The harbor at Erie, Pennsylvania, known as Presque Isle Bay (the 43rd Great Lakes Area of Concern), has been designated by the United States Environmental Protection Agency as being in a stage of recovery. The impaired beneficial uses include the high incidence of fish tumors and restrictions on dredging. The goal of this study was to assess the level of chemical contamination in Bay sediments, and to some extent its tributaries, and to compare the findings to prior studies. Parameters investigated were the previously identified contaminants of concern: selected heavy metals and polycyclic aromatic hydrocarbons. The acid volatile sulfide/simultaneously extracted metals ratio was used to estimate bioavailability of metals.

It appeared that the deepest sediments and the most recent sediments were less contaminated with heavy metals and polycyclic aromatic hydrocarbons than those sediments in the middle region studied, that is, from about 10 cm below the surface to about 30 cm below the surface. These more highly contaminated sediments are estimated to represent deposition during the second half of the 20th century.

The most recent results for the most superficial sediments are encouraging for stakeholders because there were fewer cases of heavy metals exceeding their respective probable effects concentrations. The metals of greatest concern continued to be Cd, Ni and Pb, in that order. The sites most contaminated with heavy metals were those off Cascade Creek and the site least contaminated with heavy metals was the one nearest the mouth of Mill Creek. Data on stream sediments suggest that more work must be done within the watershed, especially the Cascade Creek and Myrtle Street drainages, to reduce the level of metal contamination reaching the Bay via sediment particles. In the most recently studied sediments certain metals continued to exist at levels approaching or exceeding their probable effects concentrations. However, the acid volatile sulfide/simultaneously extracted metals hypothesis suggests that those metals are not bioavailable.

The pattern of polycyclic aromatic hydrocarbon contamination is similar to that of the metals, with the most highly contaminated sites being those along the city's waterfront and in the center of the Bay. The hydrocarbons measured in the most recent sediments were all lower than those measured in 1994, and were well below the probable effects concentration for total polycyclic aromatic hydrocarbons. While there may be other important factors not studied or reviewed in this paper, these findings suggest that heavy metals and polycyclic aromatic hydrocarbons, as of 2003, likely exert only a moderate adverse impact on the aquatic ecosystem of Presque Isle Bay.

Introduction

The harbor at Erie, Pennsylvania, known as Presque Isle Bay (the Bay), is approximately 7.9 km long with a maximum width of 2.9 km. It has a surface area of 9187 ha, a volume of 52.5 × 106 m3, and a mean depth of 4.0 m (PADER, 1991). The Bay connects to Lake Erie through a narrow channel maintained by the U.S. Army Corps of Engineers. The Bay was designated as the 43rd Great Lakes Area of Concern in 1991. Identified impaired beneficial uses were the high incidence of tumors in Brown Bullheads (Ameiurus nebulosus), and restrictions on dredging. Numerous studies evaluated selected parameters at various times over the past 20 years (ACOE, 1982, 1986; Potomac-Hudson, 1991; PADER, 1992, 1993, 1996; Gannet Fleming, 1993; Batelle, 1994a, b, 1997). The Batelle (1994a) study employed the USEPA Great Lakes National Program Office's (GLNPO) Sediment Triad Approach (USEPA, 1994), in which coordinated chemical analysis, benthic macroinvertebrate community structure, and bioassays were used to evaluate sediments collected at the same time.

Important events occurred in Erie during the 1980s and 1990s. The city's bayfront began a transition from an industrial zone to a recreational zone. A former coal storage and shipping pier was converted to condominiums. Major ship-building and ship maintenance facilities were converted to marinas for recreational boats. At the beginning of the 1990s, a coal-fired electric power plant which had operated on the city's bayfront for many years was decommissioned and demolished, ending the discharge of water from fly and bottom ash settling ponds and cooling water to the Bay. In the late 1990s, a major combined sewer overflow abatement project was completed which essentially eliminated the discharge of sewage-contaminated stormwater to the Bay. As a result of these actions, and due to the findings of the aforementioned studies, upon the recommendation of the local Public Advisory Committee and the Pennsylvania Department of Environmental Protection, the US EPA declared in 2002 that the Bay was in a Recovery Stage, and mandated continued monitoring.

This paper presents results from data collection activities which were initiated in June, 2000, to determine if there had been changes subsequent to the Batelle study. In June 2000, grab samples and core samples were collected. In June 2003, only grab samples were collected. Associated measurements were made in Bay tributaries between June 2000 and August 2003. Reference sites included open water Lake Erie sites, ‘outer harbor’ sites located just outside of the channel, and an inland lake (Canadohta Lake). Physical and chemical factors are discussed and patterns of contamination are examined.

Methods

A quality assurance project plan (QAPP) was developed and approved by GLNPO prior to sample collection activities in June 2000. The Bay sites selected for study (Figure 1) were a sub-set of those previously studied (Gannett Fleming, 1993; Batelle, 1994a) which exhibited either high concentrations of contaminants or toxicity in bioassays. Contaminants of concern were identified as selected heavy metals (cadmium, copper, lead, nickel, and zinc) and polycyclic aromatic hydrocarbons (PAHs), based on those previous studies. June 2000 grab samples of surficial sediments were obtained by Ponar dredge. Samples for acid volatile sulfide/simultaneously extracted metals (AVS/SEM) were carefully removed from grab samples without mixing and without long exposure to air. Otherwise, material from each Ponar grab was homogenized before sample collection. In June 2000, sediment cores (∼ 1.0 m) were obtained at each site by the vibratory coring apparatus mounted on the USEPA's R/V Mudpuppy. The cores were divided immediately into two layers, the upper 30 cm (top layer), and the next 60 cm (bottom layer). Material in each layer was homogenized before samples were placed in appropriate bottles.

Figure 1.

Sites selected for sampling during the June 2000, and June 2003 studies.

Figure 1.

Sites selected for sampling during the June 2000, and June 2003 studies.

