OPINION and ORDER
FEIKENS, District Judge.
I.
BACKGROUND
Plaintiff United States of America (plaintiff), initiated this action based on the Comprehensive Environmental Response Compensation Liability Act (CÉRCLA)
to recover costs incurred by the Environmental Protection Agency (EPA) during the environmental cleanup of a landfill located in Lapeer County, Michigan. Plaintiff and a number of potentially responsible parties (PRPs) engaged in settlement negotiations prior to this suit’s filing on March 31, Í995.
The three defendants in the present dispute, Dow Chemical Company, Akzo Nobel Coatings, Inc., and Gage Products Company, (defendants) were not parties to the settlement agreements that arose out of these negotiations. This opinion and order rules upon their joint motion for summary judgment on statute of limitations grounds and upon plaintiffs cross motion for summary judgment on the same issue. For the reasons that follow, defendants’ motion is denied and plaintiffs motion is granted. Plaintiffs complaint was timely filed.
II.
FACTS
The Metamora Landfill, located in Lapeer County, Michigan, opened for business in 1955 on the site of a former gravel pit. For a quarter century, the landfill accepted municipal and industrial waste. It closed in 1980. Before closing, the landfill drew the attention of the Michigan Department of Natural Resources (MDNR), the state agency that investigated fires on the site.
MDNR’s investigation revealed that chemical wastes had been dumped in several areas of the 80 acre site. In the early 1980’s, drums were discovered during construction of a solid waste transfer station on the site. MDNR sampled the drums and discovered that they contained hazardous waste.
Further tests followed. An MDNR magnetometer survey conducted in 1982 estimated that “approximately 35,000 whole drums exist” in five areas of the site. Areas 1 and 4 contained the lion’s share with about 25,000 drums, while the remaining areas accounted for the rest. Accordingly, remedial efforts were directed primarily at areas 1 and 4. In the fall of 1982, excavation of both the drums and surrounding soils confirmed that areas 1 and 4 contained hazardous waste. Pursuant to this series of tests, the site was evaluated under CERCLA’s Hazard Ranking System
and two years later was placed on the National Priorities List (NPL)
and thereby made a Superfund Site.
Once designated as a Superfund site in September 1984, remediation efforts at the landfill began to take shape. MDNR hired E.C. Jordan Co., a private contractor, to conduct a preliminary site investigation (SI). This hydrogeologic investigation began in March 1985 and accomplished a number of objectives including the determination of groundwater flow and site geology, the installation of thirteen new monitoring wells, and the collection and analysis of soil and groundwater samples.
The Si’s findings of soil and groundwater contamination prompted MDNR and EPA to order a “Phased Feasibility Study” (PFS). The stated goals of this study were to:
1. evaluate existing information to determine the extent to which each of the several suspected drum disposal areas can be assessed,
2. identify potential receptors and develop remedial action objectives,
3. identify and screen potential remedial alternatives,
4. conduct a detailed review of the alternatives identified to compare the effectiveness of the technologies and to aid in the selection of a remedy, and
5. submit to MDNR a final report documenting the study and presenting recommendations.
The PFS, which lasted from October 1985 to May 1986, explicitly indicated that it was only a prelude to a more “comprehensive” Remedial Investigation/Feasibility Study (RI/FS); the PFS Final Report contemplated that the RTFS would more fully explore issues like source removal and the management of contamination migration.
The PFS Final Report did, however, describe in some detail the dire condition of the site:
Data available on the Metamora Site indicates that drums containing hazardous chemicals are present and that some of these chemicals have been detected in groundwater underlying the Site. During drum excavation activities, drums were found in varying stages of integrity. Con
tinued degradation of drums and generation of additional contaminants to the groundwater is expected. Therefore, there is potential that further contamination of groundwater will occur- and water supplies located down gradient from the Site could be impacted by these chemicals if nothing is done to reduce the potential for further contaminant loading and contamination migration away from the landfill.
These concerns led to the report’s recommendation that drums in Areas 1 and 4 be excavated and sent off site for incineration.
This recommendation was only tentative and, as some of the public comment on the PFS had suggested, contingent on information that would be unearthed during the more detailed RI/FS to follow.
