The opinion of the court was delivered by
ARNOLD M. STEIN, J.A.D.Badische Corporation. (BASF) appeals from the Tax Court’s review of the 1992 tax assessment of its property at 50 Central Avenue in Kearny. The property is an owner-occupied chemical plant located on 27.136 acres of industrial property. As of the valuation date of October 1,1991, the plant was no longer operating.
The challenged assessment was:
Land: $1,899,500
Improvements: $3,670,500
Total: $5,570,000
The Chapter 123 Ratio, N.J.S.A. 54:51A-6, for the Town of Kearny was 53.77% for the 1992 tax year. Applying this ratio to the assessment indicates an assessed fair market value of $10,358,-936 ($5,570,000/.5377).
In a reported decision, the Tax Court found the true value of the subject property on October 1,1991 to be:
Land: $3,527,700
Improvements: $3,832,553
Total: $7,360,253 (rounded to $7,360,200)
Badische Corp. v. Town of Kearny (Badische II), 14 N.J.Tax 219, 231 (Tax 1994). After applying the 53.77% ratio, the Tax Court reassessed the property at:
Land: $1,345,600
Improvements: $2,612,000
Total: $3,957,600
Id. at 232.
Before making adjustments for economic obsolescence and environmental contamination, BASF’s valuation expert appraised the land value at $100,000 per acre, for a total of $2,713,600 which he rounded to $2,715,000. His valuation was based on the average adjusted per acre sale price of six other industrial properties (comparable sales):
*175Sale Sale Date Acres Unadjusted Price/acre
1 Mar. 1985 13.820 $104,962
2 Nov. 1985 8.310 $110,000
3 Feb. 1987 4.756 $110,000
4 Oct. 1989 3.256 $117,644
5 Jan. 1988 3.376 $103,673
6 Nov. 1991 12.312 $159,593
Adjustments to Sale Price Adjusted
Sale Time Size Location Utility Approvals Price/acre
1 17.85% - - -10.00% - $113,199
2 14.79% - - 5.00% -15.00% - $104,271
3 8.39% -10.00% -10.00% - $ 97,228
4 - - -10.00% -10.00% - $ 94,115
5 3.72% -30.00% -20.00% -10.00% - $ 86,797
6 - - - -10.00% -20.00% $111,715
The Tax Court rejected the expert’s land value estimate of $100,000 per acre as flawed because three of the six comparable sales involved much smaller lots. Badische II, supra, 14 N.J.Tax at 230.
“[Cjomparable sales include all sales that will lend logical, coherent support if they show ‘comparable building density ratios, functional similarities, proximity of sale dates to assessing dates, similarity of age, construction and condition, and in some cases, size.’ ” Ford Motor Co. v. Township of Edison, 127 N.J. 290, 308, 604 A.2d 580 (1992) (quoting Shulton, Inc. v. City of Clifton, 7 N.J.Tax 208, 218 (Tax 1983), aff'd, 7 N.J.Tax 220 (App.Div.1984)). “Appellate courts have long recognized that the trial court must be granted ‘a wide discretion’ in determining the admissibility of sales sought to be relied on as comparable.” Id. at 307, 604 A.2d 580 (quoting County of Los Angeles v. Fans, 48 Cal.2d 672, 312 P.2d 680, 684 (1957)). The property covered 27.136 acres. The discarded comparable sales ranged in size from 3.256 to 4.756 acres. It was well within the judge’s discretion to find the size difference between sales #3, #4, and # 5 and the subject property “vitiates comparability,” and to base his valuation *176on the three remaining properties. Badisehe II, supra, 14 N.J.Tax at 230.
The judge then calculated the land value on the basis of the remaining comparable sales. He eliminated the expert’s downward adjustments for utility, finding them “subjective and not supported by hard data.” Ibid. The adjusted price per acre for the remaining properties was calculated as: # 1, $123,700; # 2, $126,700; and # 6, $143,633. Ibid. Averaging and rounding the three figures, the judge found the value for the property to be $130,000 per acre. Id. at 230-31. The total value for all 27.136 acres was then calculated as $3,527,700. Id. at 231.