During the summer of 2001, stream bed samples were collected from pool areas containing fine-grained sediments along selected reaches of Cascade Creek, Mill Creek, the Myrtle Street storm sewer outfall, and Scott Run, whose mouth is located at the extreme western end of the Bay. Samples were analyzed for heavy metals using the same analytical methods as for the Bay samples. A minimum of four samples were collected from each reach and analyzed separately for heavy metals. On one occasion, two samples of suspended sediment (SS) were obtained from storm runoff collected by an automated sampler installed in Cascade Creek, the second largest tributary to Presque Isle Bay, and were analyzed for heavy metals. In accordance with the QAPP, various replicates, duplicates, field blanks, and so on, were analyzed.

Bay sediment samples were evaluated for particle size distribution, total organic carbon (TOC), and moisture content (USEPA and USACE, 1998), and were digested using EPA SW-846 Method 3050b for total recoverable metals. For AVS/SEM, sediments were analyzed according to the method of Allen et al. (1993). After digestion in either case, digestate solutions were quantified by direct aspiration atomic absorption spectrometry. In June 2003, samples were collected by Petite Ponar dredge. No cores were collected. Samples were shipped on ice by overnight freight to be analyzed by the US EPA Region III laboratory in Fort Meade, MD for physical factors as well as for metals and PAHs. The digestion method for metals was EPA Method 200.2, followed by analysis with ICP-AES or ICP-MS; PAHs were quantified by GC-MS.

Results and discussion

Age of sediments

A prior study (Batelle, 1995) used 210Pb dating to arrive at an average sediment accumulation rate in Presque Isle Bay of 0.92 cm yr−1, with a range of 0.87 to 1.01 cm yr−1. The mixing zone was confirmed to be about 10 cm. Sedimentation rates certainly vary from place to place around the Bay, and have probably varied over time. However, rough age estimates can be made using this value. Since the Ponar dredge typically penetrated approximately 15 cm into the bottom, the grab samples represented sediments with an average age of about 8 years. The top core layer roughly represented sediments ranging from 0 to 33 years of age, with an average age of 16 years, while the bottom layer represented the period from 33 to 98 years before collection, with an average age of 65 years.

Physical and aggregate parameters

The sediments collected in 2000 and 2003 were generally black or dark brown in appearance, pudding-like and sticky in texture. Some grab samples had a reddish-brown surface coating, presumed to be oxidized iron hydroxides. Some samples had a mild sulfide odor, but generally did not have an odor of petroleum. Oily films were not observed on or in the samples. The moisture content of samples was generally lower in the deeper samples. Grab samples had average moisture contents of 68%, while core samples had average moisture contents of 57% (top) and 41% (bottom).

The size categories used for this study were sand (2.0 mm to 0.05 mm), silt (0.05 mm to 0.002 mm), and clay (< 0.002 mm). Zebra mussel shells were found in a few of the locations. In order not to distort the size distribution, zebra mussel shells were excluded from the particle size distribution analysis. Sand was increasingly common for deeper samples. Grab samples averaged 16.5% sand, while the top core layer averaged 20% sand and the bottom core layer averaged 28% sand. Silt was the most abundant size category, ranging from a mean of 42.8% (grab) to 49.6% (bottom layer). Clay-sized particles made up 40.8% of the grab samples, 35.6% of the top core layer, and 22% of the bottom layer.

Total organic carbon (TOC) content of the June 2000 samples was generally lower in the deeper samples. Grab samples ranged in value from 2.6 to 5.6% TOC dry weight (DW), with a mean of 3.8%, while top core layer samples had a mean TOC of 2.5%, with a range of 0.8 to 4.3% and bottom layer samples had a mean of 1.1%, with a range of 0.3 to 3.5%.

Samples collected in June 2003 by Petite Ponar had a higher moisture content, a higher TOC, and a higher fraction of sand and silt than the grab samples in June 2000. None of these differences were statistically significant, but they do suggest that the change in sampling device had an effect on the nature of the sediments collected.

Heavy metals

Interpretation of sediment chemistry results is complex (USEPA, 2002a). The US EPA has not yet published a definitive set of sediment quality guidelines. Because of a proliferation of effects levels defined by various investigators, Ingersoll et al. (2000) and MacDonald et al. (2000) reviewed the relevant studies and published what they referred to as ‘consensus-based’ effects levels. They developed two levels: threshold effects concentration (TEC), that is, ‘concentration below which harmful effects are unlikely to be observed,’ and a probable effects concentration (PEC), that is, ‘concentration above which harmful effects are likely to be observed.’ Table 1 presents the reference levels relevant to this study.

Consensus-based toxic-effects levels for contaminated sediments.

Table 1.
Consensus-based toxic-effects levels for contaminated sediments.
 Consensus-based effects concentrations 
Analyte (mg kg−1 DW) TEC PEC 
Cadmium 0.99 4.98 
Copper 31.6 149 
Lead 35.8 128 
Nickel 22.7 48.6 
Zinc 121 459 
Total PAHs 1.61 22.8 
Source: MacDonald et al., 2000
 Consensus-based effects concentrations 
Analyte (mg kg−1 DW) TEC PEC 
Cadmium 0.99 4.98 
Copper 31.6 149 
Lead 35.8 128 
Nickel 22.7 48.6 
Zinc 121 459 
Total PAHs 1.61 22.8 
Source: MacDonald et al., 2000

In the June 2000 dataset, metals of interest were present at levels in excess of the TECs in all grab and top core layer samples from every site and in many cases were present at concentrations above the PECs (Table 2). There were instances of metals being present below the TECs in the bottom core layers for Cd, Cu and Pb, but not for Ni or Zn. The metals concentrations in the grab sample from the open water Lake Erie reference site were above the TECs as well, but did not exceed the PEC for any metal. Copper was not found above the PEC in any samples at any site. On the other hand, Cd exceeded the PEC in 16 of the 30 samples; Pb in 13 samples; Ni in 23 samples, and Zn in 19 samples. The number of exceedences was greater for the surficial samples compared to the deeper samples. Thus, younger sediments were more contaminated than older sediments. Based purely on the number of PEC exceedences, the most contaminated sites were in the central Bay (PIB07 and PIB15), with 10 exceedences each. The least contaminated sites using this metric were sites at the extreme ends of the Bay (PIB01 and PIB25) with 4 and 3 exceedences respectively. PIB01 is the site most remote from present or past industrial activity.

Total extractable metals analysis results for June, 2000, Presque Isle Bay sediments and an open water Lake Erie sample.

