Assessment of Water Column data collected in response to a fly ash release on the Delaware River
Delaware River Basin Commission
John R. Yagecic, P.E.
June 19, 2006
Executive Summary
• Ash basin leak at PPL’s Martin’s Creek power plant August 23rd-27th, 2005 released ~100 million gallons of ash and water into the Delaware River; • The States of PA & NJ requested that DRBC compile and assess water column data from agencies and PPL; • The results of this effort are as follows: Arsenic • Some apparent exceedance of PA human health arsenic criteria immediately after the release in the vicinity of the release. These exceedances are not observed upstream of the release. • Apparent signature of release observable downstream but below PA arsenic criteria.
Executive Summary (continued)
Lead • Apparent exceedances of PA and NJ lead criteria immediately after release in the vicinity of the release. Aluminum • Some short term local exceedance of PA aluminum criteria from release, but background and high flows account for most of the observed concentration. Manganese • Manganese concentrations appear to be background concentrations.
Executive Summary (continued)
Copper • Short term copper exceedances of PA criteria at release and downstream. Selenium • Selenium concentrations appear to be background concentrations. Mercury • Mercury data almost completely below reporting limits.
Brief Background
• Ash basin leak at PPL’s Martin’s Creek power plant August 23rd-27th, 2005 released ~100 million gallons of ash and water into the Delaware River; • Agencies and PPL met at DRBC on November 3, 2005 to review data collected up to that point; • DRBC agreed to compile and assess data from agencies and PPL; • Agencies and PPL agreed that DRBC and PPL should perform separate but parallel assessments; • Agencies and PPL previewed DRBC draft assessment on March 2, 2006. Recommended expansion of the data set and minor changes. • This Document:
– Results of DRBC’s assessment; – Narrowly focused on data – not an assessment of regulatory issues.
Post-Release Water Column Sampling Effort
Data received by DRBC as of 12/2005
Samples Collected >1,700 47 36 21 9 4 2 >1,819 Sampling Days 84 2 8 21 2 1 2 Analytical Parameters 60 28 up to 117 24 30 5 10
Organization PPL NJDEP PADEP NJWSA DRBC PWD USGS Total
Locations 16 9 16 1 4-5 4 1
Delaware River Flow at Belvidere (CFS)
10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000 5,000
0
Post-Release Sampling Effort
PADEP
Discharge (CFS)
Release
PPL DRBC PWD
Date
NJDEP USGS NJWSA
8/20/2005 8/22/2005 8/24/2005 8/26/2005 8/28/2005 8/30/2005 9/1/2005 9/3/2005 9/5/2005 9/7/2005 9/9/2005 9/11/2005 9/13/2005 9/15/2005 9/17/2005 9/19/2005 9/21/2005 9/23/2005 9/25/2005 9/27/2005 9/29/2005 10/1/2005 10/3/2005 10/5/2005 10/7/2005 10/9/2005 10/11/2005 10/13/2005 10/15/2005 10/17/2005 10/19/2005 10/21/2005 10/23/2005 10/25/2005 10/27/2005 10/29/2005 10/31/2005 11/2/2005 11/4/2005 11/6/2005 11/8/2005 11/10/2005 11/12/2005 11/14/2005 11/16/2005
1
10
0.1
100
Number of Samples Collected
Wide Range of Flows During Sampling
1,000,000
October 9, 2005 78,900 CFS Exceeded 0.59% of time.
100,000
Flow at Trenton (CFS)
October 14, 2005 49,900 Exceeded 2.29% of time.
Mean daily flow from August 26th to October 7th, 2005. 3,070 CFS Exceeded 95.3% of time.
10,000
1,000 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Percentage of Time Flow Exceeded
Flow at Trenton (CFS) High Flow Events Mean Flow August 26th to October 7th, 2005
Criteria at the time of release
(some criteria have subsequently changed)
Pennsylvania • Chapter 16. Water Quality Toxics Management Strategy – Statement of Policy; • CCCs, CMCs, and Human Health Criteria. • Governing (most stringent) PA criteria;
– Human Health for arsenic; – CCC for all others.
