowa’s Class B(WW1) criteria for dissolved oxygen, pH, or ammonia/nitrogen.
The assessment of support of the Class C (drinking water) uses is changed from “partially supporting (IR 5a) to “fully supporting” (IR 2a) based on an assessment error. This change represents a de-listing of the Class C impairment for arsenic. Iowa DNR’s previous assessments of support of the Class C (drinking water) uses in this river segment were based on violations of the state human health (fish + water) water quality criterion for arsenic III. Monitoring data from Illinois EPA station M-02 at Lock & Dam 15 have shown that average levels of total and dissolved arsenic in this river segment typically exceed Iowa’s human health criterion (fish + water) of 0.18 u g/l. For example, of the 16 samples analyzed for total arsenic from 2010-2011 at ILEPA Station M-02, 13 samples (88%) with detectable levels of arsenic (maximum of 8.0 ug/l) exceeded Iowa’s Human Health criterion of 0.18 ug/l (average of 2.1 ug/l). According to Iowa DNR’s assessment/listing methodology, if the mean level of a toxic metal or pesticide is greater than an MCL or human health criterion, impairment ("nonsupport") of the drinking water use is indicated. Average arsenic levels for both the 2010-12 samples analyzed as dissolved arsenic (2.4 ug/l) and as total arsenic (2.1 ug/l) were well above Iowa’s Human Health criterion of 0.18 ug/l. These results are consistent with results of past monitoring periods. A review of this assessment, however, has shown that the impairment of the drinking water use for this river segment was incorrectly identified by comparing Illinois EPA data for either total arsenic or dissolved arsenic to Iowa’s water quality criterion for arsenic III. Results of water quality monitoring for arsenic III would be needed in order to accurately identify violations of Iowa’s Human Health criterion for arsenic III, and these data do not exist. Because Section 303(d) listing has regulatory implications, accurate identification of impairments is essential. Thus, this impairment is proposed for de-listing for the current (2014) Integrated Reporting cycle.
Monitoring results for other parameters do not show violations of Class C (drinking water) criteria. Maximum and average levels of nitrate in the 16 samples collected (2.9 mg/l and 1.85 mg/l, respectively), atrazine (maximum value of 0.23 ug/l in 21 samples), and Class C toxic metals at station M-02 were well below their respective MCLs and Class C human health criteria during the 2010-2012 period.
The Human Health designated use (including the fish consumption use) is assessed (evaluated) as “partially supporting” (IR Category 3b-potentially impaired). One of the 16 samples analyzed by ILEPA for dieldrin at station M-02 during the 2010-2012 period (the only detectable level reported) exceeded the Iowa human health-fish (HH-fish) criterion for dieldrin of 0.00054 ug/l; the detected level of dieldrin was 0.0021 ug/l. According to IDNR’s assessment/listing methodology, if the mean level of a toxic metal or pesticide is greater than its respective human health criterion, impairment ("nonsupport") of the Human Health use is indicated. However, the samples with detectable levels of dieldrin were too few (two) and the method detection level (0.0014 ug/l) was too high (2.5 times the HH criterion for dieldrin) to allow calculation of a meaningful summary statistic (i.e., mean) for dieldrin at station M-02.
Fish contaminant sampling has been conducted in this river segment as part of the U.S. EPA/IDNR RAFT program. Results of this monitoring have shown that levels of dieldrin in the composite samples of fillets from common carp and channel catfish analyzed for the 2001 RAFT were below the levels of detection (<0.003 and <0.0052 mg/kg) (the 2006 RAFT samples were not analyzed for dieldrin). Although Iowa does not have a fish consumption advisory trigger level for dieldrin, the very low dieldrin levels seen in the 2001 RAFT samples from Davenport (i.e., < 5 ppb) is far below the Iowa’s previous advisory trigger of 300 ppb of dieldrin (i.e., the U.S. FDA action level for dieldrin). Levels of dieldrin in Iowa fish have declined significantly since the early and mid-1980s when levels of 300 ppb to 500 ppb were not uncommon. Due to the low levels of dieldrin in fish tissue samples from this river segment, and due to the low frequency with which violations of the human health (fish + water) dieldrin criterion occurs, this assessment (based on the detection of dieldrin in water) is considered appropriate for Iowa’s IR Category 3b (potential impairment). IR Category 3b waters will be added to Iowa’s list of waters in need of further investigation.