Indeed, the SI and PFS merely laid the groundwork for the site’s more detañed RI/FS. This latter study had its genesis in a 1985 agreement between MDNR and the United States Environmental Protection Agency (EPA). Among other things, the agreement pledged the federal funds by July 1985 necessary to conduct the RI/FS.
Whüe the RI/FS was stñl in its infancy, the EPA published a Record of Decision (ROD I) in September 1986 outlining the work that would be covered in the first phase or “operable unit” (OU-1) of the remediation. EPA contemplated that OU-1 would involve the excavation and disposal of drummed wastes and surrounding soñs in Areas 1 and 4 of the landfiU. The ROD selecting remedial source control measures acknowledged that EPA and MDNR were simultaneously undertaking an additional RI/FS “to evaluate the necessity for soü, groundwater and other remedial action.” ROD I left open the possibility that the information garnered in the RI/FS might necessitate further corrective action and, therefore, the publication of a new or amended ROD.
ROD I estimated that it would cost $90,000 to construct the two staging/storage pads upon which excavated drums would be placed prior to disposal. Construction of the pads began on or about November 3, 1986, following the ROD I’s publication by approximately one month. Plaintiffs do not dispute defendants’ characterization of the work necessary to complete construction of the 125 feet long by 125 feet wide pads:
After the soü was buüdozed and-compacted to provide a flat base, a 6-inch layer of mason sand was spread over each area with a front-end loader, and compacted with a vibratory compactor. Synthetic geofabrie and a 10-mü PVC liner were placed on top of the sand. Another 6-ineh layer of sand was spread atop the fabric and the liner, and again compacted. Then another layer of geofabrie and liner was placed, foUowed by a third layer of compacted sand. A three-foot high berm, in which soü and mason sand were interleaved with geofabrie and PVC liner, was then constructed around each pad. A single access ramp was bmlt into the staging/storage pad for Area 4, and two ramps were bmlt into the pad for Area 1. A security fence was erected around the two pads to control access.
According to defendants, calculations, the pads alone required about 175 15-ton truck loads of mason sand (2604 tons) and about 7,000 square yards of geofabrie and PVC liner.
MDNR began using the pads on December 5, 1986, when it started excavating drums from Areas 1 and 4. The excavation continued for two weeks. The 235 drums excavated during this time were stored in “over-pack” containers and placed on the pads. The drums were sampled in January 1987 and, again, in March and April 1987 to determine their content for manifesting and incin
eration as well as to conduct further chemical analysis. They remained on the pads for approximately 18 months while these tests were conducted. MDNR contracted with a firm called MAECORP to dispose of the 235 tested drums and, from March 1988 to June 1988, the contract was completed.
E.C. Jordan utilized the information obtained during the early 1987 RI/FS testing and sampling effort to prepare a February 1988 report. According to the report, it
was prepared in conjunction with the plans and specifications for the Metamora Landfill Drum Removal Operable Unit [ i.e., OU-1]. The report- is intended to provide a record of the background information and assumptions used to prepare the design. This is not an official bid document, nor is it meant to be used as one. Specific requirements for work to be performed in the drum removal are contained in the specifications and drawings.
Although this report was a significant step toward implementation of the work called for in ROD I, before the ROD I drum removal effort could begin, at least three important hurdles stood in the way. First, EPA and MDNR did not secure the first $20 million in federal funds necessary for the RI/FS, remedial design, and implementation of remedial action that ROD I specified until February 17, 1988. Second, MDNR had to secure a contractor for the source control remedy (i.e., drum excavation and disposal). This step was accomplished in September 1988 when MDNR designated Chemical Waste Management (CWM) of Chicago, Illinois as contractor. The contract required CWM to use the staging/storage pads constructed in November 1986 for the duration of the remedial action. Finally, EPA had to formally approve the final remedial design and authorize OU-1 to proceed. This approval and authorization did not occur until March 31, 1989 at about the same time as E.C. Jordan’s publication of the RI for the second phase or operable unit (OU-2) of the clean up.
