1. A geophysical survey.
2. Clearing and grubbing of the 2.1 acre site by a qualified disadvantaged business.
3. Removing the rail and ties from an abandoned railroad bed. The cleared railroad bed serves as the access road to the site.
4. Further delineation, by sampling, of the "hot" zones both north and south of the active railroad tracks that bisect the site.
5. Obtaining access agreements on various parcels of property that affect the site itself and other necessary activities on adjacent properties.
6. Surveying and staking property lines.
7. Establishing relationships with City and State officials, and property owners.
8. Establishing a design team of qualified engineers in many disciplines. vCompletion of the final design and cost analysis for the 1,000 g.p.m. water treatment plant. The treatment plant is designed with flexibility and can be scaled to address water treatment as a component of the Lemon Lane remedial action with minimal added costs and with no break in the operation of the plant.
9. A detailed estimate of the operational costs associated with the plant for a one-year "shake-down" period.

After the issuance of the June 22, 1999, Action Memorandum, the full scale construction of the ICS water treatment system began. To date, the following additional tasks have been completed, or are underway.

1. An all weather access road into the plant site and into the site was built.
2. A construction office and a "lay down" yard are completed.
3. Both a temporary fence around the "lay down" yard and a permanent fence around the treatment system area are erected.
4. The site has been leveled (a cut and fill).
5. Permanent power runs into the site.
6. Two "jack and bores" under the adjacent Indiana Central Railroad tracks to the site of the spring have been completed to assist in storm water diversion.
7. Blasting & excavation of the rock where the spring receiving sump will be located has been completed.
8. Construction of the spring receiving sump building is complete.
9. Construction of the treatment plant is 90% complete.
10. The yard and utility piping is complete.
11. All concrete work in and around the plant is complete. This includes concrete work for floors and foundations, head walls, elevated equipment pads etc.
12. Work has started on the two, two acre feet storage tanks (30% complete).
13. Installation of the process equipment (50% complete).
14. Installation of site's power requirements (70% complete).

Work has not yet started on:

1. planting of "tree barrier" adjacent to the Pro-Mark property

B. Background Regarding Design Changes

At the time of the June 22, 1999, Action Memorandum, it was determined that as a result of a detailed cost analysis, funds obligated in the original Action Memorandum of September 30, 1998, were not adequate to proceed very far into the construction of the plant, much less to complete it or to adequately enter a "shake-down" period. The June 22, 1999, Action Memorandum identified two main reasons for the differences in scope and cost of the removal that supported the consistency exemption and ceiling increase Action Memorandum. First, during the removal action design process, U.S. EPA identified areas where technical issues that were addressed in the original work of the remedial action contractor needed to be revisited and more thoroughly addressed than before. Second, during the removal action design process, new technical issues were identified. These issues were in areas that had not been previously thought of, but that needed to be addressed to ensure a protective clean-up and to ensure that the removal action was conducted in a cost-effective manner consistent with the remedial action at the Lemon Lane Landfill site. ²

The differences between the Tetra Tech (Remedial Contractor) estimate and the Earth Tech (Removal Contractor) costs as determined after design were explained in the June 22, 1999, Action Memorandum, and are reiterated in more detail below.

The original rough cost estimate described in the action memorandum was developed by the Lemon Lane Landfill remedial contractor in generally analyzing was a broad spectrum of the various treatment technologies available for the ICS site, and the various costs of these technologies. This original assignment was not intended to be used as a design estimate. Indeed, the time needed to do a design estimate was not given to the remedial contractor. Rather, the task of developing the design estimate, along with the design and construction, was given to the removal contractor.

In considering the differences in both the scope of the project and the costs between the remedial contractor (Tetra Tech) and the removal contractor (Earth Tech), it must be noted that the primary purpose of the remedial contractor (who assisted U.S. EPA in developing the conceptual outline of this removal action) was not to provide an engineer's estimate of the cost of the project, but was simply to compare the cost of the various removal action options being considered and to assist in identifying a cost effective alternative. Once the project was assigned to Earth Tech, the engineering and design was initiated. This work took a full four months at a cost of $413,000 to produce an engineer's estimate and, to date, is the only true estimate of the project's costs. During the course of the design process issues came into focus that had not been previously considered during the conceptual phase, but that none-the-less needed to be addressed. Other issues that were generally considered during the conceptual phase came into sharp focus requiring, in some instances, a re- evaluation of (and in some instances changes to) the conceptual approach.

