Allstate Life Insurance Co. v. BFA (2011)

MEMORANDUM OF DECISION

This is an action, brought pursuant to environmental indemnity agreements, to recover the costs for monitoring and remediating a contaminated property.   The language of the agreements was interpreted in the case of Allstate Life Ins. Co. v. BFA Limited Partnership, 287 Conn. 307, 948 A.2d 318 (2008).   The following pertinent facts are contained in that opinion:  “In January, 1991, the plaintiff loaned the defendants $8.2 million and the defendants executed a promissory note secured by a mortgage deed on a parcel of commercial property in Bristol (property).   As a loan condition, the plaintiff also required the defendants to execute an indemnity agreement entitled ‘Environmental Indemnity Agreement’ ‘in order to induce [the plaintiff] to accept the Property as security for the Loan and to enter into the Loan and to disburse the proceeds of the Loan ․’ Under the indemnity agreement, the defendants were required ‘to unconditionally indemnify, defend, and hold [the plaintiff] harmless against any loss, liability, damage, expense or claim arising under any Hazardous Material Law ․ [or] ․ resulting from the presence of Hazardous Material on the Property.”  Id., 309.

“In 1998, the plaintiff agreed to advance up to $2 million in additional funds to the defendants and to extend the maturity date of the loan.  ‘As a condition precedent to entering into the [loan] restructuring,’ the plaintiff required the defendants to execute the reaffirmation of the indemnity agreement on April 22, 1998 ․

“The defendants subsequently defaulted by failing to make payments on the loan, and the plaintiff commenced a foreclosure action on the property.   The parties entered into a stipulated judgment of strict foreclosure, and the plaintiff obtained title to the property on January 3, 2002.   The plaintiff did not move for a deficiency judgment.

“When the plaintiff sold the property on or about May 9, 2003, it was required to comply with the Connecticut Transfer Act, General Statutes § 22a–134 et seq., by performing an environmental site assessment and undertaking remediation of contamination on the property.   The site assessment revealed the presence of hazardous materials on the property that required remediation and monitoring, and the plaintiff consequently has incurred in excess of $100,000 in ongoing costs and fees.   The defendants refused to indemnify the plaintiff, and the plaintiff brought this action against the defendants to recover its costs and fees, claiming that the defendants had breached both the indemnity agreement and the reaffirmation of the indemnity agreement.”  Id., 310–11.

The Court reversed a summary judgment that had entered in favor of the defendants and remanded the case for further proceedings.  Id., 323.   The case was assigned to this court and a hearing was conducted over six days.   With the agreement of the parties, the hearing was bifurcated on the issues of liability under the agreements, and if necessary, damages.   The parties agree that in order for the defendants to be liable, the plaintiff must prove by a preponderance of the evidence that the contamination of hazardous materials occurred at the property prior to the date the plaintiff took title.   The parties have stipulated that the plaintiff took title to the property on January 4, 2002.¹

The parties do not dispute that the property is presently contaminated with tetrachloroethene (PCE), an organic solvent commonly used in dry cleaning.   The parties also do not dispute that a dry cleaning business operated on the property from approximately 1994 to the end of December 2002, and that the dry cleaning equipment used in the business was dismantled in April 2003.

The first documented findings of PCE contamination at the property occurred in October 2002, while the dry cleaning business was still in operation, six months before the dry cleaning equipment was removed from the property.   This first documented contamination was determined from soil samples taken from under the slab floor of the dry cleaning business, in the vicinity of the dry cleaning machines.

Additional testing was performed at the request of the plaintiff by Giatech.   In May 2003, four ground water monitoring wells were installed.   Groundwater samples taken at that time were negative for PCE. Four soil vapor samples were also taken in May 2003, and they were positive for PCE. Samples from the groundwater wells were tested in October 2003, and the results showed PCE in levels that exceeded regulatory criteria.

The plaintiff maintains that a common sense view of the evidence establishes that is more probable than not that the release of PCE occurred during the time the defendants owned the property.   The plaintiff points out that the dry cleaning establishment was in operation for approximately eight years when the defendants owned the property, and operated for only approximately one year while the plaintiff owned the property.   Spills of PCE can occur during the routine operation of a dry cleaning business due to spills from spot cleaning bottles, periodic filling of dry cleaning machines with PCE, and during servicing.   In essence, the plaintiff argues that because the period of operation of the business during the defendants' ownership was eight years, and the period of operation during the plaintiff's ownership of the property was one year, common sense suggests that the contamination most likely occurred during the longer period of time when the defendants owned the property.   The court is not persuaded.