The parties sharply dispute whether, prior to this formal approval, physical on-site construction for OU-1 began. Aside from the construction of the staging/storage pads, defendants have included exhibits which they claim prove MDNR installed mobile office trailers and utility hook ups to the site prior to March 31,1989. Exhibit 11 of defendants’ summary judgment brief is a contract release between MDNR and American Mobile Office dated 3/06/89. Although nothing on the record confirms that it was honored, the release contains, a delivery request date of 3/13/89. Likewise, defendants point to CWM daily work sheets and daily operations logs dated 3/22/89 and 3/28/89 as proof that electrical and portable sanitation services were installed on those dates.
Plaintiff points to other documents and deposition testimony to the contrary. Plaintiff cites a CWM daily work sheet dated April 5, 1989 that says telephone and electrical service was installed on that date.
Another CWM daily work sheet dated
4/4/89
states that the portable office trailers were not delivered until that date .
It is clear, however, that in May 1989, using the pads constructed in November 1986, CWM commenced the full scale excavation, . sampling, storage, and disposal of drummed wastes from Areas 1 and 4. Between May 1989 and mid-December 1989 CWM removed 23,630 drums.
While CWM’s drum removal (OU-1) work proceeded, the design for the second-phase of the project (OU-2) was well underway. Using the March 1989 Remedial Investigation for OU-2 as a starting point, E.C. Jordan completed and published its November 1989 Feasibility Study (FS) which evaluated proposed remedies to the threat identified in the RI. Approximately one year later, EPA issued a Record of Decision for OU-2 (ROD II), which selected the remedies for the “groundwater contamination, exposure to contaminated soils and potential contamination caused by leachate from the landfill.”
The remedy “inelude[d] a cap of the landfill and a groundwater pump-and-treat system” as well as the installation of fencing, the installation of a passive gas system, and, finally, the drafting of institutional controls to limit use of both the site and its groundwater.
Shortly after EPA published ROD II, EPA and a group of potentially responsible parties (PRPs) settled the United States’ claims relating to prospective OU-1 site remediation costs. On December 15, 1990, this group of PRPS and the United States entered a consent decree in which the PRPs agreed to assume responsibility for performing the work called for in OU-2 and for completing the excavation, sampling and disposal of the Metamora drums called for in OU-1.
III.
LAW
A.
Standard of Review
The parties’ cross motions for summary judgment are evaluated under Fed.R.Civ.P. 56. Under that rule summary judgment:
shall be rendered forthwith if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law----
The United States Supreme Court interpreted this language in
Matsushita Electric Industrial Co., Ltd. v. Zenith Radio Carp.,
475 U.S. 574, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986),
on remand,
807 F.2d 44 (C.A.3 1986):
Where the record taken as a whole could not lead a rational trier of fact to find for the non-moving party, there is no “genuine issue for trial.”
Id.,
106 S.Ct. at 1356.
The United States Court of Appeals for the Sixth Circuit clarified another of the so-called Celotex
trilogy,
Anderson v. Liberty Lobby, Inc.,
477 U.S. 242, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986), by holding that there is no triable issue:
unless there is sufficient evidence favoring the nonmoving party for a jury to return a verdict for that party---- Thus,the [Supreme] Court concluded, “The mere existence of a scintilla of evidence in support of the [non-moving party] plaintiffs position will be insufficientjthere must be some evidence on which the jury could reasonably find for the [non-moving party]...”
Street v. J.C. Bradford & Co.,
886 F.2d 1472 (C.A.6 1989) (Citations omitted.)
While it is clear that the above standard requires more than a scintilla of evidence by the nonmoving party to defeat the motion, nothing in the standard requires anything close to conclusive evidence — only enough evidence to convince a reasonable fact finder that the nonmovant is entitled to relief. In the present case, plaintiff presents evidence sufficient to defeat the defendant’s motion for summary judgment on the statute of limitations issue. However, defendants have failed to present the evidence necessary to defeat plaintiff’s motion. Accordingly, plaintiff’s motion is granted and defendants’ motion is denied. Plaintiff filed its complaint within the statute of limitations.
B.