We reject BASF’s claim that the judge’s finding conflicts with his 1988 assessment of $120,000 per acre for the same property, Badische v. Town of Kearny, (Badische I), 11 N.J.Tax 385, 403 (Tax 1990), because of an “obvious declining real estate market.” BASF presented no evidence to support this claim. In fact, the most recent comparable sale in BASF’s appraisal report had the highest per acre price of $159,593. At most, the report indicates a flat real estate market since October 1988. No adjustments for time were made past that date and a 5% annual adjustment was made for pre-1988 sales. The Tax Court’s fact-findings are supported by substantial credible evidence. Southbridge Park, Inc. v. Borough of Fort Lee, 201 N.J.Super. 91, 94, 492 A.2d 1026 (App.Div.1985).
Nevertheless, the issue must be remanded because it appears the judge may have inadvertently miscalculated the adjusted price for sales # 2 and # 6. Sale # 1 appears correct. $104,962, adjusted 17.85% for time, equals $123,697.71. Rounded this equals $123,700, the court’s figure.
However, the judge did not make a -5% adjustment ( — $5,500) for location for sale # 2, and faded to indicate whether he rejected this adjustment. If he did not, the price per acre should have been calculated as $120,700 ($110,000, adjusted 14.79% for time, *177equals $126,269, rounded to $126,270, the court’s figure. $126,270 minus $5,500 equals $120,770).
For sale # 6, the judge made no finding as to the -20% approvals adjustment. Although he said in his opinion that he rejected BASF’s utility adjustment of -10% ( — $15,959.30), Badische II, supra, at 230, it appears this utility adjustment was applied and the approvals adjustment ( — $31,918.60) rejected instead. ($159,593 adjusted -10% equals $143,633.70. Rounded to $143,633 — the court’s figure). If the judge rejected only the utility adjustment as indicated by his written opinion, the adjusted price of sale # 6 should have been $127,674 ($159,593 adjusted - 20% equals $127,674.40 rounded to $127,674). If he rejected both adjustments, the price should have remained $159,593 per acre.
Accordingly, we remand for determination of whether he rejected the location adjustment of -5% for sale # 2; and either the - 20% approvals adjustment or the -10% utilities adjustment, or both, for sale #6; and, if necessary, for recalculation of the adjusted prices and the land value of the property.
The judge rejected BASF’s 32% adjustment to the improvement value for economic obsolescence. BASF’s expert based this deduction on the fact that the buildings and equipment were at the end of their useful life and because it was no longer viable to continue producing the product. He testified he applied the same theory of economic obsolescence that he applied in his 1988 appraisal of the same property before the same Tax Court judge. See Badische I, supra, 11 N.J.Tax at 394-95.
The judge rejected BASF’s economic obsolescence adjustment, finding (1) it was influenced by management’s decision to close down, (2) there was no credible evidence that plaintiff had been losing money for some years prior to the valuation date, and (3) it was not supported by “capitalized rent loss or by comparing sales of similar properties that were subject to the negative influence with those that were not.” Badische II, supra, 14 N.J.Tax at 231 (citing Appraisal Institute, The Appraisal of Real Estate, 358 (10th Ed.1992)).
*178Economic, or external, obsolescence “reflects a reduction in the value of property caused by factors extraneous to the property itself, such as changes in population characteristics and economic trends, excessive taxes, and governmental restrictions.” Transcontinental Gas Pipe Line Corp. v. Bernards Tp., 111 N.J. 507, 541, 545 A.2d 746 (1988); Appraisal of Real Estate, supra, at 320. We reject Kearney’s contention that the judge’s 1988 finding of 23% was already part of the 1991 assessment as reflected in the building and tank depreciation. Adjustments for economic obsolescence are distinct from, and additional to, any depreciation in the condition of the buildings. Ibid.
“[Wjhere the factual situation and questions presented are the same, prior adjudication of a similar nature may be controlling.” Pantasote Co. v. City of Passaic, 100 N.J. 408, 416, 495 A.2d 1308 (1985) (quoting Gottdiener v. Township of Roxbury, 2 N.J.Tax 206, 213 (Tax 1981)). BASF argues that the Tax Court is bound by its earlier decision in Badische I, where the judge accepted BASF’s 23% economic obsolescence factor and applied it to the buildings and structures. Badische I, supra, 11 N.J.Tax at 402. At that time, the judge found:
Plaintiffs expert concludes that the need ... to operate the plant around the clock even though the market for plaintiffs products caused diminution of production to 80% of the plant’s capacity, resulted in economic obsolescence. A credible method of quantifying such obsolescence is to ascertain the difference in plant depredation and the more rapid depreciation of the tanks on the theory that when the tanks are no longer useful the plant will shut down. That difference, according to plaintiffs expert is 23% (74% tank depreciation less 51% plant depreciation). While the court does not accept plaintiffs estimate of 51% depreciation on the buildings ... plaintiffs expert’s estimate of economic obsolescence is reasonable and is supported by sound analysis.