The evidence did establish that dry cleaners can release PCE contaminants;  the evidence did not establish that all dry cleaners must release contaminants.   It is important to note that neither party presented any evidence surrounding the operations of the subject dry cleaning business during the nine years it was located upon the property.   The parties did not call any witnesses who were familiar with the operation of the dry cleaning business, or who were present when the dry cleaning equipment was dismantled.  “Drawing logical deductions and making reasonable inferences from facts in evidence, whether that evidence be oral or circumstantial, is a recognized and proper procedure in determining the rights and obligations of litigants, but to be logical and reasonable they must rest upon some basis of definite facts, and any conclusion reached without such evidential basis is a mere surmise or guess.”  (Emphasis added;  internal quotation marks omitted.)  Paige v. St. Andrew's Roman Catholic Church Corp., 250 Conn. 14, 34, 734 A.2d 85 (1999).   A calendar alone does not provide sufficient proof to satisfy the plaintiff's burden.

It does not appear that the plaintiff took any soil, vapor or water samples prior to the time that it took title to the property, on January 4, 2002.   The plaintiff is a sophisticated commercial entity and it is reasonable to infer that it understood the importance of determining whether contamination existed during the period of time that the plaintiffs owned the property.   The plaintiff commenced its foreclosure in May 2001, and conducted a Phase 1 Environmental Site Assessment in June 2001, prior to taking title through the foreclosure.   That assessment stated:  “Allstate may wish to consider a subsurface investigation in the vicinity of the dry cleaners to determine whether there has been impact to the soil or groundwater from the dry cleaning operations.”   Defendants' exhibit I. As noted earlier, the first samples were taken in October 2002, ten months after the plaintiff took title to the property, and while the dry cleaning business continued in operation.

Because the plaintiff did not test the property before it took title, it must use empirical evidence to establish that the PCE contamination occurred before January 4, 2002.   The established fact is that PCE was detected at a groundwater monitoring well, MW 110, on November 29, 2007.   This monitoring well was a minimum of 145 feet from the assumed source location, which was the midpoint of the exterior wall of the dry cleaner premises.   In a model based upon a time-speed-distance equation, the length of travel time for a contaminant can be calculated.   In a typical time-speed-distance equation any one element of the equation can be determined, if any of the other two elements of the equation are known with reasonable certainty.   In this case the speed of the subsurface ground water was never established by objective measurement.   The evidence established that analytical calculations can be used to determine the velocity or speed of subsurface ground water and contaminants.   As will be discussed below, the analytical calculations utilize various factors.

The challenge in using an analytical calculation for subsurface ground water speed is similar to determining the speed of an automobile, not from a speedometer, but from calculations using component factors such as the automobile's engine revolutions per minute (r.p.m.), gear ratios and tire circumference.   The calculations are made more difficult when, as will be explained in a groundwater model, those component factors are not known from objective measurement, but must be assumed from a range of possible values, i.e. a range of r.p.m.s, various gear ratios and various tire circumferences.   An incorrect assumption for any one, or more, of these variables will affect the reliability of the determination of speed.   In a time-speed-distance equation involving the automobile's travel, a lack of reliability in the speed value affects the reliability of the result of the equation.

Similarly, in this case, the speed of the subsurface groundwater is not known from objective measurement and must be calculated using an equation that has component values that are derived from a range of possible variables.   Expert testimony established the component values in determining groundwater speed.   The first is hydraulic conductivity, which is an expression of resistance to groundwater when passing through soil.   The second value is horizontal gradient, which is a measure of how much the surface of the groundwater rises or falls over a given distance;  the steeper the decline in the surface of the water table the faster the groundwater moves.   The third value is porosity, which reflects the amount of space between the particles and stones in the soil.   Those values can be used to establish the speed of the ground water;  what is, however, of interest in this case is the travel speed of the contaminants in the groundwater.   Expert testimony demonstrated that contaminants travel in groundwater at a speed that is slower than uncontaminated groundwater.   This phenomenon is known as retardation, and it is a factor that is also affected by variables such fraction of organic compound ² and porosity.

The retardation factor, when applied to the groundwater speed, results in the speed of transport of the contaminants.   To further the automobile analogy, it is similar to attempting to determine the speed, from the aforementioned calculations, of an automobile that is also towing a trailer, however the weight of the trailer is unknown;  it can be a very light trailer, or a very heavy trailer.   Any incorrect assumption as to the trailer weight, and its effect on the underlying calculations, will affect the overall calculation to determine the speed of the vehicle.

Stephen Quigley, a well qualified environmental engineer, testified as an expert witness on behalf of the plaintiff.   He performed a contaminant speed analysis for the property and determined that it took the PCE contaminants between 5.9 and 7.7 years to travel the 145 feet from the wall of the dry cleaners to monitoring well MW 110, where PCE contaminants were measured on November 29, 2007.   Mr. Quigley then calculated backwards and concluded that if PCE from the assumed location required the determined travel times to reach well MW 110, the latest release date was January 2002, and the earliest release date was March 2000.   It was his expert opinion that the release occurred before January 4, 2002.   In addition to the transport speed calculations, Mr. Quigley indicated that his opinion was influenced by the fact that the level of PCE detected at the monitoring well was relatively high.   He opined that to reach such a high level, the PCE from the assumed source must have been traveling in the groundwater for a longer period than the shortest determined travel time.   When asked by the court, Mr. Quigley could not say there was an analytical calculation that supported that conclusion.

