[CPEO-BIF] Douglas, Michigan: Proposed Development over a TCE Plume

[Lenny Siegel <lsiegel@cpeo.org>]

[To download this as a formatted 1 MB, 5-page PDF file, complete with 
photos and Greek letters, go to http://www.cpeo.org/pubs/DouglasTCE.pdf.]

Douglas, Michigan: Proposed Development over a TCE Plume
Lenny Siegel
February, 2007

The tools for evaluating, mitigating, and remediating vapor intrusion - 
the upward migration of toxic vapors from the subsurface - at sites with 
existing buildings are well proven and relatively easy to learn about. 
Nationally, however, much less thought has gone into determining when 
and where it is appropriate to develop or redevelop properties with 
underlying contamination with volatile compounds. In my visits to 
development sites with potential vapor intrusion, I have made four 
recurring observations:

1) Most communities - including local government officials as well as 
grassroots residents - know little about the vapor intrusion pathway or 
how it is successfully addressed.
2) While some states and EPA regions have extensive experience 
investigating and responding to vapor intrusion, others have limited 
experience or outdated policies or guidance.
3) It is less expensive and more effective to investigate, remediate, 
and mitigate vapor intrusion before or during construction, rather than 
after the fact.
4) Plume contour maps that have been drawn to address potential drinking 
water contamination may not be precise enough to make judgments about 
the potential for vapor intrusion.

On February 12, 2007 I visited the Village of Douglas, Michigan, a small 
community located between Kalamazoo Lake and Lake Michigan, southwest of 
Grand Rapids. I spoke at an event sponsored by the Douglas Lakeshore 
Homeowners Association. A small number of City Councilors and Planning 
Commissioners were in the audience.

A developer plans to build 64 single-family homes and 78 triplex units 
on about 15 acres of the former West Shore Golf course. A plume of TCE 
and other pollutants flows from the former Chase Manufacturing plant, 
just to the southeast, to Wick's creek, a tributary of Kalamazoo Lake 
that bisects the development parcel.

In 2003 and 2005, the Michigan Department of Community Health conducted 
health consultations on the property. It reported TCE concentrations in 
the shallow-most aquifer as high as 23,000 parts per billion. The 
author, a Department toxicologist, concluded, "There is no apparent 
current public health hazard via inhalation of indoor air, however the 
hazard is indeterminate for the future." The 2005 Health Consultation 
noted, "Future construction activities in the area near and above the 
plume, however, could result in preferential vapor pathways leading 
toward structures."

In downplaying the potential for unacceptable vapor intrusion, the 2005 
Health Consultation relied upon the Department's Groundwater 
Volatilization to Indoor Air Inhalation Criterion (GVIIC). It found, 
"The maximum TCE concentration found (23,000 ppb) is less than twice the 
Residential/Commercial I GVIIC of 15,000 ppb. Because the magnitude of 
the exceedance is not significant and because a site-specific criterion 
could be greater than 23,000 ppb, it is unlikely that any vapors 
currently originating from the TCE in the groundwater would accumulate 
in indoor air to a degree that would cause health effects. However, 
future underground construction could lead to preferential pathways 
along which vapors could easily migrate."

In my experience, however, any time there is a proposal to build above a 
shallow groundwater plume of TCE where contamination levels are measured 
in parts per million (thousands of parts per billion), there is cause 
for caution. At least, there is cause for additional investigation. But 
the residents and officials of Douglas do not have the experience with 
vapor intrusion to question, or even utilize the Department of Community 
Health's findings.

Finding the Health Department's study useful, but its general conclusion 
unprotective, I've attempted to deconstruct the GVIIC. I'm sorry about 
all the calculations, but that's the approach that most regulatory 
jurisdictions take: Figure out how much pollution is likely to cause 
unacceptable risk, and then allow exposures up to that point.

For starters, it's important to recognize that there is a weak 
quantitative correlation between groundwater sampling results and vapor 
intrusion. Most guidances on the subject call for active soil gas 
sampling to predict indoor air concentrations from subsurface sources. 
(In Douglas, there was extensive passive soil gas sampling, useful in 
some ways, but as the Community Health Department recognizes, not 
sufficient to model vapor intrusion.)

In the absence of soil gas data, California's vapor intrusion guidance 
projects soil gas concentrations by multiplying the groundwater 
concentration by the Henry's law constant for the compound in question, 
and then multiplying by a unit conversion factor. Henry's Law describes 
how a volatile compound partitions into vapor and aqueous phases. I 
looked up the Henry's law constant for TCE, and at the reference 
temperature it's .0091. Thus, 1000 ppb of TCE in groundwater corresponds 
to a soil gas concentration of 9100 micrograms per cubic meter (µg/m3).

Michigan, where average groundwater concentrations are colder, divides 
that number in half. So in Michigan, 1000 ppb of TCE in groundwater 
corresponds to a soil gas concentration of 4550 µg/m3.