[Id at 394^95 (emphasis added).]
BASF’s expert testified that he used the same method in Badische II as in Badische I. His numbers appear to support this testimony. He estimated tank depreciation at 92% and plant depreciation at 60%. The judge found these numbers credible and probative of the improvement value. Badische II, supra, 14 N.J.Tax at 231. The difference between the two figures is 32%. *179Thus, application of the method found credible by the judge in Badische I to the Badische II numbers results in 32% economic obsolescence.
We remand to the Tax Court for reconsideration of the refusal to consider economic obsolescence as a valuation factor, when the same method of measuring economic obsolescence was accepted in Badische I and rejected in Badische II. Pantasote, supra, 100 N.J. at 416,495 A.2d 1308.
The tax judge also rejected BASF’s adjustment for environmental contamination. The decision to close the plant “triggered” the application of the Industrial Site Recovery Act (ISRA) (formerly designated the Environmental Cleanup Responsibility Act), N.J.S.A. 13:lK-6 to -14. Pursuant to N.J.AC. 7:26B-1.6, BASF filed a General Information Submission notifying the Department of Environmental Protection (DEP)1 of its intention to close. A Site Evaluation Submission (SES), filed in August 1990 pursuant to N.J.A.C. 7:26B-3.2, set forth a sampling plan to determine the extent of environmental contamination.
Phase I of the sampling plan, identifying areas of environmental concern, began in December 1990. Dale Webster, BASF’s manager of site remediation in charge of ensuring compliance with ISRA, began receiving test results approximately 90 days later. A map depicting the soil test results was available in June 1991. Based on the results, Webster estimated approximately 35,000 cubic yards of soil would require decontamination. In addition, a map depicting the results of ground water testing was available in June 1991.
Although the Phase I report was not complete until November 1991, Webster testified he was aware of the data used in the report as of October 1, 1991. Webster testified that as of October 1, 1991, the assessment date, he was reasonably certain of the *180nature and extent of cleanup BASF would incur. He said there was no need to further test the ground water, and further sod testing would not indicate less contamination than already estimated. Based on the test results, the technology existing as of June 1991, and his knowledge of other contaminated sites, Webster estimated it would cost $10 million to clean the soil and groundwater. A $10 million reserve was set up by BASF to cover these costs.
At the time of trial, there was no established mandatory cleanup level. Webster conceded his $10 million estimate was subject to change based on the cleanup level negotiated with the DEP. However, BASF’s testing level of 100 milligrams of contaminant per kilogram was then used by the DEP “as an action level, above which some type of action in the area of remediation would be strongly encouraged or promoted or ... required.” Stephen Roland, senior vice-president of O’Brien and Gere, the firm that prepared the SES, characterized this level as “an aggressive cleanup level,” which the DEP had accepted on other sites. He testified that his estimate of the cleanup cost as of October 1,1991 was at least $10 million.
As of the trial date, BASF had not submitted a cleanup plan to the DEP, nor had the DEP placed any restrictions on the use of the buildings. Webster estimated that the final cleanup plan would not be approved until 1994.
BASF also submitted a decommissioning plan to the DEP on January 7, 1991. Roland testified the purpose of the decommissioning process was to remove contamination from the above ground facilities so they “could either be sold, scrapped or removed from the site.” This included an asbestos removal program. The decommissioning program began in late fall 1991 and the asbestos was removed by the spring of 1992. The cost for asbestos removal was not specified.
Bids for decommissioning were solicited prior to September 1991. The contract was awarded to O.H. Materials, which submitted line item bids for the specific tasks totalling $5.4 million. The *181line item bids were not made part of the record. The actual decommissioning cost, which was not known until December 1992, was approximately $3 million. This was not broken down on the record into line item costs.
