Free Additional Attachments to Main Document - District Court of Arizona - Arizona

- 4 -

I I
1
I

sorbed state and dissolved state and dissolved state and vaporized state. Conditions allowing such conversion simply do not exist at the PGA site. The conversion also assumes a single discrete value for the soil-water partition co-efficient based on an organic fraction in soil of 0.5%. Much of the matter in the soil column, however, is likely to be sand and gravel with little organic content. The ADWR model has assumed no retardation (Kd=0) for the aquifer.' As the Kd approaches 0.0, the ratio of soil sorbed TCE ^to soil vapor TCE approaches 0.0. Thus, estimates of TCE residuals in the soil would be much smaller if the Kd equals 0.0 as estimated by ADWR. Kaiser believes that a more .accurate estimate of TCE residuals would be between 20,000 to 30,000 pounds; of which 1,000 to 5,000 pounds would be sorbed onto the soil particles or contained in interstitial water and between 15,000 to 29,000 pounds present as vapor in the pore space in the soil. Because EPA has overestimated the amount of TCE in the soil and because no federal or state standards for the cleanup of soil contaminated with VOC's exist, Goodyear advocates the adoption of specific methods and criteria to address the field conditions as they are encountered during the cleanup process itself. This approach will require a consensus on acceptable cleanup levels based on more realistic estimates of soil contamination. To further emphasize this concern, Goodyear notes that if EPA proceeds with its current soil-vapor extraction rate, this remedy, as presently conceived, may produce the undesired outcome of extracting TCE up from the groundwater through the soil. Because of the uncertainty over the actual mass of residual TCE in the soil and the operable migration pathways, Goodyear recommends a "decision tree" for determining when TCE soil evacuation can be terminated. See attachment B. While Goodyear understands that EPA has agreed to this approach in concept at various technical meetings, the parties have not yet agreed on the actual criteria levels that would result in terminating the soil extraction process at a certain point or the target area for cleanup. A meeting to discuss these various target -cleanup levels has been scheduled for July 7, 1989. If this_meeting achieves any consensus on the decision-tree approach, Goodyear reserves the right to comment on the target cleanup levels established for the decision tree. Until these target cleanup levels are established, neither EPA nor Goodyear can estimate the potential costs involved with the soil evacuation remedy or, more specifically, whether soil extraction is a more cost effective remedy than capping.

· ·
_ I P |jr · ^ I P ·

|

£ · TM
I M _ R ~ I

m I _ *I ·

I I
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 1 of 28

I I I I I I I I I I I I I I I I I I I

- 5 -

Groundwater

The Public Comment Draft RI/FS offers four alternatives to address the remaining groundwater contamination at the site: (1) pumping and treating at an accelerated rate using existing wells to meet existing standards; (2) pumping and treating at an accelerated rate using new wells to meet existing standards; (3) pumping and treating at an average rate using new wells to exceed existing standards; and (4) pumping and treating at an accelerated rate to exceed existing standards.
At the outset, Goodyear notes that the existing maximum concentration level ("MCL") under the Safe Drinking Water Act for TCE is 5 parts per billion (ppb).._ As the Agency's risk range for Superfund remedies is 10 to 10 and as the proposed revisions to the National Contingency Plan ("NCP") no longer require consideration of alternatives that exceed standards (53 Fed. Reg. 51506, December 21, 1988), Goodyear believes that alternatives 3 and 4 would exceed the requirements of CERCLA, and would not be cost effective.
Alternatives 1 and 2 both focus on meeting existing standards, but require pumping at an accelerated rate. Goodyear believes, for reasons discussed in greater detail below that the preferred alternative from a cost effective perspective should be one that pumps the contaminated groundwater at an average rate using existing wells. EPA's discussion of groundwater in the RI/FS is flawed in several respects. Principally, EPA's final remedy has failed to take into account how contamination in Aquifers B and C will be eliminated by the operation of the interim remedy in Aquifer A. Second, EPA has failed to establish this pumping rate using any valid groundwater model. Indeed, it appears that the Agency has failed to use any existing groundwater contamination model, such as the ADWR model, at all. To our knowledge, no work is scheduled to refine the ADWR conclusions. Because the Agency has failed to use the ADWR model (or for that matter any valid model), the current RI/FS discussion of groundwater contamination is completely inadequate as a basis for selecting a remedy. Consequently, it has been virtually impossible for Goodyear to evaluate the selected alternatives. Goodyear urges EPA to recalculate the groundwater scenario using the ADWR model and specifically reserves its rights to comment on EPA's revised groundwater discussion. Finally, recognizing that federal standards for TCE exist, Goodyear believes there should be a process to terminate pumping and treating groundwater after certain action levels are met.

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 2 of 28

- 6 -

I I
· · I ·

Although EPA apparently favors air stripping over carbon treatment as part of this overall final remedy, Goodyear nevertheless desires to emphasize the efficacy of air stripping. As EPA is aware, Goodyear sought modification of the Consent Decree for that part of the operable unit addressing pumping and treating Subunit "A" because it felt that carbon treatment far exceeded the applicable, relevant and appropriate air quality standards for this area, and was consequently not a -cost effective alternative. This position was based on a sophisticated risk assessment of area-wide air emissions that was performed by ICF Technology, Inc. Goodyear is including this risk assessment and a copy of its correspondence to EPA on this subject as part of these comments (Attachment C). Finally, none of the groundwater alternatives currently envision that any of the treated groundwater would be used for any purpose other than reinjection. If other viable uses become apparent during the implementation of this remedy, Goodyear notes that water rights administered by the State of Arizona will have to be dealt with, and that a reasonable degree of flexibility should be factored into implementation of the final remedy to meet these potential requirements. Conclusion Goodyear and its consultant have reviewed the June 7, 1989, RI/FS for the PGA site and have found the document to be flawed in several major respects. Moreover, as the June 7, 1989, draft final RI/FS currently exists, it provides an inadequate basis for commenting on, or selecting, a cost-effective remedy for the PGA site. Goodyear would be happy to answer any questions that the Agency may have on these comments.

_ ~

· I m I

I
« ·

Sincerely,

·

Manager Environmental Engineering
David L Chapman pah

«

Attachments

I I I
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 3 of 28

I I I I I I I I I I I I I I I I I I I

89187RDS0690
ATTACHMENT A
TECHNICAL COMMENTS

P 2-37. The discussion of metals in soil encompasses all metal data generated regardless of the probable source of the metal
or background levels in the area of the PGA. This discussion is

particularly misleading with respect to arsenic since natural
arsenic levels are sufficiently high to generate risk levels of concern and there is no record of use of arsenic on site. The failure to segregate site-related contaminants from naturally occurring ones results in soil ingestion risks being driven by

arsenic which cannot be remedied since it is ubiquitous in the native soil. A few statements to .this effect would prevent the reader from being misled about site-related risks. P 2-40. No attempt has been made to differentiate Cr(III) from Cr(Vl) or leachable chromium from fixed or insoluble chromium.
As a consequence, total chromium values are reported and used for the purposes of estimating public health impacts even though availa-

bility and valence state greatly effect the nature and magnitude of risks.