New Jersey • Surface Water Quality Standards N.J.A.C. 7:9B (June 2005); • For the Delaware River, NJ defers to DRBC criteria where DRBC have criteria; • Where DRBC doesn’t, use FW2-NT criteria for non-saline waters.
Prioritization of Analytical Parameters*
100% 90% % of Results Exceeding Either a PA or NJ Criteria
88%
80%
75%
70%
60%
50%
50%
40%
To prioritize the comparison of results to applicable criteria, we first compared samples with paired metals and hardness data to the computed criteria based on the sample specific hardness. This screening showed that arsenic, lead, aluminum, manganese, and copper were potentially the parameters of interest. This screening did not differentiate between J-flagged (estimated) results and un-flagged results. In addition, this screening did not consider the location (upstream or downstream of release) of the sample. Finally, only a small subset of results included paired metals and hardness data. After we identified the parameters of interest using this screening approach, we performed a more comprehensive comparison to criteria. ‡NJ Manganese criteria were not applicable to freshwater, but manganese was retained as an indicator parameter.
13%
30%
20%
At the request of the states, we also assessed selenium results.
10%
0%
0%
Copper Total Arsenic Total Barium Total Manganese Total ‡ Lead Total Aluminum Total
0%
Beryllium Total
0%
Chromium Total
0%
Cobalt Total
0%
Molybdenum Total
0%
Nickel Total
0%
Vanadium Total
0%
Zinc Total
Analytical Parameter
*For prioritization only – not for regulatory purposes
Hardness Ranges from Lower Delaware Existing Water Quality Study
90 80 70
60
Hardness (mg/L)
50
40
30
20
10
Where PA or NJ criteria were based on hardness, we computed the criteria using hardness measurements from the Lower Delaware River Monitoring Program for establishment of existing water quality. These values were measured between 2000 and 2004. Over 430 hardness measurements were made within the area of interest during that monitoring period. Quartiles of hardness results are shown here.
0 -20 -10 0 10 20 30 40 50 60 70
Miles Downstream from Fly Ash Release
25th Percentile Median 75th Percentile
Spatial Grouping of Results
based on Arsenic and Aluminum Samples
700
Upstream
600
At Release
Near Downstream
Easton
Estuary
No. of Samples Collected
500
400
300
The data showed that metals concentrations were changing in both space and time. To better isolate temporal changes, we broke the data sets into spatial groups based on location and the number of samples collected in each area.
200
100
0
0.030 0.100 0.200 5.130 5.180 8.900 8.970 -3.640 -1.370 -0.800 -0.130 -0.100 -0.090 -13.490 -0.020 9.100 10.380 10.900 19.210 19.400 26.500 29.900 36.860 37.200 37.700 38.800 45.470 45.500 56.800 57.170 59.500 59.860 74.960 75.750 83.960
Miles Downstream From Release
No of Samples analyzed for Aluminum or Arsenic
Arsenic Results
Arsenic Concentrations Upstream of Release
Arsenic Water Column Concentrations Measured Upstream of Release
100
10
Arsenic Total (ug/L)
1
Most results upstream of the release were non-detect. Of the quantified results, all were below the PA arsenic criteria, but all were above the NJ arsenic criteria.
0.1
0.01 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Arsenic Total (ug/L) PA Arsenic Criteria (ug/L) Arsenic Total J-Flagged (ug/L) NJ Arsenic Criteria (ug/L) Arsenic Non-Detects (ug/L)
Arsenic Concentrations at Release
Arsenic Water Column Concentrations Measured in the immediate vicinity of the Release
1000
100
Arsenic Total (ug/L)
10
1
0.1
In the immediate vicinity of the release, initial concentrations in excess of the PA arsenic criteria were observed, with concentrations falling below the PA criteria shortly after the release. All quantified concentrations exceeded the NJ arsenic criteria.