In order to determine the groundwater velocity, Mr. Quigley had to establish hydraulic conductivity values.   As noted earlier, conductivity is a measure of resistance of groundwater through the soil.   To determine this value, he utilized slug test ³ results from four wells, using two accepted methods of analysis, Bouwer–Rice and Dagan.   Of the four wells utilized, two were within 145 feet of the dry cleaners, and two were located approximately 700 feet from the dry cleaners.   Mr. Quigley stated that using a geometric mean for the results from the four wells located throughout the site would give a more representative value for the soil located in the path of travel of the contaminants.   Common sense would indicate that slug test results from wells in the vicinity of the travel path, rather than distant wells, would more likely give a more accurate representation of soil conductivity in the area in question.   The court understands, however, the basis of Mr. Quigley's opinion on this point.   The next component of the groundwater speed equation is gradient.   This component is derived from using groundwater level contours for the site.   Mr. Quigley used the elevation difference in two wells MW 110, and MW 3 which is located approximately sixty feet from the designated source area.   The final variable in determining groundwater speed is porosity, which consists of the spaces between the rocks and organic material particles in the aquifer.   The higher the porosity value, the more groundwater is slowed.   Mr. Quigley used a porosity factor of 0.30.

It is important to note that the specific location of the release(s) of contaminants, known as a “hot spot(s),” has never been determined at this site.   Mr. Quigley reasonably assumed that the dry cleaning business was the origin of the contaminants.   To establish the starting point for his distance calculation, he used the midpoint of the exterior wall of the dry cleaner that was closest to MW 110 as an assumed location of the release of the PCE contaminants.   This midpoint may result in a shorter distance to the monitoring well, if the release area was inside the dry cleaning business, and it may be also farther from the release point, if the release occurred outside of the dry cleaning business, in an area between the wall and the monitoring well.   Obviously, variations in the travel distance, shorter or longer, would affect the time result in the time-speed-distance calculation.

Unfortunately, no objective measurement of groundwater speed using tracer materials and monitoring wells was ever performed.   Consequently, the only evidence regarding the speed of the groundwater and PCE contaminants comes from calculations performed by Mr. Quigley.

The defendants presented expert testimony from Marc Casslar, a Connecticut licensed environmental professional.   He has extensive experience in evaluating and remediating PCE contamination from dry cleaners.   His testimony regarding dry cleaner operations and mechanisms of contaminant spills was helpful to the court.   He testified that Phase I studies conducted at the property in 1998 and 2001 showed no evidence of environmental contamination.   Although he acknowledged that a small amount of PCEs were detected in sub-slab soil samples in October 2002, he emphasized that tests performed on soil vapor samples conducted in May 2003 revealed a significant level of contamination.   It was Mr. Casslar's opinion that the spill most likely occurred between December 2002 and May 2003, after the January 4, 2002 foreclosure date.   He testified that it was possible the release of PCE contaminants was caused by improper dismantling and disposal of the dry cleaning machines.   Mr. Casslar also testified as to his familiarity with the principles of modeling employed by Mr. Quigley.   Mr. Casslar was not, however, qualified to give an opinion as to the groundwater/contaminant travel times.

Tracy Roth, a hydrogeologist, testified on behalf of the defendants.   She has extensive experience with subsurface groundwater models.   She had previously reviewed the data and values utilized in Mr. Quigley's calculations that formed the basis of his opinion as to the date of release of PCE contaminants.   She acknowledged that the calculations and equations used by Mr. Quigley are appropriate in some circumstances, however she characterized them as simplistic, preliminary and one dimensional.   Her opinion was that the values used by Mr. Quigley biased the calculations to favor a slower groundwater/contaminant speed.   These values resulted in a longer travel time, which supported Mr. Quigley's opinion as to a release prior to January 4, 2002.

Ms. Roth testified that the conductivity values derived from a geometric mean utilizing data from the two distant wells did not accurately reflect the soil conditions in the contaminant travel path.   This testimony validates the court's common sense observation, mentioned earlier.   She noted that the data for the wells closest to the travel path indicated faster conductivity times.   As to gradient, she believed that the measurements used did not take into consideration the bedrock conditions in the vicinity of the dry cleaners and the gradient in the travel path.   Ms. Roth opined that the measurements used by Mr. Quigley resulted in a lower gradient which would produce a slower travel time.   She further opined that the porosity factor used by Mr. Quigley, 0.30, was too high, and a porosity value of 0.10 to 0.15 should have been used.   Ms. Roth also noted that Mr. Quigley's calculations can only be considered estimates since the actual arrival time of the PCEs at the monitoring well, and the exact location the source of the contaminants, are not known.