P 2-54. An estimate of the inventory of TCE in soil of 450 Ibs was made from existing soil boring data. When an amount equal to this was removed during pilot soil evacuation work, a second
estimate was attempted using soil vapor data. The latter estimate The

came to as much as 115,000 Ibs depending on the assumptions made
with respect to vertical distribution of TCE residuals. TCE, water-bound TCE, and soil vapors.

algorithm used to calculate total soil TCE mass from soil vapor data relies on an assumed equilibrium condition between soil-sorbed
For simplification, a single partition value was used to calculate soil/water ratios. This value was also used in conjunction with the Henry's law constant to predict soil/vapor ratios. The partition value selected was based on a prescribed soil organic level.
Use of any value other than 0.0 contradicts the assumptions made by the Arizona Department of Water Resources (ADWR) in preparing the

ground water model for the site. While the ADWR assumption is probably overly conservative, an assumed constant value throughout a 60 foot depth is also misleading. It is highly likely that deep sands and gravels will have little or no affinity for the TCE.
Hence, use of the algorithm will overpredict soil-bound TCE from the existing TCE vapor data.

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 4 of 28

- 2 -

I I
· j§

The likelihood of overprediction is illustrated by analysis of the existing data. The highest soil vapor values were found in the area of the soccer field. Borings in that same area revealed no measurable TCE in subsoils. Hence, the algorithm is assigning TCE at significant concentrations to soils that have no evidence of contamination. Similarly, soil vapor readings from the area of the Phillips well were as high as 1.7 ug/L even through this property is 3 miles from the site. These vapor levels are either derived from other sources or reflect the ground water plume at that point. There is no evidence that they are associated with soil contamination.

_ · ·
· ·

·

4.

P 2-61. Calculations are made to estimate the total volume of soil in excess of Arizona Department of Health Services (ADHS) soil action levels. These volumes are meant for use in determining the cost of remedial action. The.volumes are misleading, however, since they encompass all soils and subsoils with VOC concentrations in excess of the action level. The action level was devised for surface soils, not deep subsoils. Most TCE residuals lay 20 to 30 feet below the surface. Alternate action levels are needed for these soils on the basis of their ability to affect ground water
quality. --

I ·
· j§
--

·

5.

P 2-61. Vadose zone calculations are made suggesting that 16,000 Ibs of TCE will move to the ground water in 20 years. _These calculations are based on an assumed recharge that is without documentation. They also appear to take no recognition of unsaturated zone transport times. Using EPA time-of-travel algorithms, recharge at 0.32 in/yr would take 117 years to move 20 feet downwind under current conditions. If the TCE has a partition coefficient of 0.49 L/Kg, its travel time would be retarded by a factor of 2.6 and hence would be 304 years. '

·

«
I
·

6.

P 3-46. The risk calculations are based on current TCE concentrations at various wells around the PGA site. No attempt was made to use the ADWR model to see how those concentrations will change over time. Since cancer risks are based on 70 years of exposure, the assumption is tantamount to saying that the ground water at any one well will not see any appreciable change in TCE concentrations over a 70 year period. That is unrealistic. Simple application of plume size and the estimated velocities in the affected aguifier suggest the concentrations will drop an order-of-magnitude in seven years. If that does occur, the actual risk at the site will be one tenth that predicted in the RI/FS. The analysis also fails to consider the affects of the Operable Unit 16 remedy which is currently under construction.
P 3-46. Well logs from construction of extraction and injection wells for the Operable Unit 16 remedy suggest that the boundaries between subunits A, B and C are not always distinct and then in some areas, the units may be indistinguishable. Previous descriptions imply rather clear cut interfaces which is misleading.

· jj
^ · ·

·
_ ·

7.

· ·
·

I
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 5 of 28

I I I I I I I I I I I I I I I I I I I

8.

P 4-1. Risk estimates for suspended particulate are based on current emission rates being sustained over a 70 year period. A simple calculation shows that in a period of 7 years, the finer suspendable particles will be depleted to a depth of 1.5 cm. This in effect will leave the larger, nonsuspendable particles to armor the surface and minimize further resuspension. AS a consequence, risks will actually be an order-of-magnitude less than predicted. The bulk of the risk from suspended particles is attributable to arsenic in the soil. Since arsenic is naturally present and not a site-related contaminant, the risk calculations provide a misleading picture of incremental risk and risks that can be addressed by a site remedy. All soils in the area pose the same level of arsenic driven risk.

P 5-41. The ultraviolet-ozone oxidation process is dismissed prematurely. Recent studies show,this process to be very effective in removing organic contaminants from water. In areas where air stripper emissions must be treated with carbon, the UV-ozone process can be cost competitive. tt 10. P 6-13. Target Area 1 is inappropriate. ADHS action levels were designed to address surface soils, not subsoils 20 to 30 feet beneath the surface. IP a target area is to be defined using ADHS action levels, it should be based solely on TCE concentrations in surface soils.

b»

Target Area 3 is not based on any defensible rationale. No attempt is make to relate soil vapor concentrations to site risk values. Since soil vapor results do not correspond with subsoil concentrations of TCE, the use of soil vapor to delineate a target area is illogical. At a minimum soil vapor values should be converted to equivalent soil concentrations and the target area defined on the basis of the latter.
11. P 6-21. The discussion of the capping alternative appears to contradict other portions of the RI/FS. The implication of this discussion is that recharge is insignificant with respect to TCE movement. And yet, the calculations of vadose zone movement and soil residual effects on ground water quality are based on a prescribed recharge rate of 0.32 in/year. Either recharge is driving TCE downward and capping will minimize or prevent this migration, or recharge is insignificant and sub-soil contamination can be left in place without remedy.
P 8-2. A very simplistic analysis is employed to calculate aquifer flushing times. This is difficult to explain since a great deal of money has been spent developing a sophisticated ground water model to predict flushing times and plume movement. The RI/FS should rely on model results for flow and transport predictions.

12.

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 6 of 28

- 4 -

I I
m · I · · _ ·

13.

Figure 8-3. The contaminant plumes have been depicted as large areas joining points wherever VOCs were detected in ground water without regard to the relative concentrations at adjoining wells. Geostatistical analysis should be used to prepare these plots. The relatively high values at the Phillips well and lower concentrations at points between Phillips and the site open the possibility of multiple sources or a more concentrated transient plume that is passing by Phillips to be followed by water of better quality. Since risk was estimated on the basis of continued exposure to current levels, a better characterization of the actual plume could have a big impact on conclusions concerning risk and the nature of required remedies. P 8-30. Simple equations are applied to estimate ground water travel times. The ADWR model was developed to provide much more accurate predictions of travel times and should be employed for that purpose. P 8-36. A simplified approach is taken to calculate the time required to achieve clean up. Once again, the ADWR model should be employed for this purpose. Furthermore, the estimates do not consider' implementation of the Operable Unit 16 remedy or continued inputs from the vadose zone. This static evaluation of aquifer cleansing is unrealistic. P 9-7. The analysis of end use options for the treated ground water does not give ample consideration to problems associated with water rights. A brief discussion is given of water rights after discharge. However, it is not clear if the water is currently owned by a party who can subsequently dictate where the treated · water should go. If the City of Phoenix or some similar entity owns the ground water, they may not allow it to be delivered for private or public use by other entities. A much more jthorough evaluation of ownership is required before discharge alternatives can be considered.

