Troubled Waters in Chesterfield: A Review of Upper Swift Creek Watershed Reveals Disturbing Pollution Issues Tuesday, December 13, 2005
Posted by Conaway B. Haskins III in Uncategorized.trackback
By Thomas A. Pakurar, Ph.D.
South of the James Guest Contributor
Tom Pakurar, PhD is a water quality expert and Co-Chairman of Hands Across the Lake, a nonprofit community organization of Chesterfield County residents concerned with maintaining the water quality of the waters of Chesterfield County & Swift Creek Reservoir. A retired DuPont scientist, Tom is on the water board of the Virginia Conservation Network and is a member of the water advisory committee of the Sierra Club. He was a member of the Chesterfield County Watershed Management Committee from 1993-1997. He holds a bachelors degree in chemical engineering from Cornell University and masters and doctoral degrees in chemical engineering from the University of Delaware.
The presence of coffee-colored water flowing into the creeks and drinking water reservoir for days or weeks after a storm raises serious doubts about the adequacy of pollution removal in the Upper Swift Creek Watershed. The big question is how much phosphorus pollution is really removed by the settling ponds via the Best Management Practices (BMPs) being designed by the county consultants? These settling ponds are designed to the legal standards set by the state instead of the actual performance efficiencies requested by the Board of Supervisors in 1992. According to a recent Freedom of Information Act (FOIA) search, the performance data does not exist. Local residents, however, take pictures and report muddy runoff to the county. They see coffee-colored runoff flowing into the creeks and reservoir for days and weeks after a storm. These settling ponds do not seem to work very well to remove the suspended soils and phosphorus pollution attached to the soil particles.
In 1993, the Chesterfield Board of Supervisors asked for technical considerations to guide the political decisions the Board would make regarding water quality in the county. By consensus, the county Watershed Management Committee, from 1993-1997, recommended phosphorus pollution as the key yardstick and the Board of Supervisors adopted the performance-based “phosphorus laws” in 1997 to manage pollution in the Upper Swift Creek (USC) watershed.
The technical report written in support of the draft Upper Swift Creek Plan (under review) appears to be inaccurate in two critical areas:
- It appears to have understated the total amount of phosphorus pollution generated by the land use changes identified in the plan; and
- It appears to have overstated the actual ability to remove pollutants in the colloidal clay soils of the region. Both of these concerns can be addressed.
I recommend the county comply with the 1992 Board directive and conduct tests to measure the pollution removal efficiency of its settling ponds. Then, the county should use the more accurate pollutant removal efficiencies before finalizing the land use plan and accompanying water quality plan. The next review of the plan for development in the Upper Swift Creek Area is currently scheduled in March 2006 at a Chesterfield Planning Commission work session. The technical details for water quality are presented as Appendix D. Tim Hare and Jamie Lynn Conner of consulting firm CH2MHill wrote the report dated August 15, 2005 and addressed the work needed to limit pollution from the planned development.
The county established several phosphorus pollution laws and guides in 1997 based on a consensus recommendation of the county Watershed Management Committee (1993-1997). Joan Salvati, Manager of the county’s Office of Water Quality led the committee; the County Board of Supervisors (BOS) appointed developers, environmentalists, citizens at large, and representatives of community associations from all watershed areas of the county as committee members. Phosphorus, a major pollutant, was selected as the key indicator of the health of the reservoirs. The committee believed that if the phosphorus levels were satisfactory, then all other critical indicators of water health would also be satisfactory.
In the technical paper “Appendix D,” CH2MHill appears to have understated the amount of pollution generated in four areas:
- The phosphorus pollution load from land-disturbing activities may have been omitted. The phosphorus pollution from these activities is 240 times the desired post-development pollution load used in Appendix D. The Chesapeake Bay Foundation believes it to be 1000 times higher.
- The pollution from the new cross-reservoir highway connecting Powhite to Hull Street Road may have been omitted.
- The pollution calculated from the deferred growth area (~5000 acres) may not have considered the ultimate land use potential.
- The computer model appears to have used 39,642 acres instead of 41,950 acres in the watershed (Watershed Management Committee, 1995). The 2308-acre difference may contribute a significant pollution load if an intense land use were assigned to them.
Phosphorus pollution removal is accomplished by using settling ponds or BMPs. The soil particles containing the pollutants are allowed to settle out so the water leaving the pond is significantly cleaner than the water entering. The Virginia Stormwater Management Handbook tables allow up to 65% pollutant removal efficiencies if the pond volumes are big enough. The tables were developed using average soils in the mid-Atlantic region. They do not apply to the actual colloidal clays found in the Upper Swift Creek (USC) watershed. These clays behave like the butterfat in homogenized milk – neither one separates from the carrier fluid.