Two other areas of Ms. Roth's testimony were of interest to the court.   The first pertained to a dual-domain approach for contaminant transport.   This approach considers the heterogeneity of sediments in the aquifer and recognizes that contaminants may travel though more permeable soils, such as sand and gravel, at a significantly higher rate than average velocities.   Mr. Quigley testified that the subject property did have “lenses” or areas of sand and gravel.   The other area of interest pertained to the utilization of other lines of evidence.   Ms. Roth testified that there are tests available to determine the rate of breakdown of PCEs, which could have been used as a check on Mr. Quigley's travel time calculations.

Ms. Roth did not offer an opinion as to travel time of the contaminants involved in this case or an opinion as to the date of release.   Ms. Roth did testify persuasively and convincingly regarding her professional criticisms of Mr. Quigley's approach and his calculations.   The plaintiff offered no rebuttal to Ms. Roth's testimony.

The plaintiff's case essentially rested upon Mr. Quigley's opinion regarding the date of release of contaminants which was derived from his groundwater/contaminant travel time calculations.

“It is well established that [i]n a case tried before a court, the trial judge is the sole arbiter of the credibility of the witnesses and the weight to be given specific testimony ․ The credibility and the weight of expert testimony is judged by the same standard, and the trial court is privileged to adopt whatever testimony [it] reasonably believes to be credible ․ On appeal, we do not retry the facts or pass on the credibility of witnesses.”  (Internal quotation marks omitted.)  United Technologies Corp. v. East Windsor, 262 Conn. 11, 26, 807 A.2d 955 (2002).  “[I]n its consideration of the testimony of an expert witness, the [trier of fact] might weigh, as it sees fit, the expert's expertise, his opportunity to observe the defendant and to form an opinion, and his thoroughness.   It might consider also the reasonableness of his judgments about the underlying facts and of the conclusions which he drew from them ․ [T]he [trier of fact] can disbelieve any or all of the evidence ․ and can construe that evidence in a manner different from the parties' assertions ․ It is the trier of fact's function to consider, sift and weigh all the evidence ․” State v. Griffin, 78 Conn.App. 646, 656–57, 828 A.2d 651 (2003).

Mr. Quigley was a knowledgeable, prepared, and articulate expert witness;  the court does not doubt his personal credibility.   After sifting and weighing all the evidence, the court concludes that the groundwater/contaminant speed calculations that form the basis of Mr. Quigley's opinion do not have the necessary degree of reliability needed to establish the validity of the opinion.   The reliability of the opinion is adversely affected by Ms. Roth's persuasive and cogent criticism of the variables used, and the fact that dual-domain conditions were not reflected in the calculations supporting the opinion.   The reliability of Mr. Quigley's opinion was also diminished by the lack of objective measurements and data which could have supported or corroborated the opinion.

After a careful consideration of the evidence presented, and a though review of the well written briefs of the parties, the court concludes that the plaintiff has not established by a preponderance of the evidence that release of contaminants at issue here occurred before January 4, 2002.   The plaintiff has not established liability so there are no damages to be determined.

Judgment may enter in favor of the defendants.

Domnarski, J.

FOOTNOTES

FN1. The difference between the date stated in the facts of Allstate Life Ins. Co. v. BFA Limited Partnership, 287 Conn. 307, 310, 948 A.2d 318 (2008), January 3, 2002, and this stipulation, January 4, 2002, was not explained at trial..  FN1. The difference between the date stated in the facts of Allstate Life Ins. Co. v. BFA Limited Partnership, 287 Conn. 307, 310, 948 A.2d 318 (2008), January 3, 2002, and this stipulation, January 4, 2002, was not explained at trial.

FN2. Fraction of organic compound is an expression of the amount of organic carbon or matter that is part of the soil.   PCE has an affinity for organic carbon, which can retard the flow of PCE through the soil..  FN2. Fraction of organic compound is an expression of the amount of organic carbon or matter that is part of the soil.   PCE has an affinity for organic carbon, which can retard the flow of PCE through the soil.

FN3. A slug test measures groundwater recovery times.   Measurements are taken when a slug or rod is introduced into the groundwater, which the slug displaces, and are also taken when the rod is removed and the ground water returns to its static elevation..  FN3. A slug test measures groundwater recovery times.   Measurements are taken when a slug or rod is introduced into the groundwater, which the slug displaces, and are also taken when the rod is removed and the ground water returns to its static elevation.

Domnarski, Edward S., J.