BASF’s expert adjusted the value of the subject property for environmental contamination using a discounted cash flow analysis at a 10% discount rate. He applied the cleanup and decommissioning expenditures over a ten-year period, where the $5,296,000 decommissioning cost was to be expended in the first year, and the $10,000,000 cleanup cost over the next nine years. He then assigned a reversionary future value to the cleaned property of $2,715,000. Applying these figures, the expert concluded the subject property had a value of -$8,925,000 as of October 1,1991. When the judge asked whether the negative value meant the seller would pay a buyer eight million dollars to take the property, the expert replied “that’s exactly what I’m saying.”
The Tax Court found that BASF did not meet its burden of proof in establishing cleanup costs and therefore did not adjust the value of the subject property for environmental contamination because as of the assessment date: (1) plaintiff did not submit a cleanup plan to the DEP; (2) the DEP had not approved a cleanup plan; (3) no contract beyond the $5.4 million decommissioning contract was awarded; (4) the projected cleanup cost of $10 million was merely an estimate subject to negotiation with the DEP; and (5) the $10 million estimate was lacking in probative value. Badische II, supra, 14 N.J.Tax at 230.
Inmar Assocs., Inc. v. Borough of Carlstadt, 112 N.J. 593, 549 A.2d 38 (1988), involved challenges to tax assessments by two separate owners of environmentally contaminated land. GAF Corporation operated an asphalt siding plant that was still in use during the tax year. Id. at 596, 549 A.2d 38. Inmar Associates, Inc. owned a 5.9 acre tract of land in the Hackensack Meadow-lands on which sixty-seven abandoned chemical storage tanks were located. Id. at 598, 549 A.2d 38. The Inmar property had been placed on the federal Superiund list pursuant to the Compre*182hensive Environmental Response, Compensation, and Liability Act (CERCLA), 42 U.S.C. §§ 9601 to 9675. Id. at 598, 549 A.2d 38.
As to the GAF property, the Court affirmed the Tax Court’s finding that it was “unable to quantify the effect that compliance with ECRA requirements would have had on the market value of the property on the assessing date” due to insufficient proofs. Id. at 597, 609, 549 A.2d 38. At trial, GAF estimated cleanup costs of at least $450,000; however, no sampling study had taken place and no cleanup plan was approved by the DEP. Id. at 597, 549 A.2d 38.
However, the Court remanded the Inmar case, rejecting the Tax Court’s decision not to make an adjustment because “no firm or fixed obligation to do restoration work had been incurred.” Id. at 599, 610, 549 A.2d 38. The Court found Inmar presented sufficient proofs of cleanup costs, even though no cleanup had taken place, nor had the DEP ordered Inmar to clean the property as of the assessment date. Id. at 598, 549 A.2d 38.
It appears the Court distinguished between the GAF and Inmar properties depending upon whether they were in use as of the assessment date:
In the [GAF] case, even though ECRA might have prevented sale of the property on the assessment date, the property had a distinct “value in use” to the owner so long as the owner continued to operate the facility. Hence, when the properly is in use, normal assessment techniques will remain an appropriate tool in the appraisal process.
[Id. at 607, 549 A.2d 38.]
Therefore, in accordance with “normal assessment techniques,” GAF was not entitled to a deduction for the cost to cleanup its property. Because the Inmar property was not in use as of the assessment date, different treatment was warranted:
Another suggestion, more appropriate when property is not in use, is the exercise of “more extensive investigation and ingenuity by appraisers in determining and considering factors that affect the value of special purpose properties.” ... Carlstadt’s appraisal expert was candid to recognize that the contamination affected the value of the properly. He suggested that the cost to cure the contaminated property could be treated as a capital, improvement, which can be *183depreciated over the beneficial life of the property. Although his approach was not followed by the Tax Court ... it contains the seeds of useful doctrine.
[Ibid, (citation omitted).]
Here, the Tax Court found the subject property “indistinguishable from the GAF case.” Badische II, supra, 14 N.J.Tax at 229. We disagree. Unlike the GAF property, the BASF property was not in use as of the assessment date. The Tax Court therefore incorrectly relied upon the GAF case in determining whether BASF submitted sufficient proof of the cleanup costs. Because the BASF property was not in use, the principles applied to the Inmar property are controlling.