Substantive Law
CERCLA’s statute of limitations, set forth at 42 U.S.C. 9613(g)(2) provides in pertinent part:
An initial action for the recovery of the costs referred to in section 9607 of this title must be commenced—
(A) for a removal action, within 3 years after the completion of the removal action
•fc
(B) for a remedial action, within 6 years after the initiation of physical on-site construction of the remedial action, except that, if the remedial action is initiated within 3 years after the completion of the removal action, costs incurred in the removal action may be recovered in
the cost recovery action brought under this subparagraph.
The timeliness of plaintiffs complaint turns, in the first instance, upon whether the recovery plaintiff seeks is for removal action or for remedial action.
The term “removal” is defined by CERC-LA, 42 U.S.C. 9601(23) to mean:
the cleanup or removal of released hazardous substances from the environment, such actions as may be necessary taken in the event of the threat of release of hazardous substances into the environment, such actions as may be necessary to monitor, assess, and evaluate the release or threat of release of hazardous substances, the disposal of removed material, or the taking of such other actions, as may be necessary to prevent, minimize, or mitigate damage to the public health or welfare or to the environment, which may otherwise result from a release or threat of release ...
The definition of “remedial” under 42 U.S.C. 9601(24) is similar:
The terms “remedy” or “remedial action” means [sic] those actions consistent with permanent remedy taken instead of or in addition to removal actions in the event of a release of threatened release of a hazardous substance into the environment, to prevent or minimize the release of hazardous substances so that they do not migrate to cause substantial danger to present or future public health or welfare or the environment. The term includes, but is not limited to, such actions at the location of the release as storage, confinement, perimeter protection using dikes, trenches, or ditches, clay cover, neutralization, cleanup of released hazardous substances and associated contaminated materials, recycling or reuse, diversion, destruction, segregation of reactive wastes, dredging or excavations, repair or replacement of leaking containers, collection of leachate and runoff, onsite treatment or incineration, provision of alternative water supplies, and any món-itoring reasonably required to assure that such actions protect the public health and welfare and the environment ... [T]he term includes offsite storage, treatment, destruction, or secure disposition of hazardous substances and associated contaminated materials.
The above statutes should be interpreted with the United States Supreme Court’s frequent admonition to strictly' construe statutes of limitation in favor of the Government where application of them might otherwise bar its rights.
Badaracco v. Commissioner,
464 U.S. 386, 391-392, 104 S.Ct. 756, 78 L.Ed.2d 549 (1984),
E.I. Dupont De Nemours & Co. v. Davis,
264 U.S. 456, 44 S.Ct. 364, 68 L.Ed. 788 (1924).
1.
Costs Associated With OU-1
The conclusion that OU-1 was a remedial action is supported by the cases defendants cite.
See Sherwin-Williams Co. v. City of Hamtramck,
840 F.Supp. 470 (E.D.Mich.1993),
Union Carbide v. Thiokol Corp.,
890 F.Supp. 1035 (S.D.Ga.1994),
State of California v. Hyampom Lumber Co.,
903 F.Supp. 1389 (N.D.Cal.1995). These cases hold that remedial actions are generally response actions conducted in accordance with an ROD over an extended period of time in amounts exceeding 1 million dollars — conditions which all apply to OU-1. The issue that I must decide, however, has less to do .with the characterization of OU-1 as either removal or remedial than it has to do with deciding when work on OU-1 was initiated by physical on-site construction.
See
42 U.S.C. 9613(g)(2)(B) (Statute of limitations triggered for remedial action when there is .an “initiation of physical on-site construction of the remedial action.”)
Hyampom Lumber
employed a four-part test to determine when physical on-site construction occurs: 1) there is “physical” action; 2) that “physical”, action happens
“on-site” 3) the action must be part of the remedial action, and 4) the action must be the “initiation” of the remedial action. 903 F.Supp. at 1391. This test establishes a floor, meaning an event must at least meet these criteria to be considered the initiation of physical on-site construction. Nothing in
Hyampom Lumber,
however prohibits additional requirements from being considered. Thus, an action may meet the four conditions and still not constitute the initiation of physical on-site activity.
Hyampom Lumber
looked at a further requirement — that the action play a critical role in implementation of the permanent remedy — to determine whether events occurring in that case constituted an “initiation of physical on-site construction.”
Id.
at 1393-1394.