Table 2.
Total extractable metals analysis results for June, 2000, Presque Isle Bay sediments and an open water Lake Erie sample.
  Total extractable metals (mg kg−1 dry sediment)  
 Open Lakea PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 PIB 25 Mean Number of PEC exceedences 
Cadmium Grab 2.2 6.1 8.5 8.9 8.0 9.0 7.4 6.7 5.0 8.1 5.6 7.3 10 
TEC = 0.99 Top layer  3.9 3.8 9.3 6.2 7.0 4.8 9.4 4.6 7.1 1.0 5.7 
PEC = 4.98 Bottom layer  0.8b < 0.2b 4.3 < 0.2b < 0.2b 2.2 3.1 7.3 < 0.2b < 0.2b 1.8c 
Copper Grab 56 97 127 104 101 116 82 82 60 73 84 93 
TEC = 31.6 Top layer  72 57 116 93 104 78 100 61 80 29 79 
PEC = 149 Bottom layer  30b 26b 76 35 38 83 85 64 29b 25b 49 
Lead Grab 39 116 192 166 157 227 133 194 107 138 127 156 
TEC = 35.8 Top layer  83 71 166 178 189 116 154 95 158 50 126 
PEC = 128 Bottom layer  20b 20b 69 44 42 75 4b 151 35b 22b 48 
Nickel Grab 46 71 88 84 83 84 69 71 56 100 78 78 10 
TEC = 22.7 Top layer  59 89 80 68 68 51 68 52 89 43 67 
PEC = 48.6 Bottom layer  36 37 106 36 35 48 63 100 38 36 53 
Zinc Grab 250 600 862 739 729 747 679 602 434 697 607 670 10 
TEC = 121 Top layer  457 383 776 685 626 451 664 414 684 216 536 
PEC = 459 Bottom layer  187 183 585 247 266 470 517 582 216 180 343 
Number of PEC exceedences  10 10  
Note: bold numbers indicate PEC exceedence; site numbers correspond to the Batelle and the Gannet Fleming studies. 

a Fine-grained open lake sample obtained ∼ 7 km offshore at ∼ 18 m depth.

 

b Below TEC.

 

cMean computed by using one-half the detection limit for values below detection limit.

 
< Indicates below detection limit. 
  Total extractable metals (mg kg−1 dry sediment)  
 Open Lakea PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 PIB 25 Mean Number of PEC exceedences 
Cadmium Grab 2.2 6.1 8.5 8.9 8.0 9.0 7.4 6.7 5.0 8.1 5.6 7.3 10 
TEC = 0.99 Top layer  3.9 3.8 9.3 6.2 7.0 4.8 9.4 4.6 7.1 1.0 5.7 
PEC = 4.98 Bottom layer  0.8b < 0.2b 4.3 < 0.2b < 0.2b 2.2 3.1 7.3 < 0.2b < 0.2b 1.8c 
Copper Grab 56 97 127 104 101 116 82 82 60 73 84 93 
TEC = 31.6 Top layer  72 57 116 93 104 78 100 61 80 29 79 
PEC = 149 Bottom layer  30b 26b 76 35 38 83 85 64 29b 25b 49 
Lead Grab 39 116 192 166 157 227 133 194 107 138 127 156 
TEC = 35.8 Top layer  83 71 166 178 189 116 154 95 158 50 126 
PEC = 128 Bottom layer  20b 20b 69 44 42 75 4b 151 35b 22b 48 
Nickel Grab 46 71 88 84 83 84 69 71 56 100 78 78 10 
TEC = 22.7 Top layer  59 89 80 68 68 51 68 52 89 43 67 
PEC = 48.6 Bottom layer  36 37 106 36 35 48 63 100 38 36 53 
Zinc Grab 250 600 862 739 729 747 679 602 434 697 607 670 10 
TEC = 121 Top layer  457 383 776 685 626 451 664 414 684 216 536 
PEC = 459 Bottom layer  187 183 585 247 266 470 517 582 216 180 343 
Number of PEC exceedences  10 10  
Note: bold numbers indicate PEC exceedence; site numbers correspond to the Batelle and the Gannet Fleming studies. 

a Fine-grained open lake sample obtained ∼ 7 km offshore at ∼ 18 m depth.

 

b Below TEC.

 

cMean computed by using one-half the detection limit for values below detection limit.

 
< Indicates below detection limit. 

Statistical analysis was performed on the metals data using ANOVA and T-test with significance set at the α = 0.05 level. There was no significant difference among the sites when consolidating all metals in all types of samples, nor was there any significant difference among sites when consolidating all metals in each type of sample (grab, top layer, bottom layer) separately. Similarly, there was no significant difference among the sites for any of the metals separately. There was no significant difference between the grab samples and the top layer samples, but there was a significant difference between top layer samples and bottom layer samples (P < 0.05, n = 100, df = 98). There was a highly significant difference between the grab samples and bottom samples (P < 0.008, n = 100, df = 98). This confirms the observation made above using the number of PEC exceedences and suggests that either the importation of these toxic metals has changed over the last hundred years, or that there is a geochemical transport of metals from deeper sediments toward the surface.

A similar pattern of metal contamination existed in the June 2003 grab samples as was found in the 2000 study, except that these sediments were generally less contaminated (Table 3). There were fewer PEC exceedences when compared with the grab samples from 2000, and one site (PIB18, near the mouth of Mill Creek) was slightly above or below the TEC for all metals. The metals with the greatest number of exceedences were Cd, Ni and Pb, in that order. Zinc concentrations were significantly lower in the 2003 samples compared to the June, 2000, grab samples (p < 0.05). A reference site (Canadohta Lake, Crawford County, PA, an inland lake) had no PEC exceedences, and all metals but Cd were below the TECs.

Total extractable metals results for the June 2003, Presque Isle Bay sediment samples.