14.

·

15.

· I · jj -- · * ·

16.

·

17.

P 10-1. The options for design of the ground water extraction system should be evaluated using the available models of the local ground water. A simple water balance approach fails to consider the Operable Unit 16 remedy and the complexities of the aquifer. With sophisticated tools readily available to support the analysis, reliance on simple approaches is indefensible.

· · ·

I I I I
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 7 of 28

I I I I I I I I I I I I I I I I I I I

ATTACHMENT B

SOIL EVACUATION TERMINATION LOGIC FLOW

ENTER

IS GW (TCE)

>5ppm?

DOES ESTIMATE TCE MASS EXCEED LEVEL REQUIRED FOR GW (TCE) >ARARS?

SHUT DOWN SYSTEM MONITOR GW

AND SOIL VAPOR

SHUT SYSTEM DOWN ALLOW SOIL VAPOR TO RETURN TO STEADY STATE

DOES GW (TCE)

RETURN TO LEVELS
>ARAR?

SOIL VAPOR (TCE) HIGH ENOUGH FOR CALCULATED MASS TO

EXIT

CAUSE GW (TCE)
>ARAR?

YES

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 8 of 28

I I I I I I I I I I I I I I I I I I I

ATTACHMENT C

LAW OFFICES
HENRY L. DIAMOND ALBERT J. BEVERIDGE. 31 GARY H. 3AISE HAROLD HIMMELMAN CHRISTOPHER H. 1UCKLEY, JR.

BEVERIDGE & DIAMOND, P. C.
1333 NEW HAMPSHIRE AVENUE, N. w. WASHINGTON, D. C. 2OO3S
(2O2) 328-O2OO

DEAN H. CANNON

VIRGINIA s.AiaRECHr
DONALD J. PATTERSON. JR. STEVEN F. HIHSCH 3RNOA MALLORY MARK A. TURCO THOMAS C.JACKSON KATHRYN E. SZMUSZKOVIC2 HAROLD L. SECALL PETER J. SACRIPANTI OUSTIN P. OROWAY KENNETH S. KAUFMAN CHARLES F. GAUVIN

STEPHEN L.GOROON
ALEXANDER W. SICRCK JOHN N. HANSON ANDREW E.MISHKIN WILLIAM N. HEOEMAN. JR. AIN£S GWATHMEY, HI YNTHIA A. LEWIS KARL S. BOURDEAU JOHN S.CUTTMANN

§

DON a. SCROCGIN

TELECOPIER (2O2) 323-0334 TELEX 372SS38 3EVOIA WSH

MARC A.ZEPPETELLO
LISA M.MARTIN'

GARY J. SMITH B E N J A M I N F. WILSON ROBERT SPACER RICHARD S, OAVIS THOMAS RICHICHI GUS 3AUMAN J. DANIEL aCRRY

CHERI C. WOLFF"
CHRISTOPHER W. MAHONEY STEVEN M.JAWETZ

JOHN O. GIGLIO

SUITE 1202
OF COUNSEL

SEVERIDGE & DIAMOND ONE BRIDGE PLAZA
FORT LEE, N. J. 07O2*

ELLIOT 0. EDER

SUSAN j. KOROSTOTF
AARON H.QOLO3ERG
F. CRAIG THORLEY CRAIG O. CALLI JOHN T. S. WILLIAMS* SUSAN K.FTHHON

JOHN FRENCH m
R.THEODORE HUME CONSULTING ECONOMIST

IOI PARK AVENUE NEW YORK, N. Y. IOI7S

OEBRA L. HOTHBEHGDOUGLAS ·. WEINF1 ELD-

(212) SS7-33B5

(2OI) SSS-3162

JIM j. rozzi

SHELLE7 V. LUCAS
WRITER S DIRECT DIAL NUMBER
·NOT ADMITTED IN O. C^

(202)

429-2726

December 7, 1988

Jerry Clifford Assistant Director of Superfund Region 9 U.S. Environmental Protection Agency 215 Fremont Street
San Francisco, CA 94105

Dear Mr. Clifford:

Goodyear appreciated the opportunity to meet with you, your staff and representatives of Region IX's Air Programs on November 30, 1988, to review the current project at the Litchfield PGA site and to discuss the extent to which carbon treatment should be required as part of the operation of the first seven extraction wells associated with this operable unit. The dialogue was informative and constructive, and this letter follows-up on those discussions with a specific proposal.

As we indicated in that meeting, Goodyear requests, pursuant to Paragraph XXV. of the Consent Decree that was entered on October 31, 1988, that a modification to Paragraph VII.C.5.(c) of the Consent Decree be considered by EPA. This request is based on information that was not available during the time in which the Consent Decree was negotiated. It was discussed at our meeting and is being formally presented to you by this letter.

oe;
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 9 of 28

BEVERIDGE & DIAMOND, P, C.

Mr. Jerry Clifford
Page 2 December 7, 1988

Goodyear recommends, as a result of this new Information (which is discussed in greater detail below) that any requirements for carbon treatment that may exist J.n the current Consent Decree be formally eliminated. In the alternative, Goodyear requests that the issue of carbon treatment be deferred for at least two years, until the final remedy has been selected and designed. As we indicated at the November 30, 1988, time is very much of the essence with respect to this request for modification as Goodyear is required, pursuant to Paragraph VII.D.5. of the Consent Decree, to submit a proposed final design to EPA by January 11, 1989. This issue relates significantly to how that design will be developed and presented.

B m · ·
--

·

" ·

As you are aware, Goodyear prepared a set of comments relating, among other things, to this issue which was submitted to the Department of the Justice ("DOJ") for consideration during the period of time in which the Consent Decree had been lodged and made available for public comment. Regretably, these comments were not considered before the Consent Decree was signed and filed on October 31, 1988. However, we are incorporating these comments as an enclosure to this letter (Enclosure 1), as they address Goodyear's position concerning the applicability of Arizona State requirements to the issue.