Defendants point to a number of events occurring prior to March 31,1989, the date exactly six years prior to the filing of plaintiff’s complaint, as potential initiations of physical on-site activity: the 1986 construction of the staging/storage pads and excavation of approximately 235 drums; the alleged March 1989 sanitation and utility hookups; and the alleged March 1989 trailer delivery. Defendants stress that all of these activities were provided for in either the ROD or Phased Feasibility Study.
Plaintiff contends that the remedial work on OU-1 did not commence until April 4, 1989, a date within the statute of limitations. Citing
Kelley ex rel. State of Michigan v. E.I. DuPont de Nemours,
786 F.Supp. 1268 (E.D.Mich.1992)
aff'd,
17 F.3d 836, 840 (C.A.6 1994),
which held that investigation and study of cleanups is “removal” action, plaintiff asserts that work occurring prior to March 31, 1989 was either part of the remedial design for OU-1 or part of the preliminary investigation necessary to formulate the RI/FS and ROD for OU-2 and, thus, a removal action.
I agree with plaintiff’s reasoning. Turning first to the activity immediately following the publication of the first ROD in Fall 1986— construction of the pads and limited drum excavation — the fact that the pads were constructed to accommodate the preliminary excavation of 235 drums for testing purposes does not necessarily lead to the conclusion that full scale remedial action was underway in late 1986. While it is true that the work performed in late 1986 resembled what was provided for in ROD I and the Phased Feasibility Study, the work was an integral part of both the RI/FS for OU-2 and the remedial design for OU-1. As plaintiff points out, “to the extent that certain 1986-1989 actions are reflective of the actions called [for] by the 1986 ROD, it is because MDNR was developing and testing — but not implementing — the source control remedy.”
ROD I clearly contemplated that this “developing and testing” would need to take place prior to implementation of the long term remedy that ROD I selected. It provided:
In order to complete the site response, an RI/FS has been initiated to study the potential impacts of contaminated soil, ground water, and other media. Test pits in areas one and four have been proposed in order to better define the number, condition, and contents of buried drums.
The field work for the test pits is expected to begin in November or December of 1986. The data from the test pits will be used during the remedial design for this operable unit so that better cost estimates for the project may be made.
This will allow potential remedial action contractors to submit more accurate bids for the construction of the operable unit. The RI/FS, which will evaluate alternatives for final site remediation, is scheduled for completion during the second quarter of FY ’88.
Another Record of Decision package shall be prepared for any additional remedial action recommended as a result of the RI/FS, or if test pit information warrants re-evaluation of this Record of Decision.
(Emphasis added.)
ROD I clearly foresaw that much investigatory work still had to be done before the actual implementation of OU-1. It clearly provides that MDNR could not properly evaluate bids for OU-1 until it knew more information about the site and the remedy. Common sense demands that, if MDNR lacked the information it needed to evaluate bids, MDNR could not possibly implement the work that those bids were supposed to cover.
United States v. Petersen Sand and Gravel,
824 F.Supp. 751 (N.D.Ill.1991), is not in conflict with this conclusion. In that case, the United States Court of Appeals for the Seventh Circuit held (and the EPA agreed) that the publication of an ROD marked the dividing line between removal and remedial actions. This holding is grounded on the well-settled rule that investigation and study of a site is “removal” action. Defendants argue that the ease stands for the proposition that, once the ROD is published, removal action terminates and remedial action begins. The holding in
Petersen Sand
is not so broad. While it is true that publication of the ROD
in that case
marked the dividing line between removal and remedial action, publication of the ROD in that case also marked the end of the site study and investigation. As stated above, publication of ROD I in the case at bar was clearly not the end of the study and investigation leading up to implementation of OU-1 and, therefore, not the end of the “removal” portion of OU-1.
ROD I clearly provided that further testing would be necessary before the actual work on OU-1 could begin. There is no indication that that was the case in
Petersen Sand
or either of the two eases it discussed,
United States v. Corbett Transport, Inc.,
785 F.Supp. 81 (E.D.Tex.1990), and
U.S. v. Jack Allen,
1990 WL 339488 (W.D.Ark.1990). In fact, the opposite was true in
Petersen Sand.