Table 3.
Total extractable metals results for the June 2003, Presque Isle Bay sediment samples.
  Total extractable metal inPresque Isle Bay sediment (mg kg−1 DW)  
 Canadohta Lake PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 PIB 25 Mean Number of PEC exceedences 
Cadmium 1.7 6.0 9.2 8.6 6.6 7.4 6.2 6.8 0.9a 5.4 4.2 6.1 
Copper 26.4a 82.7 113 102 81.3 105 82.4 73.9 16.3a 64.1 76.3 79.7 
Lead 29.4a 89.1 155 138 96.8 152 104 93.3 37.1 58.1 80.6 100 
Nickel 20.9a 50.3 66.1 66.4 51.8 59.3 46.5 54.1 13.8a 39.6 45.7 49.4 
Zinc 117a 282 399 370 272 368 291 290 73.3a 230 274 285 
Number of PEC exceedences  
Note: bold type indicates an exceedence of the PEC. 

aIndicates value below TEC.

 
  Total extractable metal inPresque Isle Bay sediment (mg kg−1 DW)  
 Canadohta Lake PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 PIB 25 Mean Number of PEC exceedences 
Cadmium 1.7 6.0 9.2 8.6 6.6 7.4 6.2 6.8 0.9a 5.4 4.2 6.1 
Copper 26.4a 82.7 113 102 81.3 105 82.4 73.9 16.3a 64.1 76.3 79.7 
Lead 29.4a 89.1 155 138 96.8 152 104 93.3 37.1 58.1 80.6 100 
Nickel 20.9a 50.3 66.1 66.4 51.8 59.3 46.5 54.1 13.8a 39.6 45.7 49.4 
Zinc 117a 282 399 370 272 368 291 290 73.3a 230 274 285 
Number of PEC exceedences  
Note: bold type indicates an exceedence of the PEC. 

aIndicates value below TEC.

 

Heavy metals in surface sediments were studied in Presque Isle Bay sediments by contractors for the U.S. Army Corps of Engineers in 1982 (ACOE, 1982) and 1986 (ACOE, 1986). The same locations were sampled in both of those studies. While none of the sites were the same as those used in this study, eight (8) of the sites were within Presque Isle Bay and in the same general area as those used in the current study. In these two studies (Table 4), all metals were present in the Bay at concentrations in excess of the TECs, and Cd, Pb and Ni were in excess of PECs.

Metals concentrations found during sediment assessments in 1982 and 1986.

Table 4.
Metals concentrations found during sediment assessments in 1982 and 1986.
  Metal concentration (mg kg−1 DW)  
ACOE site number 10 11 12 13 14 PIB Mean Lake Eriea Number of PEC exceedences 
Cadmium            
    1982 3.3 7.9 9.0 6.5 12.0 6.3 9.4 11.2 8.2 3.1 
    1986 3.2 5.2 8.0 6.1 6.3 4.8 7.4 7.8 6.1 0.8b 
Copper            
    1982 34 60 80 63 88 53 76 94 68.4 39 
    1986 35 54 78 70 73 68 81 98 69.6 17b 
Lead            
    1982 51.2 132 149 188 180 123 167 195 148 67 
    1986 49 68 120 97 96 94 120 140 98.0 16b 
Nickel            
    1982 27 56 31 117 57 48 57 64 57.0 38 
    1986 27 39 46 55 49 35 47 51 43.6 20b 
Zinc            
    1982 194 352 301 289 320 218 318 364 294 184 
    1986 200 260 340 360 300 220 290 370 293 84b 
Number of PEC exceedences   
Source: ACOE, 1982, and ACOE, 1986
Note: bold numbers indicate PEC exceedence; site numbers do not correspond to the current study or FWS site numbers. 

aReference sample from open Lake Erie, location not specified

 

bIndicates value below the TEC.

 
  Metal concentration (mg kg−1 DW)  
ACOE site number 10 11 12 13 14 PIB Mean Lake Eriea Number of PEC exceedences 
Cadmium            
    1982 3.3 7.9 9.0 6.5 12.0 6.3 9.4 11.2 8.2 3.1 
    1986 3.2 5.2 8.0 6.1 6.3 4.8 7.4 7.8 6.1 0.8b 
Copper            
    1982 34 60 80 63 88 53 76 94 68.4 39 
    1986 35 54 78 70 73 68 81 98 69.6 17b 
Lead            
    1982 51.2 132 149 188 180 123 167 195 148 67 
    1986 49 68 120 97 96 94 120 140 98.0 16b 
Nickel            
    1982 27 56 31 117 57 48 57 64 57.0 38 
    1986 27 39 46 55 49 35 47 51 43.6 20b 
Zinc            
    1982 194 352 301 289 320 218 318 364 294 184 
    1986 200 260 340 360 300 220 290 370 293 84b 
Number of PEC exceedences   
Source: ACOE, 1982, and ACOE, 1986
Note: bold numbers indicate PEC exceedence; site numbers do not correspond to the current study or FWS site numbers. 

aReference sample from open Lake Erie, location not specified

 

bIndicates value below the TEC.

 

In 1991, the U.S. Fish and Wildlife Service studied surficial sediments (FWS, 1991), but used still different sites. In this case, ten of the sites were within Presque Isle Bay and in the same general area as those used in the current study. In this study (Table 5), all metals were present in the Bay at concentrations in excess of the TECs, with all metals exceeding their respective PECs at multiple sites.

Metals concentrations found during sediment assessment in 1990.

Table 5.
Metals concentrations found during sediment assessment in 1990.
 Metal concentration (mg kg−1 DW)    
FWS site number 10 11 12 14 15 PIB Mean Outer Harbora Number of PEC exceedences 
Cadmium 10.6 10.3 6.3 6.8 3.2 1.7 3.8 3.4 8.9 11.5 6.6 2.2 
Copper 214 152 86 109 81 92 225 107 183 183 143.2 48 
Lead 163 159 86 118 94 129 143 110 223 224 144.9 24b 
Nickel 73 81 58 58 47 48 61 73 105 113 71.7 30 
Zinc 573 470 398 355 294 324 490 385 683 684 465.6 162 
Number of PEC exceedences  
Source: FWS, 1991. 
Note: bold numbers indicate PEC exsceedences; site numbers do not correspond to either the current study or ACOE site numbers. 

aMean of 3 FWS reference samples from outside the Erie Harbor.

 

bIndicates value below the TEC.