As of the assessment date, BASF had submitted a detailed SES to the DEP documenting the environmental history of the subject property. A sampling plan setting forth the method of testing to determine the extent of contamination was submitted and approved by the DEP. Soil and groundwater testing were substantially complete, and some results were available enabling BASF to estimate the amount of contamination and the cost of cleanup at $10 million. BASF established a $10 million reserve to cover these costs. These proofs at least equal, if not surpass, those submitted by Inmar and found by the New Jersey Supreme Court sufficient to warrant remand. We therefore remand to the Tax Court to adjust the value of the subject property due to environmental contamination.
BASF argues the adjustment should reflect the price a prospective purchaser would pay for it in its contaminated condition. In Inmar, the Court found this approach difficult to apply to environmentally contaminated property, because “environmental regulations do not permit the sale of environmentally damaged land.” Inmar, supra, 112 N.J. at 603, 549 A.2d 38.
The optional cost-shifting that each example contemplates assumes a regulatory neutrality that does not exist in this context. The seller cannot avoid the cost of cleanup, but cost is not invariably equated with value.
[Id. at 602, 549 A.2d 38.]
BASF’s expert estimated the value of the subject property at negative $8,926,695. He arrived at this startling conclusion by *184projecting the total cleanup expenditures of $15,296,000 over ten years. This method was clearly incorrect. In Inmar, swpra, the Supreme Court explicitly stated that costs to comply with ISRA should not be directly deducted from the value of the land.
One thing is certain: the methodology for resolving the question is not simply to deduct the cost of the cleanup from a putative value of the property.
On the other hand, if the effect of these federal and state regulatory programs is to produce the market consequence of driving down the value of commercial property potentially subject to cleanup costs, the effect of those market forces cannot be ignored in the assessment process simply because it would be counter to the environmental poliey. Rather, the question that remains to be tested is whether a strong environmental cleanup poliey will drive real estate values up or down.
[Id. at 605, 606, 549 A.2d 38.]
Nonetheless, the Court did not formulate any particular method for determining the effect of cleanup costs upon value, preferring to “leave to the competence of the appraisal community the sound measure of that adjustment.” Id. at 608, 549 A.2d 38.
We are frank to recognize the difficulty of evaluating such market data, but we have recently reaffirmed the unique capability and responsibility of the Tax Court to exercise its power, in circumstances where the presumption of validity of the local assessment does not apply, to use the information available to it to make an independent determination of value.
[Id. at 609, 549 A.2d 38.]
Accordingly, we defer to the Tax Court’s specialized knowledge and expertise and remand for the formulation of a proper method for determining the effect of cleanup costs on the value of environmentally contaminated land.
We affirm the Tax Court judge’s finding that the $5.3 million estimate for decommissioning costs were not properly deductible from the value of the subject property. Francis Calabrese, an engineer at BASF in charge of overseeing the decommissioning process, testified the reason for closure was because the plant had been losing money every year since 1979 and because it had to make two major environmental investigations. We reject the proposition that the cost to decommission a plant *185voluntarily shut, down is deductible from the property’s market value.
We remand for supplemental proofs on the question of whether BASF should be entitled to a reduction for asbestos removal costs. University Plaza Realty Corp. v. City of Hackensack, 264 N. J.Super. 353, 358, 624 A.2d 1000 (App.Div.), certif. denied, 134 N.J. 481, 634 A.2d 527 (1993) permits a court to deduct such costs, in some cases dollar-for-dollar. The Tax Court found BASF failed to segregate the cost of asbestos removal from its decommissioning costs and therefore failed to meet its burden of proof. On remand, BASF shall be permitted to segregate the cost of asbestos removal from its decommissioning costs.
Affirmed in part, reversed in part. Remanded to the Tax Court for (1) further findings and recalculation, if necessary, of the per acre valuation of the property; (2) reconsideration of the refusal to consider economic obsolescence and environmental contamination adjustments as valuation factors; and (3) taking of proofs as to the allocation of costs of asbestos removal. The judge may, of course, take such additional testimony from either party as he deems appropriate. We do not retain jurisdiction.
The DEP was redesignated the Department of Environmental Protection and Energy (DEPE) on June 20, 1991, and on July 4, 1994 was again redesignated at the DEP. N.J.S.A. 13:1D-1.