I
·

·»
· jg

·

In this letter, we cited a number of EPA documents ("Guidance on Feasibility Studies under CERCLA" dated April, 1985, page B-19 and the "Superfund Public Health Evaluation Manual" dated October, 1986)to observe that the risk range of 1Q-4 to IQ~7 is used by EPA to determine adequate protection of public health and the environment when there are no national standards that otherwise would establish an appropriate level of cleanup. We also note that this risk range has been presented by EPA for comment in its proposal for inclusion in revisions to the National Contingency Plan which were announced on November 17, 1988; but EPA also has requested comments on reducing this risk range from 10~^ to 10"^ to a range of 10~4 to 10~^. We indicated in our October 19, 1988, letter to DOJ that "the health risk of exposure to the air emissions from the air strippers, without carbon absorption, is no worse than 10~^, and also stated that Goodyear would be developing additional data to support this position which would be forwarded to you under separate cover. This data was discussed with you on November 30, 1988, and is

· ·
· f§

« ·m I ·

· |

_

incorporated into this letter as Enclosure 2.
Goodyear's consultant, ICF, Inc., performed a risk

·

assessment of the release of volatile organic compounds
("VOCs") into the air with air stripping but without carbon absorption. Taking trichloroethylene ("TCE") alone, the cancer

·
·

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 10 of 28

I I I I I I I I I I I I I I I I I I I

B E V E R I D O E & D I A M O N D , P. C,

Mr. Jerry Clifford Page 3 December 7, 1988

risk is less than 1 x 10~7. Indeed, at 100 meters from the extraction wells, the cancer risk is .74 x -10"7, and then diminishes significantly as one proceeds away from the site at 300 meter, 700 meter and 1500 meter increments. When one incorporates the cancer risk of dichloroethene (DCE) into the equation, the combined risk for both substances is less than 10~6 (actually, .24 x 10~6 at 100 meters). As we indicated in our meeting, the fundamental objective of Superfund cleanups is to adequately protect public health and the environment. While we recognize that a great deal of discretion exists within EPA in determining what is adequate protection, and that this discretion is in large part bolstered by somewhat conflicting criteria within the Superfund Amendments and Reauthorization Act ("SARA") (which EPA describes in the preamble to the proposed revisions of the NCP as a "dynamic process,") we submit that the fundamental objective should be adequate protection of public health in the most cost effective manner. EPA guidance and proposed revisions to the NCP subsequent to enactment of SARA have provided various bases for determining what is adequate protection of public health and the environment. First, EPA uses applicable, relevant and appropriate federal and state requirements ("ARARs"). While there are technology based standards relating to Arizona's State Implementation Plan ("SIP") for sources emitting TCE contaminants, Goodyear does not believe that they can be interpreted as requiring carbon absorption in the PGA case for the reasons discussed in greater detail in Enclosure 1. However, those standards are not health based; rather, they are technology based. As to health based standards which also are intended to address the adequacy of public health protection, we believe that we have demonstrated to you and your staff that air stripping from these extraction wells will more than adequately achieve that purpose without the need for carbon treatment.

During our meeting, mention was made of a developing policy within EPA concerning the requirements for Superfund remedies located in non-attainment areas. While Goodyear appreciates the issue and is very sympathetic to the air pollution problems that exist in certain areas of the country, including Phoenix, we can find no legal basis for the application of a non-existent (but emerging) EPA policy in this area. In addition, we question whether or not such a policy would fall within the cost-effective criteria of SARA, if it can be

demonstrated that air emission controls are not required to adequately protect public health and the environment. During our meeting, concern was also expressed as to whether remedies established at other Superfund sites within

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 11 of 28

BEVERIOGE & DIAMOND, P. C.

I
·

Mr. Jerry C l i f f o r d Page 4 December 7, 1988

· . · *

the Phoenix, Arizona area would be brought into question if Goodyear were not required to install carbon absorption as part of the treatment of the extracted groundwater. We have

examined the sites that were mentioned, and offer the following factors to distinguish each of them.
At the Indian Bend Wash Superfund site, we are informed that the City of Scottsdale is operating an air-stripping tower (at well no. 6) which is not equipped with emission control equipment. According to representatives of Arizona's Department of Environmental Quality and the City of Scottsdale, the City is allowed to operate this air stripper without emission control equipment because it emits less than 40 pounds of VOCs per day, which is the threshold level for emission controls under Maricopa County regulations. The City is planning five to six additional air-stripping towers where it does intend to install air emission control equipment. According to the City, however, the reason for installing air emission controls is due to the residential character of the area and the request by neighboring citizens for air emission control equipment, not because such controls would be required pursuant to any regulation. Also, we are informed that the City's future towers will be part of a final (as opposed to an interim) remedial plan.

·

M ·

· " · ·

« ·
· ·

Another federal Superfund site in Maricopa County, Motorola 52nd Street, currently is in the planning stage. The Draft Remedial Action Plan prepared by Dames & Moore (June 24, 1988), indicates that Motorola is considering use of carbon absorption emission controls in its air-stripping towers, although no firm decision has been made. Dames & Moore anticipates that the air-stripping towers will be located at Motorola's plant facility. Because the plant facility already is subject to air emission requirements, representatives of Dames & Moore believe carbon absorption may be necessary for the air-stripping towers to keep total plant air emissions below regulated levels. Motorola's plan appears to be in its early stage (10%) plan, and Dames & Moore indicates that plans to use carbon absorption may change as the design progresses.

|
« · TM
· ·

· | _ · · ·
·
_ · m

EPA apparently has allowed other air-stripping towers to operate without emission controls in Pima County, Arizona so long as those facilities met applicable air quality regulations. We have been informed that the applicable regulation in Pima County requires emission controls at "miscellaneous" sources if VOC emission exceed 2.4 pounds per day. Representatives of the City of Tucson and Hughes Aircraft informed us of currently operating air-stripping towers at federal Superfund sites that emit below the 2.4 pounds requirement. Neither of these towers has been required to

I
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 12 of 28

I I I I I I I I I I I I I I I I I I I

BEVERIOGE & DIAMOND, P. C.

Mr. Jerry Clifford Page 5
December 7, 1988 install air emission controls, as VOC emissions did not exceed

the regulatory threshold.
It is significant that emission controls have not been required for air stripping units at state-lead remedial action sites in Maricopa County where less than 40 pounds per day of VOCs are emitted. We provided you with the original of a printout obtained from the Maricopa County Health Department Bureau of Air Pollution Control. (A copy is attached for your convenience, Enclosure 3). The printout listed all stripper facilities with potential emissions of less than 40 pounds per day of volatile organic compounds which obtained air permits during the calendar years 1987 and 1988. Of the sixteen listed sites (eight of which emit VOCs at levels equaling or exceeding the maximum of 10 pounds per day expected from Goodyear's proposed facility), only one is equipped with air emission controls: an air stripper at a Texaco service station at 305 East Thomas Road. That cleanup is managed by W.W. Irwin, Inc., of Long Beach, California. Long Beach is located in California's South Coast air quality control district, which district requires such emission controls for any facility emitting over one pound of VOCs per day. W.W. Irwin has informed us that it simply did not investigate the applicable Maricopa County regulations and proceeded under the erroneous assumption that a one pound per day limit applied in Maricopa County. Such emission controls were neither required nor requested by Maricopa County's Bureau of Air Pollution Control.