 
 Metal concentration (mg kg−1 DW)    
FWS site number 10 11 12 14 15 PIB Mean Outer Harbora Number of PEC exceedences 
Cadmium 10.6 10.3 6.3 6.8 3.2 1.7 3.8 3.4 8.9 11.5 6.6 2.2 
Copper 214 152 86 109 81 92 225 107 183 183 143.2 48 
Lead 163 159 86 118 94 129 143 110 223 224 144.9 24b 
Nickel 73 81 58 58 47 48 61 73 105 113 71.7 30 
Zinc 573 470 398 355 294 324 490 385 683 684 465.6 162 
Number of PEC exceedences  
Source: FWS, 1991. 
Note: bold numbers indicate PEC exsceedences; site numbers do not correspond to either the current study or ACOE site numbers. 

aMean of 3 FWS reference samples from outside the Erie Harbor.

 

bIndicates value below the TEC.

 

Heavy metals were measured by the Gannett Fleming study in 1993, which established the sampling sites used today. The original document was not available for this review, but tables from that study were included in the 1995 RAP update (PADER, 1995). The data (Table 6) showed that Cd exceeded the PEC at all the sites of interest, while Pb exceeded the PEC at 5 sites, and Ni was found in excess of its PEC at 7 sites. Copper and Zn were found to be below their PECs. Sites mostly along the city's waterfront and along the centerline of the Bay were the most contaminated with 3 exceedences each. The least contaminated sites were PIB01 (at the western end of the Bay) and PIB18 (located near the mouth of Mill Creek). The mean concentrations for the metals from the outer harbor sites were all below the TECs, except for Cd, which was slightly above its TEC.

Metals concentrations found during Gannet Fleming sediment assessment in 1993.

Table 6.
Metals concentrations found during Gannet Fleming sediment assessment in 1993.
 Metals concentrations (mg kg−1 DW)  
 PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 PIB Mean OHa Number of PEC exceedences 
Cadmium 6.1 9.5 10.0 10.0 7.7 9.4 7.9 6.0 7.9 8.3 0.7 
Copper 76.7 112 104 97.9 104 104 79.8 75.4 78.7 92.5 13.5b 
Lead 92.0 170 150 170 170 130 92.0 99.0 120 133 7.3b 
Nickel 47.5 58.7 53.9 62.5 52.1 55.0 49.7 36.7 49.2 51.7 11.8b 
Zinc 278 428 398 381 399 377 340 264 341 356 64.7b 
Number of PEC exceedences   
Source: PADER, 1995
Note: bold numbers indicate PEC exceedence; site numbers correspond to the Batelle study and the current study. 

aOH is the mean of samples from 3 sites outside the Erie Harbor.

 

bIndicates value below the TEC.

 
 Metals concentrations (mg kg−1 DW)  
 PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 PIB Mean OHa Number of PEC exceedences 
Cadmium 6.1 9.5 10.0 10.0 7.7 9.4 7.9 6.0 7.9 8.3 0.7 
Copper 76.7 112 104 97.9 104 104 79.8 75.4 78.7 92.5 13.5b 
Lead 92.0 170 150 170 170 130 92.0 99.0 120 133 7.3b 
Nickel 47.5 58.7 53.9 62.5 52.1 55.0 49.7 36.7 49.2 51.7 11.8b 
Zinc 278 428 398 381 399 377 340 264 341 356 64.7b 
Number of PEC exceedences   
Source: PADER, 1995
Note: bold numbers indicate PEC exceedence; site numbers correspond to the Batelle study and the current study. 

aOH is the mean of samples from 3 sites outside the Erie Harbor.

 

bIndicates value below the TEC.

 

Prudence must be used when comparing the results of these studies since different sites and/or different collection and analytical methods were used. With that reservation in mind, mean values for grab samples for the Bay and the reference sites for each metal of concern were computed for each study and are presented in Figure 2. The reference samples from Lake Erie and the outer harbor consistently had lower concentrations than the Presque Isle Bay samples. Every metal showed a decline in the 2003 samples. Since this may be an artifact of the collection method, the 2003 results were omitted from the data set and new means were computed. When those new mean values were regressed against time for each metal, the Cd concentration regression line had a slope of essentially zero, implying that there has been no change in Cd concentrations over this time period. Regression lines for Cu, Pb and Ni had slightly positive slopes, while the regression for Zn showed a marked positive slope of 20.5 yr−1. Not surprisingly, ‘year’ explained virtually none of the variation for Cd, Cu and Pb, with coefficients of determination (R2) of 0.00, 0.16, and 0.16, respectively. However, for Ni, ‘year’ explained 44% of the variation, and for Zn, ‘year’ explained 79% of the variation. If this comparison of studies over the 18 years from 1982 to 2000 is legitimate, and if the decline in concentrations in the June 2003 study truly represents the most recent sediments, then perhaps there has been a real change in the previous trend in recent times.

New sediments are added to the Bay from tributaries, storm sewer discharges, and atmospheric deposition. Samples were collected from tributaries and the outfall area of one storm sewer. Metals in a SS sample from Cascade Creek were found to exceed the PEC for all metals but copper. Bed sediments from Cascade Creek and the Myrtle Street storm sewer outfall were found to have higher levels of all metals than were found in Mill Creek and Scott Run (data available but not shown). Mill Creek is the largest single tributary to Presque Isle Bay. Its bed sediments had lower concentrations of metals than the other sources, except for Scott Run, which has a small residential watershed.

Figure 2.

Comparison of various Presque Isle Bay sediment assessments; Lake Erie and Outer Harbor are reference sites from various studies and are outside of the confines of Presque Isle Bay.

Figure 2.

Comparison of various Presque Isle Bay sediment assessments; Lake Erie and Outer Harbor are reference sites from various studies and are outside of the confines of Presque Isle Bay.

Acid volatile sulfide/simultaneously extracted metals

The AVS/SEM method is thought to estimate the bioavailability of toxic metals (Hansen et al., 1996). The solubility product constants of the sulfide minerals of the divalent cations of concern here (Cd, Cu, Pb, Ni and Zn) are lower than that of iron sulfide (Bard, 1966). Thus the pore water concentrations of these toxic heavy metals in equilibrium with their respective sulfide minerals are likely to be low, that is, below the toxic effect threshold of most sediment dwelling organisms (DiToro et al., 1992). The analytical protocol (Allen et al., 1993) for measuring this ratio which involves a mild acid digestion has not been endorsed in final form by the EPA. The hypothesis suggests that if the ratio of the sum total of the molar concentrations of the SEM to the molar concentration of AVS is less than one, the probability of toxicity is low. In this study, the SEM:AVS ratio was below one for all sites, and below 0.5 in most cases. If this hypothesis is correct, then even though there are numerous exceedences of the PECs as discussed previously, the metals may not be bioavailable due to the high availability of sulfide in Presque Isle Bay sediments. The US EPA has written a draft document (USEPA, 2002b) endorsing this pore water equilibrium concentration approach to toxicity estimation, but has not yet released it as of this writing in final form to the public.