0.01 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Arsenic Total (ug/L) Arsenic Total Non-Detects (ug/L) PA Arsenic Criteria (ug/L) NJ Arsenic Criteria (ug/L)
Arsenic Concentrations near Downstream
Arsenic Water Column Concentrations Measured 0.1 to 0.2 miles downstream from release
100 100
Arsenic Water Column Concentrations Measured 5.13 to 5.18 miles downstream from release
10
10
Arsenic Total (ug/L)
1
Arsenic Total (ug/L)
9/5/2005 9/25/2005 10/15/2005 11/4/2005 11/24/2005 12/14/2005 1/3/2006 1/23/2006 2/12/2006
1
0.1
0.1
0.01 8/16/2005
0.01 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time Arsenic Total (ug/L) PA Arsenic Criteria (ug/L) Arsenic Total Non-Detects (ug/L) NJ Arsenic Criteria (ug/L) Arsenic Total (ug/L) PA Arsenic Criteria (ug/L)
Date and Time Arsenic Total Non-Detects (ug/L) NJ Arsenic Criteria (ug/L)
A comparison of time series arsenic measurements at two downstream ranges shows that most results were below the PA arsenic criteria, but all quantified results exceeded the NJ arsenic criteria. Both plots suggest resuspension of settled arsenic during the mid-October storm events.
Arsenic Concentrations near Easton
Arsenic Water Column Concentrations Measured between RM 8 and 11 (Easton)
100
10
Arsenic Total (ug/L)
1
0.1
Near Easton, all results were below the PA arsenic criteria, but all quantified results exceeded the NJ arsenic criteria. Again, some apparent resuspension is observed during the mid-October storm events.
0.01 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Arsenic Total (ug/L) PA Arsenic Criteria (ug/L) Arsenic Total J-Flagged (ug/L) NJ Arsenic Criteria (ug/L) Arsenic Total Non-Detect (ug/L)
Arsenic Concentrations in Estuary
Arsenic Water Column Concentrations Measured in the Estuary
100
10
Arsenic Total (ug/L)
1
0.1
Some arsenic results were collected in the estuary during mid-October (note the change in the X-axis). DRBC also has an arsenic criteria that applies to the estuary only. All results were below the PA and DRBC arsenic criteria, but all quantified results exceeded the NJ arsenic criteria. Since results were collected in mid-October only, no comparison can be made to the period immediately after the release.
0.01 10/8/2005
10/9/2005
10/10/2005
10/11/2005
10/12/2005
10/13/2005
10/14/2005
10/15/2005
10/16/2005
10/17/2005
10/18/2005
Date
Arsenic Total (ug/L) NJ Arsenic Criteria (ug/L) Arsenic Total Non-Detects (ug/L) DRBC Arsenic Criteria (ug/L) PA Arsenic Criteria (ug/L)
Arsenic Concentrations by Distance from Release, August 26, 2005
Arsenic Water Column Concentrations on August 26, 2005
1,000.00
100.00
Arsenic Total (ug/L)
10.00
We plotted all results collected on August 26, 2005 by river mile. This plot shows good agreement between results collected by different agencies, and demonstrates a decrease in arsenic concentration moving from the release site downstream.
1.00
0.10
0.01 -10 0 10 20 30 40 50 60 70
Miles Downstream from Release
PPL NJDEP PWD PPL Non-Detects NJWSA PA Arsenic Criteria DRBC PADEP NJ Arsenic Criteria DRBC Non-Detects PADEP Non-Detects
Arsenic Concentrations by Distance from Release, August 27, 2005
Arsenic Water Column Concentrations on August 27, 2005
1,000.00
100.00
Arsenic Total (ug/L)
A similar plot for the following day (August 27, 2005) suggests that the peak arsenic concentration has shifted downstream. All results are below the PA arsenic criteria, but all quantifiable results exceed the NJ arsenic criteria.