Maricopa County does not maintain air monitors in or near the City of Goodyear or the Phoenix Goodyear Airport. It would be difficult unequivocally to claim that the area immediately surrounding the Phoenix Goodyear Airport would be in attainment for ozone, if a monitor were placed there. Because it is largely a farming area and the prevailing wind and weather patterns in the Phoenix area are from west to east (thus tending to carry ozone and other pollutants from the metropolitan area eastward), it would not be unreasonable to assume that Goodyear, which lies to the far west of Phoenix, would be in attainment with regard to ozone levels, were a monitor present at the airport.

In summary, this modification is based on the fact that we did not have a risk assessment of the air emissions that would occur without carbon absorption when we negotiated this Consent Decree. Indeed, many of the numbers that form the basis for this risk assessment were not developed until the completion of the conceptual design. We believe that this information clearly supports a view that, at a minimum, Goodyear should be given the opportunity to defer any requirements for carbon absorption until the overall requirements for treatment become better known in the final remedy.
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 13 of 28

BEVERIOGE & D I A M O N D , P. C.

I
I

Mr. Jerry Clifford Page 6 December 7, 1988

As was stated at the meeting, we recognize that many of the requirements for this operable unit form the "cornerstone" of the final remedy. But, at the same time, as Section 121(d)(4)(A) of SARA indicates, ARARs and related requirements do not have to be considered as part of an interim remedy. Equally relevant in these considerations is the need to maintain a continuous and constructive relationship between EPA and Goodyear over a considerable number of years. The constructive relationship that we have sought to establish with EPA in working on this project is based on the elements of fairness and reasonableness that we believe now prevail, and will continue to prevail in the long term future in which both of us will be associated with this project. To request a potentially responsible party, such as Goodyear, to invest $300,000 - 500,000 in capital expenditures for carbon absorption at these first seven wells, as well as an estimated average of $90,000 per year for routine maintenance when adequate protection to public health and the environment already exist through air stripping based on EPA's existing criteria, simply does not seem right.
Goodyear hopes that you take this request for a modification in the spirit in which it is given. While we recognize that the request follows closely on the heels of the final Consent Decree, we would emphasize that the data to support it did not emerge until completion of the conceptual design, and that Goodyear has fulfilled every commitment in the Consent Decree since it undertook the initial work in April, 1988.

· ·

_ · · · jj
_ · "

· · M · · ·

· m
m

Thank you for your consideration of this request and we continue to look forward to working with Jeff Rosenbloom and other EPA staff in the future. Sincerely yours,

_ · ·

William N. Hedeman,
WNH/b

Enclosures cc: Alexis Strauss (with enclosures) Jeff Rosenbloom (with enclosures) Hugh Barroll (with enclosures) Barry Sandals (with enclosures) David Chapman (without enclosures) Mark Phillips (without enclosures) Takashi (Wally Ito) Ito (without enclosures) John Hill, ICF (without enclosures) Rolf R. von Oppenfeld, Fennemore Craig (without enclosures)

I I
· · · jj

g ·

1081g
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 14 of 28

I I I I I I I I I I I I I I I I I I I

THE GOODYEAR TIRE & RUBBER COMPANY

1144 E. Market St. Akron, Ohio 44316 (216) 796-3084

October 19, 1988

Assistant Attorney General Land & Natural Resources Division Tenth and Constitution Avenue, N.W. Ben Franklin Station P.O. Box 7415 Washington, D.C. 20044-7415 Re: United States v. The Goodyear Tire & Rubber Company; D.J. Ref. 90-11-2-186 Comments of The Goodyear Tire & Rubber Company on Proposed Consent Decree

Gentlemen:

The Goodyear Tire & Rubber Company (hereinafter "Goodyear") hereby submits its comments on the referenced proposed Consent Decree addressing the operable unit remedial action at the Phoenix-Goodyear Airport Superfund site in Litchfield Park, Arizona. Goodyear respectfully requests the Department of Justice ("DOJ"), after consultation with the Environmental Protection Agency ("EPA") to file these comments with the court along with a concurrent motion to enter a judgment that is consistent with modifications to the' draft consent decree as presented below. Before addressing the two areas of major concern to Goodyear in the proposed consent decree, Goodyear would like to make a preliminary observation. Goodyear responded to the EPA's notice letter concerning its willingness to enter into discussions with EPA concerning the negociation of the proposed consent decree in a timely manner. However, the unique circumstances at this sice, and in particular the involvement of the U.S. Navy as a potentially responsible party ("PRP") along with Goodyear, raised significant legal and policy issues that required resolution within the Department of Defense concurrently

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 15 of 28

Assistant Attorney General
October 19, 1988

Page 2

with Goodyear ' s negotiations with EPA. Resolution of several of

these issues occurred through an alternative dispute resolution ("ADR") process between Goodyear and the Department of Defense, represented by the U.S. Army Corps of Engineers, Omaha District, which did not conclude until May 21, 1988. During this period of time from September 3, 1987 (the
date on which the notice letter was sent to GoodyearJ to May 21,

1988, (the date on which Goodyear signed the consent decree), Goodyear initiated the first phases of the work described in Paragraph VII of the consent decree, and has submitted all work elements required by Paragraph VII as if the consent decree were effective and binding. This was done to ensure that Goodyear 's (and subsequently the Department of Defense's) commitment to address the problems at this site would not be .unnecessarily delayed while waiting for full resolution of the exact provisions of the filed consent decree. However, as Goodyear proceeded into the conceptual design of this operable unit ("OU"), it became aware of other alternatives associated with conducting this interim remedy not known to all of us (EPA., Goodyear, and the Department of Defense) during the development of the operable unit feasibility study ("OUFS") that yielded EPA's Record of Decision ("ROD") on September 29, 1987.

M g _ · ·

We emphasize this point to indicate that the changes m requested in the proposed consent decree that are outlined below are changes that have evolved during Goodyear ' s voluntary perform mance of the work elements displayed in Paragraph VII of the proposed consent decree. While we do not believe that any of these changes to the proposed consent decree would trigger, as a « condition precedent, a change to the ROD, we do believe that the M changes would further clarify the intent of all of the parties, would be consistent with the ROD and the requirements of the Comprehensive Environmental Response Liability and Compensation · Act ("CERCLA"), as amended by the Superfund Amendments and · Reauthorization Act of 1986 ("SARA"), and .also would provide adequate protection to public health and the environment in a · cost effective manner. _ |

A.

Disposal of Treated Water;

m

Our first proposed change relates to Paragraph VII.C.5(a) of the proposed consent decree which reads as follows:
All water from the groundwater extraction system will be treated and reinjected. Treatment shall assure that reinjected water will meet federal and state standards for treatment »

·

I
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 16 of 28

I I I I I I I I I I I I I I 1 I I I I

Assistant Attorney General
October 19, 1988 Page 3

plant discharge levels prescribed in Table I of the 1987 ROD. During start-up activities, extracted water to and from the treatment plant will be checked on a schedule as provided for in the Operations and Maintenance Plan submitted in accordance with subparagraphs D.8 and D.10. This subparagraph of the consent decree was written following the emphasis in the ROD that the only alternative for disposing of the water that was,pumped and treated from Subunit A was to reinject the treated water back into the ground. This section, as written, would require reinjection of the water and would not permit consideration of other available beneficial uses of this groundwater.