Polycyclic aromatic hydrocarbons

Sixteen PAHs had been targeted by the US EPA as priority pollutants, with the combined concentrations of these compounds in a sample referred to as PAH16. Recently, in a new equilibrium-based approach to estimating toxicity due to PAH mixtures, the US EPA has advocated the quantification of 34 PAHs (USEPA, 2002c). Here, the TEC and PEC for ‘total PAHs’ of 1610 and 22800 μ g kg−1 DW, respectively, were based on summing the concentrations of 13 (of the 16) PAHs (Ingersoll et al., 2000). Clearly, prudence must be exercised when comparing the results presented here against a single-number effects level to estimate adverse effect.

Gannett Fleming (1993) measured PAHs. The results were reported as ‘total PAHs.’ The findings are given in Table 7. The two most contaminated sites were at the mouths of the Bay's largest two tributaries: Mill Creek (PIB18) and Cascade Creek (PIB09). Only the Mill Creek location, with a total PAH concentration of 29760 μ g kg−1 DW, exceeded the PEC.

Values for total PAHs from the 1993 Gannett Fleming study of Presque Isle Bay sediments.

Table 7.
Values for total PAHs from the 1993 Gannett Fleming study of Presque Isle Bay sediments.
 Concentration (ug kg−1 DW) 
 PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 PIB Mean OH1 
Total PAHs 2,021 12,080 3,950 3,260 18,207 3,240 10,780 29,760 no data 10,412 658 
Source: PADER (1995). 
Note: PEC for total PAH is 22,800 ug kg−1 DW; bold value indicates PEC exceedence. 
1OH is the mean of samples from 3 sites outside the Erie Harbor. 
 Concentration (ug kg−1 DW) 
 PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 PIB Mean OH1 
Total PAHs 2,021 12,080 3,950 3,260 18,207 3,240 10,780 29,760 no data 10,412 658 
Source: PADER (1995). 
Note: PEC for total PAH is 22,800 ug kg−1 DW; bold value indicates PEC exceedence. 
1OH is the mean of samples from 3 sites outside the Erie Harbor. 

Batelle (1994a) also investigated PAHs. A detailed and comprehensive analysis of those results are reported in Batelle (1997). Both PAH16 and total PAHs were reported. Batelle found that total PAH concentrations ranged from 1690 to 40650 μ g kg−1 DW in grab samples, and from 310 to 57990 μ g kg−1 DW in cores. Data for the grab samples at the sites selected for this study are presented in Table 8. There was only one site (PIB18) with a PAH16 concentration below the TEC, and PIB09 had a PAH16 concentration exceeding the PEC. Batelle concluded that most of the PAHs found in Presque Isle Bay sediments were of pyrogenic nature.

PAH concentrations in Presque Isle Bay sediments collected in June, 1994.

Table 8.
PAH concentrations in Presque Isle Bay sediments collected in June, 1994.
 PAHs in surface ssediments (ug kg−1 dry sediment) 
 PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 Mean 
Acenaphthene 64 96 88 81 181 114 88 27 82.8 
Acenaththylene 64 112 109 142 113 142 105 64 95.2 
Anthracene 137 315 248 199 408 286 248 20 92 217 
Benzo(a) pyrene 714 1353 1296 1087 2041 1439 1117 96 440 1065 
Benzo(a)anthracene 615 1152 1066 879 1937 1286 1003 91 393 936 
Benzo(b)fluoranthen 1282 2362 2176 1942 3351 2377 1921 153 804 1819 
Benzo(g,h,i)perylene 616 1164 1101 970 2296 1105 830 71 398 950 
Benzo(k)fluoranthen 361 741 594 579 784 639 500 50 238 498 
Chrysene 777 1358 1238 1054 1996 1423 1158 96 508 1068 
Dibenzo(a,h) anthracene 169 353 312 269 999 329 251 21 118 313 
Fluoranthene 1240 2490 2666 1740 3668 2158 2192 171 674 1889 
Fluorene 118 171 163 131 215 166 135 56 129 
Indeno(1,2,3-cd) pyrene 736 1373 1342 1150 2757 1348 1016 86 457 1141 
Naphthalene 124 230 216 216 262 342 228 128 194 
Phenanthrene 652 1087 859 792 1830 1162 910 79 335 856 
Pyrene 1287 2412 3215 1935 3440 2413 1985 181 751 1958 
PAH16 8,956 16,769 16,689 13,166 26,278 16,729 13,687 1,138 5,483 13,211 
Total PAH 15,010 29,150 26,860 22,680 40,650 27,490 22,490 1,690 9,970 21,777 
Source: Batelle, 1994. 
Note: TEC for total PAH is 1,610; PEC for total PAH is 22,800 ug kg−1 DW; bold indicates PEC exceedence. 
 PAHs in surface ssediments (ug kg−1 dry sediment) 
 PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 Mean 
Acenaphthene 64 96 88 81 181 114 88 27 82.8 
Acenaththylene 64 112 109 142 113 142 105 64 95.2 
Anthracene 137 315 248 199 408 286 248 20 92 217 
Benzo(a) pyrene 714 1353 1296 1087 2041 1439 1117 96 440 1065 
Benzo(a)anthracene 615 1152 1066 879 1937 1286 1003 91 393 936 
Benzo(b)fluoranthen 1282 2362 2176 1942 3351 2377 1921 153 804 1819 
Benzo(g,h,i)perylene 616 1164 1101 970 2296 1105 830 71 398 950 
Benzo(k)fluoranthen 361 741 594 579 784 639 500 50 238 498 
Chrysene 777 1358 1238 1054 1996 1423 1158 96 508 1068 
Dibenzo(a,h) anthracene 169 353 312 269 999 329 251 21 118 313 
Fluoranthene 1240 2490 2666 1740 3668 2158 2192 171 674 1889 
Fluorene 118 171 163 131 215 166 135 56 129 
Indeno(1,2,3-cd) pyrene 736 1373 1342 1150 2757 1348 1016 86 457 1141 
Naphthalene 124 230 216 216 262 342 228 128 194 
Phenanthrene 652 1087 859 792 1830 1162 910 79 335 856 
Pyrene 1287 2412 3215 1935 3440 2413 1985 181 751 1958 
PAH16 8,956 16,769 16,689 13,166 26,278 16,729 13,687 1,138 5,483 13,211 
Total PAH 15,010 29,150 26,860 22,680 40,650 27,490 22,490 1,690 9,970 21,777 
Source: Batelle, 1994. 
Note: TEC for total PAH is 1,610; PEC for total PAH is 22,800 ug kg−1 DW; bold indicates PEC exceedence. 