Thus, at the conceptual stage the following items necessary items were either not included or not adequately considered:

1. No provision for emergency power, critical at "high flow", if power were to be lost.
2. No separation of clean surface runoff from the contaminated spring water. Maintaining separation of the clean surface runoff from contaminated spring water reduces the total amount of water to be treated. The separation is accomplished through an additional "jack and bore" of a conduit under the active railroad tracks. This matter was not considered originally.
3. Re-use of the original spring water conduit under the railroad tracks despite not knowing the structural integrity of the 24" pipe. The age of the existing conduit is estimated to be many decades.
4. The original 2-acre foot containment was contemplated to be in the form of a lagoon with a synthetic liner (as opposed to the storage tanks that have been designed). The lagoon approach did not take into account the Karst topography. The lagoon concept also did not address the removal of accumulated contaminated sediment, or overflow or by-pass piping.
5. No provisions for sanitary sewer, or potable water to the treatment building.
6. Grossly inadequate access road into the site.
7. No provisions for concerns of adjacent property owners i.e. Pro-Mark, with regards to landscaping, sight barriers, etc.
8. No cost provisions for a "shake down" period prior to the plant being turned over to the State.

The above items, together with the flexibility designed into the plant (including added electrical capacity, increased pipe diameters, an additional 2-acre feet of storage, and 40% added floor space in the treatment building) ensure compatibility with the final Lemon Lane Landfill remedial action. The above list includes the predominant reasons for the difference in costs between the original cost estimate (identified in the Action Memorandum) and the removal engineer's design estimate.

As explained in the June 22, 1999, Action Memorandum, the facility currently under construction is not a final remedy. Rather, it is a "flexible interim remedy" intended to be consistent with the remedial action at Lemon Lane Landfill.³ The Plant is designed to make the transition to the future final remedy as easy as possible and with minimal interruption to the treatment process. From the larger building to the excess electrical capacity and piping size, everything is geared toward not having to add building space or dig up and replace pipelines or conduits (thereby effectively terminating the plant's operation) during any final remedy retrofit.⁴ U. S . EPA anticipates that the final remedy may call for something additional, such as pumping capacity or different/extra process equipment. By scaling the plant today, however, for possible expansion if necessary, does not result in large short-term expense (the difference between an 8 inch pipeline and a 10 inch pipeline, and added electrical capacity is not a substantial funding increase) and the benefits of scaling the Plant expansion may result in overall cost savings when compared to failure to include expandability in the original design and then attempting a retrofit.^5>

Furthermore, the storage tanks are bolted steel tanks that also can be moved, as can the actual building itself.

C. 25-YEAR STORM EVENT AND FLEXIBLE DESIGN

The flexible design as approved by U.S. EPA has combined the initial guidelines of 1,000 g.p.m. of water treatment using basic water treatment technology and a storage of 2 acre feet, with the addition of more storage capacity. Outside of adding the 2 acre feet of storage, the U.S. EPA has adhered to the initial guidelines. U.S. EPA has determined that by designing the ICS plant to address a 25 year storm event by simply adding 2 acre- feet of storage capacity and additional assorted pumps and miscellaneous equipment, is protective of Clear Creek.⁶ The 1,000 g.p.m. plant, along with ability to send an additional 5,000 g.p.m. to storage (up to 4 acre-feet to be treated off peak) will be able to remove approximately 70% of the total sediment of that storm event. Although a 25 year storm event only holds a 4% chance of occurring at any given time, it comes down to whether or not such a release is harmful. U.S. EPA has determined that such an event would be harmful, particularly when it can be 70% captured with a relatively small additional financial input.⁷

U.S. EPA's June 22, 1999, Action Memorandum amended the removal action to allow for design and construction of a system capable of addressing storm water flows from a 25-year storm event (defined as a 3.7" rain in a 6 hour period by the National Engineering Handbook). This differed from the conceptual approach of a plant capable of treating contaminated water at a rate of 1,000 gpm and with the ability to store up to two acre- feet of water pending treatment.

The amount of PCBs on a mass basis found in the discharged ICS storm water increases during large storm events. The PCBs are discharged as part of the suspended solids, or sediment, coming out of the spring during high-flow events. U.S. EPA's decision to design the plant to handle the total gallons per minute of a 25-year storm event, together with a removal efficiency of 1 micron of contaminated sediment, was driven, in part, by the fact that for this relatively small investment of approximately $350,000 for additional storage capacity, an added degree of protection for Clear Creek can be achieved and a removal efficiency of 0.1 to 0.3 ppb (parts per billion)of PCB's, can be met.