In
Petersen Sand
the remedy selected by EPA was “no further action.” No further study was necessary to implement that remedy because, quite simply, there was nothing to implement. Similar reasoning was employed in
Jack Allen
where, as in
Petersen Sand,
the court acknowledged that the removal action was not “complete until the EPA formally determined that no further on-site activity was needed ...”
Petersen Sand,
824 F.Supp. at 754. No such determination was ever made in the case at bar before March 31,1989.
Defendants assert that, even if the Fall 1986 activity was not the initiation of physical, on-site activity, the events of March 1989 were. Defendants point specifically to the alleged delivery of mobile office trailers and the installation of utility and sanitation hookups. As pointed out above, plaintiffs have produced documents that rebut that these events occurred in March 1989. According •to those documents cited above, the events occurred some time in April 1989, a period within the statute of limitations. At best, defendants create a factual dispute. The question then becomes whether the dispute is triable to a jury.
I hold that it is not and agree with plaintiff that the March 1989 events cited by defendant were merely “ ‘preliminary’ or ‘precon-struction’ ” steps and not, as
Hyampom Lumber
required, an integral step in the implementation of the permanent remedy provided for under ROD I.
Hyampom Lumber
stressed that:
[TJhe term [“construction”] [as used in CERCLA’s statute of limitation] ... serves the purpose of excluding those preliminary and tentative “physical 'on-site” activities that while related to the remedial action, are not part of its construction.
903 F.Supp. at 1392. Thus, even if defendants’ version of events proves to be the correct one, none of the events would constitute the initiation of physical on-site activity.
Accordingly, I credit plaintiff’s argument that physical on-site construction of the OU-1 remedy was initiated on April 4, 1989. Plaintiff timely filed suit on March 31, 1989 to recover the costs associated with this remedial action.
2.
Costs Associated With OU-2
In addition to the costs it paid for OU-1, EPA seeks to recover certain costs associated with OU-2. Specifically, EPA claims that it is entitled to costs it paid for performing the March 1989 RI, the November 1989 FS, and the September 1990 ROD (ROD II) which selected as remedies a landfill cap and pump and treat system to treat contaminated groundwater.
The parties do not dispute that these costs relate to “removal” activities. Under 42 U.S.C. 9613(g)(2)(A), EPA has three years from the date of the removal action’s completion to file a cost recovery suit. To determine whether plaintiffs claim for these removal costs, I must decide when the removal action ended. Defendants argue that the removal action ended in September 1990 with the publication of ROD II.
For the reasons stated above, I cannot accept such a rigid demarcation between “remedial”
and “removal.”
While it is true that some courts have concluded that the publication of an ROD marked the end of removal activity
under the facts of those
coses
, the statutory language upon which the cases rely, allows for a more searching analysis. 42 U.S.C. 9601(23) clearly provides that action taken to “monitor, asses and evaluate the release or threat of release of hazardous substances.” This language has been interpretéd to cover site investigation and study, of which remedial design is an important part.
Kelley, supra
This observation was not lost on plaintiff. Plaintiff argues that its efforts to refine the design for OU-2 prior to its implementation are ongoing and involve investigation, assessment and evaluation of the release or threat of release of hazardous substance at the site. Thus, plaintiff claims its cost recovery action filed on March 31, 1995 is within the statute of limitations.
Plaintiffs interpretation is consistent with the House Judiciary Committee Report to the Superfund Amendments- and Reauthori-zation Act of 1986 (SARA), 100 Stat 1613
et seq.
(1986), which included 42 U.S.C. 9613(g)(2). The report stated that remedial action begins “after the RI/FS and
after design of the remedy.”
H.Rep. No. 99-253, Pt. 3 99th Cong., 1st Sess.21 (1985). (Emphasis added.) Importantly, plaintiffs interpretation is consistent with the language of 42 U.S.C. 9613(g)(2)(B) because the phrase “physical onsite construction of the remedial action” implies implementation of the remedy as opposed to design or development of the remedy chosen by ROD II.
IV.
CONCLUSION
For the above reasons, defendants’ motion for summary judgment on the statute of limitations issue is denied. Plaintiffs cross motion for summary judgment is granted. Plaintiff timely filed its complaint.
IT IS SO ORDERED.