Currently the natural background quality of Subunit A water addressed by the 00 is located is so poor that it is not used for potable, agricultural, or industrial purposes. Tests show average total dissolved solid concentrations of approximately 3,000 ppm. The existence of industrial contaminants slightly further degrades the extracted water's quality. As presently written, the consent decree provides no option for cost-effectively treating the water to manage the background contaminants that remain after the water is treated to remove the industrial contaminants of concern. If these background contaminants can be economically managed, it could potentially create the availability of additional water resources for use in the Arizona desert.
The 1987 ROD incorporated the OUFS discussion of potential water disposal options. The OUFS indicated that disposal options other than reinjection are not economically feasible, although other uses may be desirable. Presently, the proposed treatment water reinjection system calls for installation of 15 to 18 reinjection wells along with a distribution piping system. The capital cost of this system will range from $500,000 to $1 million. There also will be a large operation and maintenance cost associated with the reinjection system. The operation of the reinjection system may pose substantial technical challenges which are of concern to Goodyear. Goodyear believes, therefore, that from a cost-effectiveness and technical viewpoint, an option to create an economically usable water resource from the extracted water may exist, or may arise at some point in the future. Goodyear further believes the potential for managing a presently non-usable water resource to create a viable water resource should be encouraged and further explored, if determined by Goodyear to be economically practicable. Of course, if such

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 17 of 28

Assistant Attorney General October 19, 1988 Page 4

an alternative proved to be feasible, it would have to be consistent with the water laws of the state of ^Arizona, and should be approved by the Arizona Department of Water Resources. Therefore, Goodyear respectfully requests DOJ and EPA to consider and support a modification to the consent decree to allow Section VII.C.5(a) to read as follows:

All water from the groundwater extraction system will be treated. All treated water from Subunit "A" will be reinjected, or in the alternative, disposed of through.^ an economically practicable and beneficial use on terms and conditions approved by the Arizona Department of Water Resources. Treatment shall assure that reinjected water will meet federal and state standards for treatment plant discharge levels prescribed in Table 1 of the 1987 ROD. All water disposed of through a beneficial use shall be treated or otherwise meet all applicable federal and state water quality standards _and criteria.
B. Air Emissions from Air Strippers:

· j§

^
~
ft ft
·

ft

Goodyear ' s second issue relates to Section VII. C. 5 (c) of the proposed consent decree, which reads as follows: Air stripping will be used to reduce volatile organic compound { "VOC" ) contamination to meet federal and state standards as prescribed in Table 1 of the 1987 ROD. The air stripping towers will be equipped with -air emission controls in order, among otherpurposes, to meet Maricopa County requirements, including Rule 32-C and any other applicable provisions of the Arizona implementation plan under the Clean Air Act. If the Maricopa County requirements are revised and approved by EPA pursuant to the Clean Air Act to specify that sources such as the -air stripping towers are noc subject co air emission controls, then Goodyear may petition EPA to agree to amend this consent decree to remove the air emission control requirement of this paragraph. Any dispute with regard

^

ft W
· ft
K ft ft ft

ft

I I
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 18 of 28

I I I I I I I I I I I I I I I I I I I

Assistant- Attorney General October 19, 1988 Page 5

to any such petition shall be subject to dispute resolution in accordance with paragraph XXII.
Goodyear entered into negotiations with EPA and DOJ concerning the proposed consent decree guided by two principal documents: the provisions of CERCLA/SARA, including related guidance documents and the National Contingency Plan; and the requirements of the ROD.

1.

ROD Requirements;

The ROD specifies that "the air stripping towers (related to the air stripping required to reduce the VOC contamination of the groundwater) will be equipped with air emission controls in order to meet Maricopa County requirements that all new air emissions sources employ reasonably achievable control technology to reduce emissions, as promulgated by the Superfund Amendment and Reauthorization Act (SARA). Remedies should significantly and permanently reduce the volume, toxicity and mobility of the contaminants."

During recent discussions with EPA, concerns have surfaced as to precisely what the requirements of the ROD are and whether or not the above-quoted provision in the proposed consent decree adequately reflects those requirements. Specifically, Goodyear was operating under the impression that Maricopa County Air Pollution Control ("MCAPC"), Rules and Regulations, Regulation III, Rule 32(C) requires application of emission control technology for new VOC emission sources under certain conditions. Through its consultant, Dr. Lial F. Tischler of Engineering-Science, Inc., Goodyear advised EPA by letter dated October 2, 1987, (after the date of the ROD and the notice letter) that it had conferred with Mr. Lawrence Crisafulli of the Maricopa County Air Pollution Control to determine how those regulations apply to VOC stripping columns used for groundwater treatment. That conference revealed that the primary condition that the MCAPC applies to determine if emission controls are required for a new source is a minimum emission rate of 40 pounds of VOC emissions per day. At an estimated maximum rate of 10 pounds of VOC emissions per day, the air stripping requirements for the Licchfield site are well below the de minimis level and should not require emission controls. A copy of this letter is included as Attachment 1. Goodyear also has requested an opinion from Arizona counsel, Fennemore Craig, concerning the requirements for air stripping under the Arizona law and related implementation plan.

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 19 of 28

Assistant Attorney General
October 19, 1988

Page 6

We are attaching to this letter a copy of that memorandum dated October 10, 1988, (Attachment 2) that basically concludes that

Rule 32{C), which has now been renumbered as Rule 320 as a result of amendments to the Maricopa County Air Pollution Control Rules
and Regulations on July 15, 1988, does not require carbon

adsorption emission controls for the air stripping towers. When Goodyear was involved in negotiating the proposed
consent decree with EPA and DOJ, it was operating under the

understanding that the limit in Maricopa County for the discharge of VOC contaminants from air stripping towers without carbon absorption is 40 pounds per day. It also was aware that there were considerable discussions between EPA and the state of Arizona concerning the amendment of the state's implementation plan, and that these requirements could be adjusted upward or downward. Although Goodyear had not yet retained a consultant to commence the conceptual design for this project or receive the benefit of the consultant's advice in this area, Goodyear nevertheless supported the language in Paragraph VII of the consent decree in order to recognize the opportunity to adjust the design, construction and operation of the OU if changes in the Arizona/Mar icopa County requirements were to occur. As indicated by the Fennemore, Craig memorandum, changes did occur on July 15, 1988, but these changes did not alter the general statement within the regulations that up to 40 pounds per day of VOC emissions are acceptable without carbon adsorption emission controls .
2. CERCLA/SARA Requirements:

M (
» · *

IB »

1
M ·

Remedial action selected under SARA must attain a degree of cleanup of hazardous substances, pollutants or contaminants which include applicable, relevant and appropriate federal and state requirements {often referred to as "ARARS"). "Compliance [with these requirements] is required at the completion of the remedial action for hazardous substances, pollutants or contaminants that remain on site."1 Thus, EPA's guidance emphasizes that these requirements must be achieved in the context of the final remedy, but not necessarily in the context of the interim remedy such as the OU envisioned for this site.
Further emphasis of this 121(d)(4) which specifies that EPA:
1

· TM · if m

point

exists

in

Section TM

52 Fed. Reg., page 32495, "Superfund Program; Interim Guidance on Compliance with Applicable or Relevant and.. Appropriate Requirements; Notice of Guidance, August 27, 1987."