PAHs in Petite Ponar grab samples were evaluated in the June 2003 study. Concentrations for PAH16 ranged from 476 to 4692 μ g kg−1 DW(Table 9). The pattern of PAH concentrations was similar when comparing the 1993 and 1994 studies with the 2003 study, but the mean for the selected sites was significantly lower in 2003 than the means for these sites in either of the two prior studies (P < 0.001).

PAH concentrations in Presque Isle Sediments collected in June, 2003.

Table 9.
PAH concentrations in Presque Isle Sediments collected in June, 2003.
  PAHs (ug kg−1 DW) 
 Canahdota Lake PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 PIB 25 Mean 
Acenaphthene 9.0 12 10 26  13 18 19 12.6 
Acenaththylene 18 13 22 21 32 48 26 13 36 22.4 
Anthracene 42 37 47 40 13 83 94 60 61 17 86 53.8 
Benzo(a)anthracene 41 101 159 143 49 320 295 180 201 47 281 178 
Benzo(a) pyrene 62 119 201 188 64 406 339 222 224 49 339 215 
Benzo(b)fluoranthen 29 195 341 269 99 675 478 320 333 79 511 330 
Benzo(g,h,i)perylene 35 112 177 166 62 337 244 182 149 40 254 172 
Benzo(k)fluoranthen 52 103 154 143 51 279 252 156 176 43 228 159 
Chrysene nd 124 182 168 60 392 324 217 250 57 349 212 
Dibenzo(a,h)anthracene 69 nd 55 46 nd 103 73 56 nd nd 66 66.5 
Fluoranthene nd 226 305 286 97 701 622 339 543 91 661 387 
Fluorene 35 25 31 23 48 59 30 34 nd 51 34.3 
Indeno(1,2,3-cd)pyrene 125 191 180 68 364 270 201 161 40 270 187 
Naphthalene 27 20 33 30 12 47 84 40 36 15 74 39.1 
Phenanthrene 56 115 141 125 45 287 264 152 264 44 286 172 
Pyrene nd 187 272 246 82 592 525 306 423 79 538 325 
PAH16 476 1,511 2,323 2,084 720 4,692 3,971 2,500 2,886 610 4,049 2,566 
nd = not detected; note: TEC for total PAH is 1,610; PEC is 22,800 ug kg−1 DW. 
  PAHs (ug kg−1 DW) 
 Canahdota Lake PIB 01 PIB 05 PIB 07 PIB 08 PIB 09 PIB 14 PIB 15 PIB 18 PIB 20 PIB 25 Mean 
Acenaphthene 9.0 12 10 26  13 18 19 12.6 
Acenaththylene 18 13 22 21 32 48 26 13 36 22.4 
Anthracene 42 37 47 40 13 83 94 60 61 17 86 53.8 
Benzo(a)anthracene 41 101 159 143 49 320 295 180 201 47 281 178 
Benzo(a) pyrene 62 119 201 188 64 406 339 222 224 49 339 215 
Benzo(b)fluoranthen 29 195 341 269 99 675 478 320 333 79 511 330 
Benzo(g,h,i)perylene 35 112 177 166 62 337 244 182 149 40 254 172 
Benzo(k)fluoranthen 52 103 154 143 51 279 252 156 176 43 228 159 
Chrysene nd 124 182 168 60 392 324 217 250 57 349 212 
Dibenzo(a,h)anthracene 69 nd 55 46 nd 103 73 56 nd nd 66 66.5 
Fluoranthene nd 226 305 286 97 701 622 339 543 91 661 387 
Fluorene 35 25 31 23 48 59 30 34 nd 51 34.3 
Indeno(1,2,3-cd)pyrene 125 191 180 68 364 270 201 161 40 270 187 
Naphthalene 27 20 33 30 12 47 84 40 36 15 74 39.1 
Phenanthrene 56 115 141 125 45 287 264 152 264 44 286 172 
Pyrene nd 187 272 246 82 592 525 306 423 79 538 325 
PAH16 476 1,511 2,323 2,084 720 4,692 3,971 2,500 2,886 610 4,049 2,566 
nd = not detected; note: TEC for total PAH is 1,610; PEC is 22,800 ug kg−1 DW. 

Assuming that the results reported here are not simply artifacts of analytical or collection methods, the most surficial sediments were less contaminated with PAHs in 2003 than the sediments sampled in 1993 and 1994.

Acknowledgments

The results reported here were made possible by grants from a number of sources. The principle source of funding was the USEPA (GLNPO Project No. GL97504701-01-0). Additional support was provided by the Pennsylvania Department of Environmental Protection's Growing Greener Program and the Great Lakes Commission through a grant to the Erie County Conservation District. Assistance and support was provided as well by Mr. Robert Wellington of the Erie County Department of Health, and Ms. Lori Boughton, Chief of Great Lakes, Pennsylvania Department of Environmental Protection, and numerous undergraduate and graduate students at Gannon University.