Because of time constraints, driven by a Court Order that requires the start of the clean-up of the Lemon Lane Superfund Site (which is hydraulically connected to the spring and the source of the PCBs emerging from the spring) by the spring of the year 2000, there was insufficient time to conduct a "particle size study" before designing and constructing the Plant. Absent specific the information regarding particle size, there was insufficient information on which to base a design for a plant capable of adequately addressing site risks using as a general design parameter the ability to treat 1,000 g.p.m. of storm water and capable of 2 acre-feet of storage. Accordingly, U.S. EPA, based on a consensus of opinions by Earth Tech's design team, agreed with the recommendation, that it was appropriate to design the treatment plant to address a 25-year storm event (see attachment 4). It was understood that the added costs associated with the addition of an extra two acre-feet of storage, would be in the range of $350,000 (including adding extra pumps).

To enable the Project to be more cost effective and in lieu of a particle size study, U.S. EPA also instructed Earth Tech to design the plant to be capable of removing storm water total sediment down to a particle size of 1 micron. Below 1 micron, any suspended sediment is considered colloidal and to remove a colloid from the spring water would involve flocculation and coagulation and this technology would be considered "ultra filtration." It was roughly estimated that to add a flocculation and coagulation system to the treatment train would add an additional $1,000,000 to the project costs. Addressing colloids and the use of ultra-filtration is beyond the scope of this project, but will be considered in the final remedy selection process.

U.S. EPA's decision to design the plant to address the risks from storm water flows during a 25-year storm event (as opposed to the conceptual 1,000 g.p.m with 2 acre feet of storage approach) results in the plant's ability to increase its PCB percentage recovery from an estimated 40% of the PCB mass released during such an event to 70% of the PCB mass released during a 25-year storm event. This level of "protectiveness" to Clear Creek was deemed appropriate and necessary by the U.S. EPA.

U. S. EPA's system has been designed to be protective and will not permit any water to by-pass the plant for maintenance and/or equipment breakdowns. In addition, EPA's system will only allow for by-pass when the total pumping capacity is overwhelmed when influent volumes exceed 6,000 g.p.m.⁸ U.S. EPA believes that its approach is most protective of Clear Creek and is achievable with the 25 year design.

D. MAJOR DESIGN CHANGES; CONCEPTUAL Us ACTUAL DESIGN

With regards to the storage of storm water pending treatment, U.S. EPA has determined that the best approach consists of constructing two above-ground storage tanks, each capable of storing up to two acre feet of storm water flows from Illinois Central Spring. This is a shift from the "lagoon" approach contemplated at the conceptual stage. At the conceptual stage a lagoon capable of two acre-feet of storage water storage was to be located in line with the bed of Clear Creek. This "basin" would be used for both storage of storm water pending treatment, as well the settling of the suspended solids. During the design phase of the plant, U.S. EPA determined that the use of a lagoon in the type situation presented at Illinois Central Spring would not perform either the storage or the settling features very well, and that significant risks would be posed if a lagoon were relied on based on the topography of the area.⁹

The main reason U.S. EPA abandoned the lagoon concept is the instability of the Karst topography present throughout boundaries of the site.^l0 Because of the Karst topography, and the various cracks, vugs and "swallow holes," at the Site, U.S. EPA determined that significant risks of release of contaminated materials was presented by reliance on a synthetic liner system placed on this type of subsurface. Ensuring liner integrity was determined to be too problematic to warrant proceeding.

Among the other problems identified with the use of a lagoon system are the following: A limited ability to remove from the lagoon sediment and foreign material such as leaves, twigs and other debris without risking the integrity of the liner; extended periods of time where there lagoon could not be used (unless the lagoon is compartmentalized) during maintenance work or during sediment removal, thus placing the stream at risk; concern over attractive nuisance issues; concern over wildlife exposure to contaminated sediment accumulated in the lagoon; and, finally, concern over the ability to detect for rips or tears in the liner of the lagoon, an item that a double walled bottom of a storage tank would eliminate.

Abandonment of the lagoon concept (and instead using a storage tank system) also resolves problems associated with housing, mounting, and protecting the pumps and assorted electrical components needed to transport the contaminated water to the treatment plant. By constructing a spring water receiving sump and an accompanying small building, this important function now falls into place.