10.00
1.00
0.10
0.01 -10 0 10 20 30 40 50 60 70
Miles Downstream from Release
PPL NJDEP NJWSA USGS PA Arsenic Criteria NJ Arsenic Criteria
Percentiles of Arsenic Measurements
Percentiles of Water Column Arsenic Measurements from 5 Location Ranges
1000
100
Arsenic Total (ug/L)
10
1
0.1
0.01 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1
Percentiles
Upstream Estuary At Release PA Arsenic Criteria (ug/L) Near Downstream NJ Arsenic Criteria (ug/L) Near Easton
Comparison to Pre-Release Arsenic Data (1992-Present) found in NWIS and STORET
Comparison of Pre- and Post-Release Arsenic Concentrations
100
Comparison of post-release results to historical arsenic ranges at Riegelsville and Trenton indicate higher concentrations measured after the release.
Arsenic Total (ug/L)
10
1
9 Data Points 8 of which were Non-Detect at 1 ug/L.
0.1 0 10 20 30 40
36 Data Points 34 of which were Non-Detect at concentrations ranging from 1 to 3 ug/L.
50
60
70
Miles Downstream from Release
Arsenic Total Post Release (ug/L) Arsenic Total Trenton Range Pre-Release (ug/L) Arsenic Total Post-Release Non-Detect (ug/L) Arsenic Total Riegelsville Range Pre-Release (ug/L)
Lead Results
Lead Concentrations Upstream of Release
Lead Water Column Concentrations Measured Upstream of Release
100
Lead Total (ug/L)
10
Results from most lead samples collected upstream of the release were non-detect. Any quantifiable concentrations exceeded the PA hardness based lead criteria. Most results were below the NJ lead criteria. In mid-September, PPL exceeded the capacity of their contract analytical laboratory and changed to a new laboratory, resulting in a change to the lead reporting limits and diminishing the resolution of the lead data. It should be noted that the new lab provided slightly better reporting limits for arsenic, the primary parameter of interest. By December 2006, PPL returned to the original lab and the lower lead reporting limits.
1
0.1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005
12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Lead Total (ug/L) PA Lead Criteria (25th Percentile Hardness) (ug/L) NJ Lead Criteria (ug/L) Lead Total Non-Detects (ug/L) PA Lead Criteria (75th Percentile Hardness) (ug/L)
Lead Concentrations at Release
Lead Water Column Concentrations Measured in the immediate vicinity of the Release
100
Lead Total (ug/L)
10
Lead samples collected at the release show a few values that exceed the NJ lead criteria immediately after the release, with concentrations decreasing over time. All quantifiable concentrations appear to exceed the PA criteria. Again, the change in reporting limits in mid-September makes comparison to criteria impossible for non-detect data.
1
0.1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005
12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date
Lead Total (ug/L) PA Lead Criteria (25th Percentile Hardness) (ug/L) NJ Lead Criteria (ug/L) Lead Total Non-Detects (ug/L) PA Lead Criteria (75th Percentile Hardness) (ug/L)
Lead Concentrations near Downstream
Lead Water Column Concentrations Measured 0.1 to 0.2 miles downstream from release
100 100
Lead Water Column Concentrations Measured 5.13 to 5.18 miles downstream from release
Lead Total (ug/L)
1
Lead Total (ug/L)
9/5/2005 9/25/2005 10/15/2005 11/4/2005 11/24/2005 12/14/2005 1/3/2006 1/23/2006 2/12/2006
10
10
1
0.1 8/16/2005
0.1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Lead Total (ug/L) PA Lead Criteria (25th Percentile Hardness) (ug/L) NJ Lead Criteria (ug/L) Lead Total Non-Detects (ug/L) PA Lead Criteria (75th Percentile Hardness) (ug/L)
Date and Time
Lead Total (ug/L) Lead Total Non-Detects (ug/L) PA Lead Criteria (75th Percentile Hardness) (ug/L) Lead Total J-Flagged (ug/L) PA Lead Criteria (25th Percentile Hardness) (ug/L) NJ Lead Criteria (ug/L)
Lead samples collected downstream of the release are mostly below the NJ lead criteria. All quantifiable concentrations appear to exceed the PA criteria. Again, the change in reporting limits in mid-September makes comparison to criteria impossible for non-detect data.