·

I
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 20 of 28

I I I I I I I I

Assistant Attorney General October 19, 1988 Page 7

. . . may select a remedial action meeting the requirements of paragraph (1) [i.e., ARARS] that does not attain a level or standard of control at least equivalent to a legally applicable or relevant and appropriate standard, requirement, criteria, or limitation as required by paragraph (2) if [EPA] finds that -

(A) The remedial action selected is only part of a total remedial action that will attain such level or standard of control when completed; . . .
While Goodyear has been unable to obtain any "official" interpretation by EPA of the criteria that it will consider with respecc to this waiver, we have examined a recent draft guidance document entitled "CERCLA Compliance With Other Laws Manual" dated August 8, 1988 (OSWER Directive 9234.1-01). Paragraph 1.3 of that document, entitled CERCLA Waiver Criteria for ARARS,

I I I I I I I I I I

I

states as follows:
This waiver may be applicable to interim measures that are expected to be followed within a reasonable time by complete measures that will attain ARARS. The interim measures waiver may apply to sites at which a final site remedy is divided into several smaller actions.

* * * The factors that may be appropriate invoking this waiver include: * * *
· Non-interference with final remedy. The interim measures selected must not interfere with, preclude, or delay the final remedy, consistent with EPA's priorities for taking further action. (Emphasis added).

for

Goodyear believes that this guidance is "relevant" to the situation ac hand. Specifically, in a reasonable period of time, Goodyear anticipates thac EPA will complete the feasibility study for the final remedy at this site and issue a ROD that may,

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 21 of 28

Assistant Attorney General
October 19, 1988

Page 8

inter alia, address the need for additional groundwater treatment and perhaps other related matters. Goodyear also believes that it should not embark on costly requirements to design an air stripper with carbon adsorption features if those features become redundant, or are inconsistent with the overall requirements of the final remedy. Thus, Goodyear maintains that the consent decree should be adjusted to clarify that air stripping towers do not have to be equipped with carbon adsorption devices for this interim remedy measure. As indicated in the Pennemore Craig memorandum, we have received concurrence with this approach from the Maricopa County officials (Attachment 3). Section 121(b) of SARA requires EPA, in assessing alternative remedial actions, to take into account a number of criteria which include:

(1) the persistence, toxicity, and mobility of the hazardous substances; "
(2)
(3)

fl|

long term maintenance costs;
the cost effectiveness of the remedy; and £

(4) the ability of the remedy to adequately protect human health and the environment.

^ M

It could be argued that regardless of state requirements, carbon adsorption should be a requirement of this OU because it would serve to reduce the mobility of hazardous substances into the environment. However, SARA offers other criteria that also must be considered, including long term maintenance costs and cost effectiveness. In ~this regard, Goodyear estimates that the capital cost for the installation of carbon adsorption units could range from $200,000 to $900,000 in capital costs (depending on whether an onsite carbon regeneration system is included). Thereafter, annual operation and maintenance costs could range from $100,000 to $200,000. Goodyear

flj ·

tt jj « · *

believes that such an approach, based on the interpretation of · the proposed consent decree, would not meet the SARA "cost 9 effective criteria" as the "effectiveness" criteria practiced by EPA in the administration of the Superfund program is based on a · risk range of 10-4 to 10-7. Simply stated, we now know through the conceptual design that the TCE air emissions will not exceed this range regardless of whether air stripping with or without m carbon adsorption is used. Thus, carbon adsorption simply is not £ cost effective.
As the August 8, 1988 "CERCLA Compliance With Other

I

I
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 22 of 28

Assistant Attorney General October 19, 1988 Page 9

under the most conservative of circumstances, the health risk of exposure to the air emissions from the air strippers, without carbon adsorption, is no worse than 10-4. Goodyear is developing additional data to support this position, which we will forward under separate cover. This will further verify our position that there will not be an unacceptable risk --to public health if these contaminants are released into the environment in these de minimis quantities.

Laws Manual" indicates, ARARS fall into three basic categories: (1) ambient or chemical specific requirements; (2) performance, design or other action specific requirements; and (3) location specific requirements. (Goodyear does not believe that this third category has any relevance to these discussions.) While this document and the August 27, 1987 EPA interim guidance indicate that a national ambient air quality standard is a type of chemical - specific ARAR, there are no such standards for TCE. Using the risk range of 10-4 to 10-7 as presented in the "Guidance on Feasibility Studies Under CERCLA" dated April, 1985, p. V-19 and in the "Superfund Public Health Evaluation Manual" dated October, 1986 (OSWER Directive 9285.4-1, pp. 91-93, section 8.32) as a guide, Goodyear has preliminarily concluded that, 2

The second category of ARARS - "performance, design or other action specific requirements" - could arguably include carbon adsorption, if this requirement is viewed in a vacuum. However, as we have discussed above, this requirement must be viewed in light of other SARA requirements of cost effectiveness and adequate protection of public health. We believe that both of these requirements are met without carbon adsorption. In addition, assuming arguendo that these standards still apply, the waiver of such standards for an interim remedy also seems appropriate. Accordingly, Goodyear proposes that subparagraph 5(c) of Paragraph VII of the consent decree be modified as follows: Air stripping will be used to reduce volatile organic compound ("VOC") contamination to meet federal and state standards. At the present time, relevant state implementation plan and Maricopa County requirements do not require sources of VOC emissions that are well below 40 pounds per day, such as the air stripping towers, to utilize carbon adsorption emission
2 The assumptions made include an emissions rate of 10 pounds per day for 70 years, continuous worst case meteorological conditions, and continuous exposure on a 24 hour basis.

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 23 of 28

Assistant Attorney General October 19, 1988 Page 10

\ In summary, Goodyear has maintained its commitment to address the problems at the Litchfield site in a manner that will adequately protect public health and the environment. As it pursues this commitment, Goodyear anticipates that new facts will always emerge. The comments that we submit on this proposed consent decree today relate to refinements to the approach to be taken on the OU based on a better appreciation of the facts that we now have, and the applicability of CERCLA/SARA^ and related guidance documents to those facts.
Sincerely yours,

controls. If the applicable requirements are revised to specify that sources such as the air stripping towers are subject to - air emission control, then the provisions on modification in Paragraph XXV apply. Any dispute with regard to such emission controls shall be subject to dispute resolution in accordance with Paragraph XXII.