In a settling pond with BMP, the soil particles do not settle while they are in the pond. It takes weeks for these soils to settle, so the water leaving the pond is usually coffee-color indicating high pollution. The picture shows one such BMP operating near the reservoir. Click on the picture to see an enlarged view. Note the coffee-colored runoff on the left flowing directly into the reservoir. Both the Environmental Protection Agency (EPA) and the Virginia Department of Conservation and Recreation (DCR) recommend against using settling ponds in regions where the soil particles do not settle. We have that situation in the Upper Swift Creek watershed.
I recommend that the county conduct tests to measure the pollution removal efficiency of its settling ponds. The Board of Supervisors mandated this in 1992 so adjustments in the water quality plan can be made using more applicable data. Depending on the results of these tests, the County may have to treat a larger percentage of the runoff or use technologies such as pond aeration and vegetative planting to make the ponds more efficient. The county can modify the Land-use Plan to protect its waters. For example the Ruffin Mill area of the county has similar soils to the Upper Swift Creek Area. It required the number of houses per acre to be less than or equal to 1.5 dwelling units/acre; currently the Upper Swift Creek (USC) Plan allows 33 per cent higher housing density.
Failure to address pollution removal in a scientific, technically sound way can only spell disaster for the county’s major drinking water reservoir.
Hey, I appreciate learning about the water quality in our fair chesterfield. Nonetheless, I feel like there is a missing piece to this article. Specifically, I was wondering how you know that the water quality is crappy. I assume writer, being a chemist and all might be able to run some type of lab test, but it’s just not said here.
Relly, I feel if your gonna rip the county on this one, give us the methodology you used to come to this conclusion.
The planning commission has discussed this issue at some length. Although I share Dr. Pakurar’s concern about the long-term health of the reservoir, I suggested a different approach to determine whether we had a problem.
A number of years ago I listened to Dr. Pakurar discuss the phosphate load that the reservoir was receiving during construction of certain facilities particularly roads. It appears that when land is disturbed it has a higher potential to pollute than it does at final build out.
I reviewed the model presented by CH2MHill. The model predicts in-lake phosphorous by using certain variables as inputs. These variables include land use categories at full build out (e.g., acres of residential, acres of commercial, etc.); BMP efficiency assumptions, natural vegetative removal assumptions, etc. The variables do not include items such as pollutant loading during construction.
The apparent failure to include the potential increased pollutant loading during construction was a great concern. The water scientists say that once the in-lake phosphorous exceeds a certain level, the lake in dead and cannot be revived. The question of whether we could hit that level before we reach full build out is obviously of grave concern.
On the other hand, because the model is meant to be predictive in nature and not necessarily mimic the process of phosphate loading, it is not necessarily the case that because the model does not use acres under construction as a variable that it has no predictive value. The reservoir itself has the ability to absorb pollutants. If, for example, the model overstates the ability of the BMPs to remove pollutants but understates the ability of the reservoir itself, it might still be an accurate predictor of in lake phosphorous within an acceptable tolerance level.
I requested that the County use our actual data to test the predictive value of the model. We should use the data points we now have as inputs to the model, and compare the model’s predicted results to the lake’s actual results for the same period.
If the requested validation of the model’s predictive abilities shows that it is consistently conservative (that is, it predicts more in-lake phosphorous than actual), or right on, we have no problem utilizing it for our land use decisions. If the model predicts less in-lake phosphorous than actual, we have a problem and will need to look not only at the forecasted BMP removal capabilities, but also the pollutant removal capabilities of the natural riparian corridor and other conditions that impact the pollutant load in the reservoir. Fortunately, however, we also have time to address it.
I understand that the validation will be finished in the spring.
I agree with Dr. Pakurar that we need to protect the reservoir. I appreciate greatly the knowledge he has brought to the commission – it is a fine example of citizen input highlighting an area that needed to be addressed. I am concerned, however, that if we begin a process of testing the removal capabilities of the BMPs, we will divert time and resources away from the ultimate goal, protection of the reservoir. Therefore, I respectfully disagree with his suggestion that we start by measuring BMP efficiency. I hope that he agrees that calibration of the model to start the process is a more efficient use of our resources.
Mr. Gecker,
I want to thank you for taking the time to post a response to this article from Tom Pakurar. It speaks volumes when public officials take time to respond to citizen comments, especially in a forum such as a blog/online journal.
Conaway Haskins
Editor
i wonder what congressional district Dan is in? I would imagine the 7th (Cantor)? Maybe we should start a “draft Dan Gecker” campaign? He is a terrific public official!
1. Though I’m not too surprised, it still amazes me how with all of the scientific knowledge that we have out here, that the countly would decide to implement a plan WITHOUT first testing the water to see what the actual phosphorus levels are. Yes, predicitive models are useful, but I honestly do not see how hard it could have been to actually run some tests on ambient water samples to get a snapshot of what’s really going on there.