References

ACOE (Army Corps of Engineers)
.
1982
.
Chemical, physical and bioassay analysis of sediment samples, Erie Harbor, Erie, Pennsylvania
,
Prepared for the U.S. ACOE
Buffalo District by Applied Biology, Inc.
.
ACOE
.
1986
.
The analysis of sediments from Erie Harbor, Erie, PA
,
Prepared for the U.S. ACOE, Buffalo District by AquaTech Environmental Consultants, Inc. (Technical Report #G0176-07; Contract #DACW49-86-D-001)
Allen, H. E., Fu, G. and Deng, B.
1993
.
Analysis of acid-volatile sulfide (AVS) and simultaneously extracted metals (SEM) for the estimation of potential toxicity in aquatic sediments
.
Environ. Toxicol. Chem.
,
12
:
1
13
.
Bard, A. J.
1966
.
Chemical Equilibrium
,
New York
:
Harper and Row
.
Battelle
.
1994a
.
Presque Isle Bay sediment quality evaluation report for May 1994 study
,
Final Report prepared by D. E. West, H. Tulli, and J. Neff, Battelle Ocean Sciences, Duxbury, MA, for EPA Region III under Work Assignment 1-107. Contract No. 68-C2-0134. September 29, 1994.
Battelle
.
1994b
.
Evaluation of polycyclic aromatic hydrocarbons in Presque Isle Bay sediment cores for May 1994 Study
,
Data report prepared by D. E. West, Battelle Ocean Sciences, Duxbury, MA, for EPA Region III under Work Assignment 1-107. Contract No. 68-C2-0134. September 29, 1994.
Battelle
.
1995
.
Evaluation of Lead-210 in Presque Isle Bay sediment cores for May 1994 study
,
Data report prepared by V. Cullinan and E. Crecelius, Battelle Marine Science Laboratory, Sequim, WA, for EPA Region III under Work Assignment 2-107. Contract No. 68-C2-0134. August 30, 1995.
Battelle
.
1997
.
Presque Isle Bay sediment study-data review
,
Final report prepared by G. Durell and J. Neff, Battelle Ocean Sciences, Duxbury, MA, for EPA Region III under Work Assignment 4-456. Contract No. 68-C2-0134. April 30, 1997
DiToro, D. M., Mahony, J. D., Hansen, D. J., Scortt, K. J., Carlson, A. R. and Ankley, G. T.
1992
.
Acid volatile sulfide predicts the acute toxicity of cadmium and nickel in sediments
.
Environ. Sci. Technol.
,
26
:
96
101
.
FWS (United States Fish and Wildlife Service)
.
1991
.
Chemical analysis of sediments from Presque Isle Bay, Erie, Pennsylvania. U.S
,
Fish and Wildlife Service, State College, PA Field Office (C. Rice and C. Kulp). Special Project Report 91-2.
Gannet Fleming, Inc.
1993
.
Special study: Presque Isle Bay sediment quality evaluation
,
Prepared for U. S. Environmental Protection Agency Region III. Harrisburg, PA.-368.
Hansen, D. J., Berry, W. J., Mahony, J. D., Boothman, W. S., DiToro, D. M., Robson, D. L., Ankley, G. T., Ma, D., Yan, Q. and Pesch, C. E.
1996
.
Predicting the toxicity of metal-contaminated field sediments using interstitial concentrations of metals and acid-volatile sulfide normalizations
.
Environ. Toxicol. Chem.
,
15
(
12
):
2080
2094
.
Ingersoll, C. G., MacDonald, D. D., Wang, N., Crane, J. L., Field, L. J., Haverland, P. S., Kemble, M. E., Lindskoog, R. A., Severn, C. and Smorong, D. E.
2000
.
Prediction of sediment toxicity using consensus-based freshwater sediment quality guidelines
,
USEPA 905/R-00/007
MacDonald, D. D., Ingersoll, C. G. and Berger, T. A.
2000
.
Development and evaluation of consensus-based sediment quality guidelines for freshwater ecosystems
.
Arch. Environ. Contam. Toxicol.
,
39
:
20
31
.
PADER (Pennsylvania Department of Environmental Resources)
.
1991
.
Trophic State Analysis, Presque Isle Bay, Erie County.
,
Meadville, PA.
:
Northwest Pennsylvania Region
.
PADER
.
1992
. “
Presque Isle Bay Remedial Action Plan
”. In
Pennsylvania Department of Environmental Resources
,
Meadville, PA.
:
Northwest Pennsylvania Region
.
PADER
.
1993
.
Presque Isle Bay brown bullhead tumor study, Conducted from March 29, 1992 to October 7, 1993
,
Prepared by E. C. Obert, Pennsylvania Department of Environmental Protection
PADER
.
1995
. “
Update to the Presque Isle Bay Remedial Action Plan
”. In
Pennsylvania Department of Environmental Resources
,
Meadville, PA.
:
Northwest Pennsylvania Region
.
PADER
.
1996
.
A study of tumors in fish of Presque Isle Bay, 1995
,
Prepared by M.K. Walter, Pennsylvania Department of Agriculture and D.M. Dambach, University of Pennsylvania, for Pennsylvania Department of Environmental Resources. June, 1996
Potomac-Hudson Engineering
.
1991
.
Presque Isle Bay Ecosystem Study—Background Report
,
Meadville, PA
:
Prepared for the Pennsylvania Department of Environmental Resources
.
USEPA
.
1994
.
ARCS Assessment Guidance Document
,
EPA 905-B94-002
Chicago, IL.
:
Great Lakes National Program Office
.
USEPA
.
2002a
.
A Guidance Manual to Support the Assessment of Contaminated Sediments in Freshwater Ecosystems: Volume III-Interpretation of the Results of Sediment Quality Investigations
,
EPA-905-B02-001-C
USEPA
.
2002b
.
Equilibrium Partitioning Sediment Guidelines (ESGs) for the Protection of Benthic Organisms: Metal Mixtures (Cadmium, Copper, Lead, Nickel, Silver, and Zinc) (DRAFT)
,
EPA-822-R-02-045
Washington, DC.
:
U.S. Environmental Protection Agency
.
USEPA
.
2002c
.
Procedures for the Determination of Equilibrium Partitioning Sediment Benchmarks (ESBs) for the Protection of Benthic Organisms: PAH Mixtures
,
EPA-600-R-02-013
Washington, DC.
:
U.S. Environmental Protection Agency
.
USEPA and USACE
.
1998
.
Great Lakes Dredged Material Testing & Evaluation Manual
,
Washington, DC.
:
Final Draft. U.S. Environmental Protection Agency
.