To get from soil gas to indoor air, Michigan applied an old version of 
the well-regarded Johnson-Ettinger Model. J & E plugs in a number of 
site-specific factors, or defaults where data in unavailable for some 
measurements, into a formula to calculate alpha, the attenuation factor. 
The attenuation factor represents the indoor air concentration of a 
contaminant (or at least, that portion originating in the subsurface) 
divided by the concentration in soil gas, beneath or near the building.

In my experience, alpha typically varies from .02 to .001 (from 1/50 to 
1/1000). Applied to the 4550 µg/m3 calculated soil-gas concentration, 
that means the indoor air concentration for a site with 1000 ppb TCE in 
groundwater would normally be between 91 µg/m3 and 4.55 µg/m3. As the 
health consultation points out, specific conditions such as dirt 
basements or open sumps might lead to greater levels. More important, 
risk management decisions should be based upon the more conservative 
assumption - that occupants would be exposed to 91 µg/m3 - unless 
additional data shows otherwise.

Michigan apparently has no health standard or screening level for TCE in 
indoor air. A 2003 Health Consultation for the West Shore property 
claims that EPA's reference concentration for TCE is 40 µg/m3. I believe 
that a more realistic cleanup objective would be the 1 µg/m3 action 
level used in residential settings by California, Connecticut, U.S. EPA 
Region 9, and now New York State. (New York actually uses a matrix, but 
in most cases action would be taken at 1 µg/m3.) Note that this action 
level does not fully take into account EPA's 2001 risk assessment 
finding that TCE was 5 to 65 times as hazardous as previously believed. 
EPA officials withdrew the risk assessment in 2003.

Thus, a 1000 ppb shallow groundwater measurement would suggest an indoor 
air concentration 91 times higher than the acceptable level with 
standard residential construction. Calculated in reverse, a 1 µg/m3 
indoor air threshold implies a soil gas screening level of 50 µg/m3 in 
the absence of site-specific data suggesting that greater attenuation 
will take place. In turn, that implies a groundwater screening level of 
11 ppb in locations with cold groundwater, not 15,000 ppb.

Much of the proposed construction site appears to have TCE in 
groundwater above 11 ppb, but it's hard to know which proposed buildings 
are likely to be at risk. The rule of thumb for vapor intrusion 
potential is that buildings or building sites laterally within 100 feet 
of subsurface contamination should be evaluated. In the field, however, 
this may be confusing. The West Shore site has a TCE contour map. Soil, 
groundwater, and surface water have been sampled with sufficient density 
to protect potential drinking water, but in my opinion there are not 
enough monitoring wells to delineate areas with serious vapor intrusion 
potential. In fact, there's one isolated sample, near where Wick's Creek 
enters Kalamazoo Lake, where the TCE level is 140 ppb, surrounded by 
contours suggesting 5 ppb. Further testing is necessary to determine 
whether that reading was indeed accurate, and if so, why it's there.

Furthermore, though there is a pump-and-treat system installed at the 
source area, I have seen no evidence that the plume will not spread 
further. High levels (1600 ppb) of TCE in Wick's Creek, at the head of 
the plume, suggest a continuing flow from the source. If the plume moves 
laterally, more areas will be at risk of vapor intrusion.

Recommendations

Based upon my visit to Douglas, I cannot say for sure that the new 
buildings, if constructed, will have unacceptable levels of vapor 
intrusion. I have concluded, however, that there is enough of a problem 
to conduct further investigation. I recommend the following:

* Additional monitoring wells should map the full extent of the TCE 
plume over time.

* Active soil gas sampling should be conducted at each proposed home site.

* Target indoor air concentrations of TCE should be no higher that 1 µg/m3.

* The footprint for the proposed development should be altered to move 
residential units away from higher concentrations of TCE in the 
subsurface. If a protective site plan cannot be designed, then the 
development should be rejected.

* All structures should be built with vapor membranes and subslab 
depressurization systems. The latter may be operated in passive mode if 
subslab soil gas does not exceed levels corresponding to the indoor air 
screening level.

* After construction, but before occupancy, each unit should be tested, 
including measurements of subslab soil gas, indoor air, and ambient 
nearby (outdoor) air. As long as soil gas contamination at any structure 
remains above levels corresponding to the indoor air targets, such 
sampling should continue periodically.

* Institutional controls should warn building occupants of potential 
exposure, prevent construction above the highest levels of groundwater 
contamination, forbid perforation of vapor barriers, and require 
periodic monitoring for perforation caused either by natural or human 
activity.

Michigan should develop more realistic tools for evaluating vapor 
intrusion risk. All government agencies with jurisdiction over 
development and vapor intrusion decisions where buildings are proposed 
above plumes of volatile pollutants should adopt an approach that 
incorporates the type of recommendations I have made for Douglas.


--


Lenny Siegel
Director, Center for Public Environmental Oversight
c/o PSC, 278-A Hope St., Mountain View, CA 94041
Voice: 650/961-8918 or 650/969-1545
Fax: 650/961-8918
<lsiegel@cpeo.org>
http://www.cpeo.org


_______________________________________________
Brownfields mailing list
Brownfields@list.cpeo.org
http://www.cpeo.org/mailman/listinfo/brownfields