Takashi Ito Attorney
TI:afj ~ Attachments cc: Daniel W. McGovern, Regional Administrator, EPA Jeff Rosenbloom, EPA
Hugh Barroll, EPA Barry Sandals, DOJ

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 24 of 28

2901 NORTH I N T E R R E G I O N A L

AUSTIN. TEXAS 7sr:z

E N G I N E E R I N G - S C I E N C E , INC.

( siz>

477-9901

October!, 1987

Mr. Jeff Rosenbloom (T-4-2) Remedial Project Manager U.S. Environmental Protection Agency Region EX 215 Fremont Street San Francisco, California 94105

Re: Air Emission Controls on PGA Operable Unit

Dear Mr. Rosenbloom:
We received your letter of August 18, 1987 responding to our submission of the air quality modeling of the emissions from the volatile organic compound (VOC) strip'ping columns for the operable unit (OU) at the Phoenix-Goodyear Airport (PGA). As you acknowledged in your letter, the air quality modeling demonstrates that the uncontroDed emissions of trichJoroethylene (TCE) from the stripper result in

ambient concentrations well below the 0.769 p.g/m3 annual average which Table 9-6

modeling we performed for our comments on the above document demonstrates that maximum 24-hour concentrations of TCE are always less than 1 percent of the short-term exposure limit adopted by the American Conference of Governmental Hygienists. It is clear that the uncontrolled emissions of TCE from the stripping columns will not jeopardize human health or the environment.

of the Public Comment Feasibility Study for Section 16 Operable Unit, June 1987, cites as the lower cutoff limit for requiring air emission controls. In addition,

In your letter you cite two bases for a policy decision that air emissions controls will be required'on the VOC stripping columns: (1) the wording in Section 121(b)(l) of theSuperfund Amendments and Reauthorizatipn Act (SARA); and (2) Maricopa County Air Pollution Control Rules and Regulations, Regulation 3, Rule 32(c). We do not believe that the Agency is interpreting either of these rules correctly for this situation.

Section 121(b)(l) of SARA, as you state, indicates that preference should be
given to remedial actions which will result in permanent and significant decreases in wxiciry, mobility, or volume of hazardous substances. As you correctly point out, the mobility of the VOCs in the ground water is increased by air stripping.

However, the statement that the volume of the contaminants is increased is incorrect - although the volume of the media in which the hazardous substance is distributed is increased substantially, the mass of VOCs emitted is constant and in fact is

c r T-s

Case 2:03-cv-02226-ROS

Document 81-13

Filed 06/27/2006

Page 25 of 28

Mr. Jeff Rosenbloom Page 2 October 2, 1987
substantially diluted in concentration. This dilution and increased mobility, in turn, decrease the potential exposure rate for any individual as compared to the ground iter exposure potential, as demonstrated by the air quality modeling. It can be Argued that although the TCE is unaffected by the transport from the water to the air in the stripper, the overall toxicity is reduced because of the substantial dilution which occurs in going from the water to the air.
As you have also acknowledged in the August 18th letter, Section 12l(b)(l) of SARA also has explicit limitations on deciding whether a particular remedy is needed. With respect to long-term and short-term potential health effects (Section 121(b)(l)(D), we have already shown, and you have acknowledged, that the uncontrolled TCE emissions (by far the most significant air pollutant) from the stripping column pose no long-term and short-term health effects. .Section 12l(b)(l)(E) requires an assessment of long-term maintenance costs. The costs associated with removing the estimated 7 pounds per day of TCE from the VOC stripping column off-gas cannot meet any reasonable cost effectiveness test, as shown below. As we documented in our comments on the OU public comment document, about 7 pounds per day of TCE will be emitted from the stripping columns. Other VOCs will amount to a total of not more than 3 pounds per day. For the purposes of this analysis, we assume that a total of 10 pounds per day of total VOCs will be emitted by the columns. Using a carbon loading of 0.1 pound of VOC per pound of carbon, 36,000 pounds per year of carbon will be required. At S2 per pound for replacement (includes custom regeneration), the annual operating cost is 572,000 for the air emissions control unit excluding the maintenance costs and capital costs for this equipment. This works out as a cost of S43,000 per metric ton of VOC removed for the carbon replacement alone. In its proposed rule for regulation of benzene under Section 112 of the Clean Air Act, the EPA's Air Office used a cost effectiveness value of S1050 per metric ton of VOC reduction to establish a size cutoff for facilities covered by the emission standards (Benzene Fugitive Emissions Background Information for Promulgated Standards, EPA 450/3-80-032b, June 1982). EPA determined that controls at this level are not cost-effective for benzene, a carcinogenic air pollutant. In addition, EPA selected a cost-effectiveness limit for VOCs of SI600 per metric ton in setting guidance for 36 major organic chemicals
(Guideline Series (Draft) Control of Volatile Organic Compound Emissions from Air Oxidation Processes in Synthetic Organic Chemical Manufacturing Industry, CTG,

m K M · · J§

^ ·

·

M J§

^
· *

ft m
m |

» · ·

Office of Air Quality Planning and Standards, March 1984).

The cost of the air emissions controls for the PGA VOC stripping columns is more than an order of magnitude greater than the cost-effectiveness limits used by the Agency to evaluate air pollution controls for hazardous air pollutants and VOCs. We do not believe that SARA intends that control technologies which are so inefficient are to be mandated - that is, in our opinion, the reason that Sections 12l(b)(l)(D) and (E) are included in SARA. We do not believe that this basis for EPA's requirement for emissions controls on the strippers is justified.

· m
m jj

I

I
Case 2:03-cv-02226-ROS Document 81-13 Filed 06/27/2006 Page 26 of 28

I

Mr. Jeff Rosenbloom Page 3 October 2, 1987 Maricopa County Air Pollution Control Rules and Regulations, Regulation 3, Rule 32(c) requires application for reasonably achievable control technology (RACT) for new VOC emissions sources under certain conditions. We spoke with
Mr. Larry Crissafulli of Maricopa County Air Pollution Control (MCAPC) to

determine how they apply this regulation to VOC stripping columns used for ground water treatment.- The primary condition that MCAPC applies to determine if RACT is required for a new source is a minimum emissions rate of 40 pounds per day. At an estimated rate of 10 pounds of VOC emissions per day, the PGA air stripping columns are well below the de minimis level and should not require emissions controls. Mr. Crissafulli indicated to us that there are 5 to 6 stripping columns currently operating in Maricopa County and none of these have air emissions controls. One of these stripping columns is a large (32 foot high, 13 foot diameter) unit which strips TCE from a drinking water supply well in Scottsdale. The other columns all strip gasoline-contaminated groundwater which would contain benzene, toluene, and xylenes as well as other V.OCs.

It is obvious that MCAPC does not interpret their regulation to require air emissions controls on de minimis VOC sources. This means that EPA's second basis for insisting on air emissions controls is also unjustified. We respectfully request that the Agency reconsider the policy decision to require air emissions controls on the air stripping columns. Aside from the costs, we believe that the additional maintenance and