Lead Concentrations near Easton
Lead Water Column Concentrations Measured between RM 8 and 11 (Easton)
100
10
Lead Total (ug/L)
1
0.1
Lead samples collected near Easton show a few values that exceed the NJ lead criteria immediately after the release. All quantifiable concentrations appear to exceed the PA criteria. Again, the change in reporting limits in mid-September makes comparison to criteria impossible for non-detect data.
0.01 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Lead Total (ug/L) Lead Total Non-Detects (ug/L) PA Lead Criteria (75th Percentile Hardness) (ug/L) Lead Total J-Flagged PA Lead Criteria (25th Percentile Hardness) (ug/L) NJ Lead Criteria (ug/L)
Lead Concentrations by Distance from Release, August 26, 2005
Lead Water Column Concentrations on August 26, 2005
100
10
Lead Total (ug/L)
Lead results collected on August 26, 2005 plotted by River Mile show a decrease in lead concentration moving from the release site downstream. Some concentrations exceeding both the PA and NJ lead criteria are apparent at the release site, with lower concentrations downstream.
1
0.1
0.01 -20 -10 0 10 20 30 40 50 60 70
Miles Downstream from Release
Lead Total (ug/L) Lead Total Non-Detects (ug/L) PA Lead Total Criteria (Median Hardness) (ug/L) NJ Lead Total Criteria (ug/L) Lead Total J-Flagged (ug/L) PA Lead Total Criteria (25th Percentile Hardness) (ug/L) PA Lead Total Criteria (75th Percentile Hardness) (ug/L)
Lead Concentrations by Distance from Release, August 26 and 27, 2005
Lead Water Column Concentrations on August 26, 2005
100 100
Lead Water Column Concentrations on August 27, 2005
10
10
1
Lead Total (ug/L)
-20 -10 0 10 20 30 40 50 60 70
Lead Total (ug/L)
1
0.1
0.1
0.01
0.01 -20 -10 0 10 20 30 40 50 60 70
Miles Downstream from Release
Lead Total (ug/L) Lead Total Non-Detects (ug/L) PA Lead Total Criteria (Median Hardness) (ug/L) NJ Lead Total Criteria (ug/L) Lead Total J-Flagged (ug/L) PA Lead Total Criteria (25th Percentile Hardness) (ug/L) PA Lead Total Criteria (75th Percentile Hardness) (ug/L)
Miles Downstream from Release
Lead Total (ug/L) Lead Total Non-Detects (ug/L) PA Lead Total Criteria (Median Hardness) (ug/L) NJ Lead Total Criteria (ug/L) Lead Total J-Flagged (ug/L) PA Lead Total Criteria (25th Percentile Hardness) (ug/L) PA Lead Total Criteria (75th Percentile Hardness) (ug/L)
A side-by-side comparison of lead samples collected on August 26 and 27, 2005 plotted by River Mile indicates a decrease in concentration at the release site. Unlike arsenic, there is no apparent downstream movement of a concentration peak.
Comparison to Pre-Release Lead Data (1992-Present) found in NWIS and STORET
Comparison of Pre- and Post-Release Lead Concentrations
100
10
Lead Total (ug/L)
44 Data Points 21 of which were Non-Detect at 1 ug/L
1
0.1
9 Data Points 7 of which were Non-Detect at 1 ug/L
0.01 -20 0 20 40 60
Comparison of post-release results to historical lead ranges at Riegelsville and Trenton indicate results that appear to be fairly consistent with the historic range. The lower limit of each historic range is bounded by the detection limits. Therefore, J-flagged data which appears to be below the range is consistent with the range. At each site, one quantifiable measurement above the historic 80 100 range was observed.