Despite the potential impairment of the Human Health/Fish Consumption uses suggested by levels of dieldrin in water samples, the results of EPA/DNR fish tissue (RAFT) monitoring near Davenport in 2001 and 2006 suggest that fish consumption uses should be assessed as "fully supported". The 2001 composite samples of fillets from common carp and channel catfish had generally low levels of contaminants. The levels of total PCBs in both samples, however, exceeded the IDNR/IDPH trigger level of 0.20 ppm for a one meal per week consumption advisory: the level of PCBs in the sample of common carp fillets was 0.476 ppm and in the sample of channel catfish fillets was 0.544 ppm. According to the IDNR/IDPH advisory protocol, two consecutive samplings that show contaminant levels are above the trigger level in fillet samples are needed to justify issuance of an advisory. Because, however, the levels of contaminants in the 2001 sample were more than twice the 1 meal/week trigger level, and due to the history of elevated levels of PCBs in this river reach, the fish consumption uses were assessed as “fully supported/threatened” (impaired, IR Category 5) for the 2006 Section 303(d) listing cycle. A consumption advisory had existed for the Iowa side of Pool 15 from 1989 through 2000 due to levels of PCBs in common carp and carpsuckers (Carpiodes spp.) that exceeded the FDA action level of 2.0 ppm. Additional monitoring was conducted in 2006 to determine whether a fish consumption advisory was again needed for this waterbody.
Results of the 2006 RAFT sampling at Davenport, however, showed much lower levels of contaminants than did the 2001 sampling. Levels of primary contaminants in the composite sample of common carp fillets were as follows: mercury: 0.09 ppm; total PCBs: 0.09 ppm; and technical chlordane: <0.03 ppm. Levels of primary contaminants in the composite sample of largemouth bass fillets were as follows: mercury: 0.184 ppm; total PCBs: 0.09 ppm; and technical chlordane: <0.03 ppm. These results showed a much lower level of total PCBs in common carp (0.09 ppm vs. 0.476 ppm in 2001) and indicated that a consumption advisory was not justified for this river segment. Thus, the fish consumption uses were assessed (monitored) as “fully supported” for the 2008 through 2012 Section 303(d) listing cycles.
Fish contaminant monitoring was conducted in this segment of the Mississippi River in September 2012 as part of ongoing studies by ALCOA to update the levels of PCBs in UMR Pool 15 fish (channel catfish and common carp) and to identify any trends in PCB contamination based on periodic monitoring in this Pool over the last 25 years. Sampling was conducted at three locations: Pool 15 near the ALCOA facility, Pool 14 near LeClaire, IA, and Pool 15 on the Illinois side of the river. The LeClaire and Illinois sites were chosen as reference sites. ALCOA used performance standards of 0.226 mg/kg for channel catfish and 0.231 mg/kg for common carp (these performance standards approximate Iowa’s threshold for issuing a 1 meal/week consumption advisory). Results showed that levels of PCBs in channel catfish were low at all sample sites and met the performance standard. Levels of PCBs in common carp met performance standards at the LeClaire (Pool 14) and Illinois-side sites, but levels near the ALOCA facility were above the performance standard. It was hypothesized that the larger size of fish collected in Pool 15 near ALCOA was, in part, responsible for the higher levels of PCBs seen. An Iowa DNR review of the 2012 ALCOA data show that levels of PCBs in the size of common carp collected for the Iowa/U.S. EPA fish tissue monitoring program (RAFT) (12-21 inches, total length) were well below Iowa’s advisory threshold for PCBs of 0.2 mg/kg at all sites sampled in 2012. For common carp between 12 and 21 inches in total length, none of the PCB levels in fish from any of the ALCOA sites sampled exceeded 0.1 mg/kg (i.e., half the Iowa consumption advisory trigger). PCB levels tended to exceed the Iowa advisory trigger in fish greater than 24 inches in total length (the largest common carp analyzed was nearly 30 inches in total length). None of the PCB levels seen in the larger common carp, however, exceeded Iowa’s “do not eat” advisory threshold of 2.0 mg/kg: the maximum PCB concentration was 0.715 mg/kg of total PCBs. Based on the low levels of PCBs in samples of channel catfish from all study locations, (2) low levels of PCBs in common carp from all sites except near the ALCOA facility, and (3) low levels of PCBs in common carp of sizes that meet EPA/Iowa DNR fish tissue monitoring guidelines, the human health/fish consumption uses of this assessment segment are assessed as “fully supported”. Based on results of the 2012 ALCOA study, however, follow-up fish tissue monitoring will be conducted in UMR Pool 15 in 2013 as part of the EPA/DNR fish tissue monitoring program.