During design, U.S. EPA also determined that there was a need to handle sediments collected during the capture and treatment process. Earth Tech completed a flow diagram showing how the sediment will be managed by the plant once captured. By use of a thickener and a filter press, the PCB laden sediment can be dewatered and shipped off site for disposal. The fact that U.S. EPA is using a filter press, and providing a means for dealing with sediments that will result from treatment does not make this system any less of an interim system. Rather, it recognizes the potential for large amount of sediment to be released from Lemon Lane during the Lemon Lane remediation, scheduled to begin in the spring of 2000, and provides a cost-effective method for addressing that potential.

The design also provides for an emergency power source to ensure continued operation in case of a commercial power outage. This is also needed to ensure that this removal is protective. The most turbid and contaminated water flow occurs during high flow, and that high flow probably will happen during a storm event. Commercial power to the plant is most likely to be interrupted due to storms. Accordingly, back-up power is appropriate.

U.S. EPA's conceptual approach estimated $24,000 for the access road and $13,000 for clearing and grubbing of the site. The actual clearing and grubbing cost for an "8-A" firm was $62,000. The all weather access road's cost was $71,000. At the conceptual stage U.S. EPA had erroneously failed to provide power and utilities, such as natural gas and telephone, and had no provision for emergency power ($150,000 for a diesel generator). These expense are for the most part not new, but the were estimated incorrectly before the design was prepared. In addition, while these elements are durable, they do not transform this interim system into a final system. Rather, these are components that are compatible with both.

Another significant design change involved the manner in which the contaminated spring water is collected and then transported to the treatment plant. The conceptual design made no provision for separating the clean surface runoff from the contaminated spring water. It also assumed that the decades old conduit under the adjacent railroad would be adequate to continue to transport the spring water. It was an easy decision not to treat clean surface water, which is achieved through construction of an earthen berm and a separate clean water by-pass. It also followed logic to construct a new spring water pipe and use the original pipe as a reserve. Three 2-inch PVC were also run in the annular space between the casing and the pipe for permanent electricity, should it be needed at the spring.

One last major design difference was that the conceptual report included no cost provisions for a "shake down" period before the treatment plant was turned over to the State. Of course the person who prepared the conceptual report had no way of anticipating that such a agreement would be in place.

IV. EXPECTED CHANGE IN THE SITUATION SHOULD THE ACTION BE DELAYED OR NOT TAKEN

There most certainly would be an increased risk to public health and to the environment if the Illinois Central Spring water treatment plant is delayed or not built. The high levels of PCB that are currently affecting the Clear Creek ecosystem will continue to affect the ecosystem and those attempting to use the resources of the creek downstream.

V. RECOMMENDATION

This memorandum represents a further explanation of the selected removal action for the Illinois Central Spring Site located in Bloomington, Indiana, developed in accordance with CERCLA, as amended, and is not inconsistent with the NCP. This decision is based upon the Administrative record for the site. Conditions at the site continue to meet the NCP Section 300.415 (b) (2) criteria for a removal action and I recommend your approval of the explanations presented in this document.

APPROVE: Wm. E. Munro
Director, Superfund Division

FOOTNOTES

¹ Using the 1996 calendar year, approximately 20% of the total PCB mass released during the 1996 calendar year is attributed to non-peak flow discharges and 80% of the total PCB mass released is during peak or storm flow events. The total mass of PCBs released during calendar year 1996 is estimated to be 18.7 kilograms or approximately 41.14 pounds. One pound of PCBs has the potential to contaminate 454,000 kilograms (or approximately 998,800 pounds) of stream sediment at a concentration of greater than 1 ppm PCBs.

² The plant as designed does not differ conceptually from the plant as originally conceived in any substantive manner. The plant, as designed, is capable of treating 1,000 gpm of storm water flow, as did the conceptual plant. Both plants provided for storm water storage. At the conceptual stage a 2-acre foot lagoon storage system was envisioned. As discussed more fully later in this memorandum, the lagoon approach was abandoned during the design process in favor of a 4-acre foot above-ground storage tank system. The difference in cost between the two approaches is approximately $350,000.

³ U. S. EPA's system is not a "mobile portable" system that can be readily moved en masse to a new location. However, all of the equipment in the process was designed with portability in mind, as witnessed by the many pipe connections which were selected for easy disassembly and reassembly if so desired. Thus, the equipment used in the treatment of storm water (such as the carbon vessels) can be removed and re-used somewhere else.