Miles Downstream from Release
Lead Total Post-Release (ug/L) Trenton Pre-Release Range (ug/L) Lead Total J-Flagged (ug/L) Riegelsville Pre-Release Range (ug/L) Lead Total Non-Detect Post-Release (ug/L)
Aluminum Results
Aluminum Concentrations Upstream of Release
Aluminum Water Column Concentrations Measured Upstream of Release
10,000
1,000
Aluminum Total (ug/L)
100
10
Upstream of the release, aluminum water column concentrations in excess of both the PA aluminum criteria and the EPA recommended aluminum criteria are observed coincident with the mid-October storm events, suggesting resuspension of bottom material as a likely contributor to water column aluminum.
1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005
12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Aluminum Total (ug/L) Aluminum Total Non-Detects (ug/L) PA Aluminum Criteria (ug/L) EPA Criteria (ug/L)
Aluminum Concentrations at Release
Aluminum Water Column Concentrations Measured at Release
100,000
10,000
Aluminum Total (ug/L)
1,000
100
10
In the vicinity of the release, higher aluminum concentrations in excess of the PA and EPA recommended aluminum criteria are observed immediately after the release. Higher concentrations are observed again during the mid-October storm events, consistent with the upstream sites.
9/5/2005 9/25/2005 10/15/2005 11/4/2005 11/24/2005 12/14/2005 1/3/2006 1/23/2006 2/12/2006
1 8/16/2005
Date and Time
Aluminum Total (ug/L) Aluminum Total Non-Detects (ug/L) PA Aluminum Criteria (ug/L) EPA Criteria (ug/L)
Aluminum Concentrations upstream and at release
Aluminum Water Column Concentrations Measured Upstream of Release
100,000 100,000
Aluminum Water Column Concentrations Measured at Release
10,000
10,000
Aluminum Total (ug/L)
1,000
Aluminum Total (ug/L)
1,000
100
100
10
10
1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005
12/14/2005
1/3/2006
1/23/2006
2/12/2006
1 8/16/2005
9/5/2005
9/25/2005 10/15/2005 11/4/2005 11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Aluminum Total (ug/L) Aluminum Total Non-Detects (ug/L) PA Aluminum Criteria (ug/L) EPA Criteria (ug/L) Aluminum Total (ug/L)
Date and Time
Aluminum Total Non-Detects (ug/L) PA Aluminum Criteria (ug/L) EPA Criteria (ug/L)
A side-by-side comparison of the upstream and release site plots suggests an initial increase in water column aluminum at the release site, followed by a storm related increase that is consistent with the upstream sites.
Aluminum Concentrations at near downstream locations
Aluminum Water Column Concentrations Measured 0.1 to 0.2 miles downstream from release
10,000
10,000
Aluminum Water Column Concentrations Measured 5.13 to 5.18 miles downstream from release
One measurement (11,300 ug/L) off the scale
Aluminum Total (ug/L)
Aluminum Total (ug/L)
9/5/2005 9/25/2005 10/15/2005 11/4/2005 11/24/2005 12/14/2005 1/3/2006 1/23/2006 2/12/2006
1,000
1,000
100
100
10 8/16/2005
10
8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005
12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Aluminum Total (ug/L) Aluminum Total Non-Detects (ug/L) PA Aluminum Criteria (ug/L) EPA Criteria (ug/L) Aluminum Total (ug/L)
Date and Time
Aluminum Total Non-Detects (ug/L) PA Aluminum Criteria (ug/L) EPA Criteria (ug/L)
A side-by-side comparison of plots for the 2 near downstream ranges shows an initial increase in water column aluminum concentration after the release, followed by storm related increases in mid October. The structure of the initial concentration increase is more apparent at the second downstream range, even though that range is further from the initial release site. We observed this difference in structure for other analytical parameters as well. One possible explanation is that the second downstream range is shallower and has a higher velocity, keeping more material in suspension. By contrast, the first downstream range is more of a pool and may tend to dampen the signal of the release. As with lead, a change to a higher aluminum reporting limit interferes with direct comparison to the EPA recommended criteria for non-detect data.