2. For various reasons, I can understand why Phosphorus was chosen as the pollutant from which pollution is determined. Still, I think it would be safer to pair phosphorus with, say, fecal coliform measurments. As the article says, the Swift Creek Reservoir is a primary drinking water source for the county. Let me say that again…primary drinking water source!! For that alone, I would expect that the water is monitored for fecal contamination as this is certainly a potential human health issue that could arise.
3. I attended a seminar about a month ago in which the guest speaker discussed retention ponds and how they are thought to be potential cess pools for contamination- though the original intent of these retention ponds was to allow pollutants to settle out resulting in cleaner water. All the more reason for the county to not just focus on phosphorus levels.
With due respect to Mr. Vaught’s efforts to be a one-man Gecker cheering squad, I wonder, since he doesn’t live in the area how he knows any more of what is going on than what his friend tells him? There are contrary opinions. . .
Thank you Mr. Gecker for your comments about “Troubled Waters”. Your thoughtful and timely response within a few hours of the article posting shows you also care.
We are in agreement on a number of key points:
1. We both care deeply about water quality issues.
2. We are both focusing on the long-term quality of the reservoir. The present quality is satisfactory.
3. We are both concerned about the excessive pollution generated during land disturbing activities such as road construction, site preparation and building construction.
4. We are both concerned about the lack of BMPs (ponds) in the developed areas close to the reservoir. While you were chairperson of the Chesterfield Planning Commission, you approved key interim measures to help trap this pollution.
5. We both agree that we need to validate the BMP model used by CH2MHill.
6. We both agree that validation is a one-time effort and not a sustained program.
Let me amplify point(s) five then three:
Point 5
We both agree we need to validate the computer model. I chose BMP efficiency for model verification, as it is the key variable that might affect land use decisions in the Upper Swift Creek area. I asked Scott Flannigan, Interim Manager Office of Water Quality, to make these measurements on a one-time basis in time for the March work session.
To show the interaction of BMP efficiency and land use, I used the data in Attachment D for a couple of simple calculations. Using scenario B (2.0 houses per acre) of Table 5, I calculated the percentage of runoff that needs to be treated at different assumed BMP efficiencies. At an assumed 65% BMP efficiency (which is the current figure being used by the county), we would have to treat only 40% of the runoff to remove enough pollutants to assure an acceptable pollutant level is achieved in the reservoir. This would imply that BMPs close to the reservoir might not be needed in the long term. In a second calculation, I assumed the BMP efficiency was 40% for illustration purposes. We would need to treat 100% of the runoff to remove the same amount of pollutants. The BMPs near the reservoir would be required. At efficiencies below 40% we would need to make changes to BMP design to increase their efficiency or change some land use decisions in the Upper Swift Creek Plan.
I concur with you Mr. Gecker, that we should determine the predictive capability of the model from actual data from the reservoir. However, I feel that determining the efficiency of our BMPs should be done concurrently and can be done without excessive resources. This additional information would aid significantly in the overall picture. In fact, as mentioned above, Scott Flannigan has already agreed to do this within the next several months.
Point 3
We are both concerned about the excessive pollution generated during land disturbing activities. I used a sophisticated lab in Delaware to measure the diameter of over a million soil particles being washed into Little Tomahawk and Tomahawk Creeks. The average diameter was the diameter of a human hair (~0.003 inches). According to the USGS (United States Geological Survey), these particles contain most of the phosphorus pollution on their surface and wash through the environmental barriers like marbles flowing through tennis net. The county data supports this finding.
The use of a coagulating agent like Anionic PAM will make the soil particles so large that the environmental barriers will trap them. There are concerns in using PAM and problems cleaning the traps and basins so as to not return the pollution to the reservoir. These concerns and problems would need to be answered prior to any large-scale use of this technology.
On a separate matter, one question that hasn’t been asked yet – how will this Watershed Management Plan cope with the new state standard for total phosphorus for Swift Creek Reservoir (0.040 mg/l vice 0.050 mg/l)?
There are several ways to determine if the water quality is “crappy”. As you said, you can go out, collect water samples, then run tests on them in the lab. The benchmark for determing if the water quality is good or not is referring to the concentrations and numbers that have been codified in the states’s water quality standards.
Another important component to assessing water quality is conducting biological assessments- collecting samples of macroinvertebrates, fish, and plants in order to determine their present biological condition (or health)in relation to the waterbody they live in. There are several species of organisms that are indicative of pollution in an aquatic environment (high abundances of pollution-tolerant organism paired with low abundances of pollution-sensitive organisms) indicated that there is something going on that is affecting the organisms. The biological response of organisms is most often a cumulative response, meaning that there could be some type of contaminant that has been compromising water quality over some time, and the organisms soon begin to show this.
Pairing laboratory tests with direct field measurements of water quality and biological condition is the most holistic way to determine the status of water quality.