⁴ At the conceptual stage, building costs were estimated at $45,000. Using a low hypothetical cost of $50 per sq. ft., the space afforded at the conceptual stage would be in a 30' x 30' building. This size building would be far too small even for the treatment process that was being considered at the conceptual stage, and did not include costs for things like foundation costs, wiring, piping, etc. U.S. EPA's building, as designed will be able to house the added thickener and filter press along with space to house on office, and a small laboratory with lavatory.

The building also contains 40% added space above and beyond what will be used by the Earth Tech process. This added floor space is projected to be used in the "final remedy" phase perhaps to house the "ultra filtration" equipment and is part of the "flexible remedy" that U.S. EPA has designed for in its project. The cost of the building "as built," including costs for foundation, piping, wiring, etc., is approximately $2,048,500. The cost of the additional 40% space is approximately $281,700, or approximately 14% incremental additional cost.

⁵ A final design will be approved as part of the final treatment system. That design may include various water treatment technologies that can be best described as "ultra filtration". This technology, will involve either (or both) reverse osmosis or flocculation/coagulation, in addition to the processes that will be in place. For any new technology to be implemented, certain piping, power, and building requirements will have to be met. It makes sense to design and build to those requirements now where significant costs savings will be realized and the cost of retro-fit avoided. The capital cost of a 10" pipeline does not differ significantly from that of a 12" pipeline. This also holds true for added electrical capacity and additional square feet of building space. These three items demonstrate that the Illinois Central Spring water treatment plant will be able to transition from a "flexible to a final" plant with little or no impact on its operation and with little additional costs, other than potential additional treatment equipment costs.

⁶ In the 1998 action memorandum, U.S. EPA determined that a two-acre feet storage system would be sufficient for storm water flows. U.S. EPA in part founded its decision in this regard based on data regarding one isolated year of storm data. Earth Tech reviewed charts and data from CBS and determined that it is more appropriate to design a system capable of addressing storm water flows during a 25-year storm event. A 25 year storm event is a common event to design to and has been used by other government agencies.

U.S. EPA's flexible interim design plant has a 1000 g.p.m. treatment capacity. However, U.S. EPA's plant also will be able to concurrently send an additional 5,000 g.p.m. to storage (up to four acre feet) for treatment after the storm surge diminishes (25-year storm event). A portable plant would have to allow untreated storm water flow by-passes to Clear Creek and, thus, all of the PCBs contained in the suspended sediment that is known to emanate from the spring during a storm, would go untreated directly to Clear Creek. To stop the treatment process at 1,000 g.p.m. without a storage and settlement train in the treatment process will allow more PCB sediment to enter Clear Creek during one storm event than a low flow portable plant could capture in many months of operation at low flow.

⁸ The main purpose for the plant is to capture the maximum amount of PCB sediment released during a storm event or high flow, as the amount of suspended solids released during low flow is minor when compared to a high flow situation. To maximize the amount of PCB sediment laden water that can be treated during a storm event, U.S. EPA decided to partition off the surface runoff from the spring runoff at the site of the spring emergence. Because the spring is located in a valley, this can be simply done by constructing a storm water separation wall which will greatly reduce the amount of clean surface runoff that otherwise would be mixing with the contaminated runoff during a storm event. The clean surface water will be routed untreated to the Clear Creek at a location down stream from the plant, via a 24" bypass, while the water containing the PCB sediment will proceed to the plant via its own 24" pipeline. Accordingly, U.S. EPA has considered the matter of capacity and has taken those steps that are cost effective to reduce the total volume of water that needs to be treated. U.S. EPA decided not to trust the integrity of the decades old conduit under the railroad tracks of the Indiana Central Railroad and, therefore, constructed two "jack and bores", one the 24" clean water by-pass and the other the 24" contaminated spring water pipeline, at a cost of $150,000.

Further, the plant had to be located somewhere. The fact that the City of Bloomington owned a piece of property that it was willing to make available without cost to U.S. EPA, that was isolated, the right size, and that was easily accessible for construction, are among the reasons the plant was located where it is.

The Quarry Spring location advocated by CBS property is private property. Thus, long acquisition negotiations and/or condemnation proceedings could be anticipated. Further, given the area and nature of the Quarry Spring location it is likely that the City and/or the residents might not want either the plant or the construction traffic there.

¹⁰ U.S. EPA has not been presented with any geophysical work that contradicts U.S. EPA's work and its conclusion that it is unsound to rely on a lagoon built on karst for storm water storage.