Regression of Paired Aluminum Concentrations
Regression of Paired Aluminum Observations Upstream and At Release
4,000 8,000
Regression of Paired Aluminum Observations Upstream and Near Easton
3,500
7,000
Aluminum at Release (ug/L)
3,000
Aluminum Total Near Easton (ug/L)
6,000
y = 0.9229x + 177.73 2 R = 0.4546
2,500
y = 0.8253x + 140 2 R = 0.6899
5,000
2,000
4,000
1,500
3,000
1,000
2,000
500
1,000
0 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000
0 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000
Aluminum Upstream of Release (ug/L)
At Release Match (1:1) Linear (At Release)
Aluminum Total Upstream of Release (ug/L)
Near Easton Match (1:1) Linear (Near Easton)
Bivariate plots of aluminum concentrations at the release site and at Easton versus the upstream concentrations, suggests that portion of the variability in the downstream concentrations is explained by the upstream concentration. At each location, however, a subset of values not following this relationship is apparent.
Our interpretation is that the initial peaks are associated with the release, but subsequent peaks strongly follow the upstream storm related pattern.
Manganese
Manganese Concentrations Upstream of Release
Manganese Water Column Concentrations Upstream of the Release
1000
Manganese Total (ug/L)
100
10
A plot of manganese water column concentrations upstream of the release shows a strong response to mid-October storms, including concentrations in excess of the NJ manganese criteria.
1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005
12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Manganese Total (ug/L) Manganese Total Non-Detects (ug/L)
Manganese Concentrations Upstream of and at Release
Manganese Water Column Concentrations Upstream of the Release
1000 1000
Manganese Concentrations at Release
Manganese Total (ug/L)
100
Manganese Total (ug/L)
9/5/2005 9/25/2005 10/15/2005 11/4/2005 11/24/2005 12/14/2005 1/3/2006 1/23/2006 2/12/2006
100
10
10
1 8/16/2005
1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005
12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Manganese Total (ug/L) Manganese Total Non-Detects (ug/L)
Date and Time
Manganese Total (ug/L) Manganese Total Non-Detects (ug/L)
A side-by-side comparison of plots of manganese water column concentrations upstream and at the release indicates that both sites show a strong response to mid-October storm events, including concentrations in excess of the NJ manganese criteria. There are no apparent elevated concentrations of manganese at the release site relative to the upstream sites. In fact the range of concentrations for the upstream sites is slightly higher than for the release site.
Percentiles of Manganese Concentrations Upstream of and At Release
Percentiles of Manganese Concentrations Upstream of at At Release
1000
Manganese Total (ug/L)
100
10
A comparison of the percentiles of measured manganese water column concentrations further confirms that concentrations at the release are not higher than upstream concentrations throughout the range of measured concentrations.
1 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Percentiles
Upstream At Release
Copper Results
Copper Concentrations Upstream
Copper Water Column Concentrations Measured Upstream of Release
100
Copper Total (ug/L)
A review of water column copper concentrations upstream of the release indicates a handful of quantifiable concentrations, including a few apparently exceeding PA copper criteria. Again, a change in reporting limits diminishes the resolution of the measurements after mid-September, and makes meaningful comparison to the PA criteria impossible for non-detects.
10
1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Copper Total (ug/L) PA Copper Criteria (25th Percentile Hardness) (ug/L) Copper Total Non-Detects (ug/L) PA Copper Criteria (75th Percentile Hardness) (ug/L)
Copper at Release
Copper Water Column Concentrations Measured at the Release
100
At the release, copper concentrations may be decreasing immediately after the release. This trend, however, is no longer discernable after the change in reporting limits.
Copper Total (ug/L)
10
1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Copper Total (ug/L) PA Copper Criteria (25th Percentile Hardness) (ug/L) Copper Total Non-Detects (ug/L) PA Copper Criteria (75th Percentile Hardness) (ug/L)
Copper Near Easton
Copper Water Column Concentrations Measured between RM 8 and 11 (Easton)
100
Near Easton, water column copper concentrations appear to be falling after the initial release, but the change in reporting limits obscures the trend. A few samples exceeded the PA copper criteria.
Copper Total (ug/L)
10
1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Copper Total (ug/L) Copper Total Non-Detects (ug/L) PA Copper Criteria (75th Percentile Hardness) (ug/L) Copper Total J-Flagged (ug/L) PA Copper Criteria (25th Percentile Hardness) (ug/L)
Copper by Miles from Release on August 26th and 27th, 2005
Copper Concentrations by Distance downstream from Release on August 26, 2005
100 100
Copper Concentrations by Distance downstream from Release on August 27, 2005
Copper Total (ug/L)
10
Copper Total (ug/L)
-20 -10 0 10 20 30 40 50 60 70
10
1
1 -20 -10 0 10 20 30 40 50 60 70
Miles Downstream from Release
Copper Total (ug/L) Copper Total Non-Detects (ug/L) PA Copper Criteria (Median Hardness) (ug/L) Copper Total J-Flagged (ug/L) PA Copper Criteria (25th Percentile Hardness) (ug/L) PA Copper Criteria (75th Percentile Hardness) (ug/L)
Miles Downstream from Release
Copper Total (ug/L) Copper Total Non-Detects (ug/L) PA Copper Criteria (Median Hardness) (ug/L) Copper Total J-Flagged (ug/L) PA Copper Criteria (25th Percentile Hardness) (ug/L) PA Copper Criteria (75th Percentile Hardness) (ug/L)
A side-by-side comparison of copper concentration measured on August 26 and 27, 2005 plotted by River Mile suggests that higher concentrations observed near the release on August 26th have decreased by August 27th. Unlike arsenic, there is no apparent downstream movement of a concentration peak.
Selenium Results
Selenium Concentrations Upstream of and at the Release
Selenium Water Column Concentrations Upstream of the Release
100 100
Selenium Water Column Concentrations Upstream of the Release
Selenium Total (ug/L)
10
Selenium Total (ug/L)
9/5/2005 9/25/2005 10/15/2005 11/4/2005 11/24/2005 12/14/2005 1/3/2006 1/23/2006 2/12/2006
10
1 8/16/2005
1 8/16/2005
9/5/2005
9/25/2005
10/15/2005
11/4/2005
11/24/2005 12/14/2005
1/3/2006
1/23/2006
2/12/2006
Date and Time
Selenium Total (ug/L) Selenium Non-Detects (ug/L) PA Selenium Criteria (ug/L) NJ Selenium Criteria (ug/L) Selenium Total (ug/L)
Date and Time
Selenium Non-Detects (ug/L) PA Selenium Criteria (ug/L) NJ Selenium Criteria (ug/L)
A side-by-side comparison of water column selenium concentrations measured upstream of and at the release are nearly identical, suggesting that the release did not contribute to water column selenium concentrations.
Mercury
• All mercury measurements (1,805) were non-detect except for one sample collected 0.2 miles downstream of the release on 10/13/05; • About 65% of the samples had a reporting limit of 0.2 ug/L and about 35% of the samples had a reporting limit of 0.5 ug/L.
Conclusions
Arsenic • Some apparent exceedance of PA arsenic criteria immediately after the release in the vicinity of the release. These exceedances are not observed upstream. • Apparent signature of release observable downstream but below PA arsenic criteria. • Any quantifiable concentration of arsenic exceeds NJ criteria (both upstream and downstream).
Conclusions (continued)
Lead • Apparent exceedances of PA and NJ lead criteria immediately after release in the vicinity of the release. • Most quantifiable lead concentrations exceed PA criteria (including upstream). Aluminum • Some short term local exceedance of PA aluminum criteria from release, but background and high flows account for most of the observed concentration.
Conclusions (continued)
Manganese • Manganese concentrations appear to be background concentrations. Copper • Short term copper exceedances of PA criteria at release and downstream.
Conclusions (continued)
Selenium • Selenium concentrations appear to be background concentrations. Mercury • Mercury data almost completely below reporting limits.
Assessment of Water Column data collected in response to a fly ash release on the Delaware River
Delaware River Basin Commission
John R. Yagecic, P.E.
June 19, 2006