Initially, this litigation began with the issue of whether a certain number of improvements were to be characterized and assessed as realty or personalty. In Chevron U.S.A., Inc. v. Perth Amboy, 9 N.J.Tax 205 (Tax Ct.1987) (Chevron I), this court held that the improvements, identified as process units, machinery, related support facilities, instrumentation, piping, and storage tanks, located on the site known as the Chevron-Perth Amboy refinery were “ordinarily intended to be affixed permanently to real property” pursuant to N.J.S.A. 54:4-1a.(3), and thus, constituted real property to be assessed and taxed as such.
In a subsequent determination, reported as Chevron U.S.A., Inc. v. Perth Amboy, 9 N.J.Tax 571 (Tax Ct.1988) (Chevron II), this court further held that certain of Chevron’s pollution control equipment was to be considered as part of the total assessable value of Chevron’s real property. Since Chevron’s counsel, however, was under the mistaken impression, permitted by Perth Amboy’s assessment records, that those facilities were tax-exempt, this court allowed Chevron the opportunity to supplement the existing record with appraisal testimony as to the value of such pollution control facilities.
The supplementary hearing was conducted and now remaining for determination by this court is the question of whether the real property at the Chevron site was, pursuant to N.J.S.A. 54:4-23,1 properly assessed for local property tax purposes for *117the tax years of 1984 and 1985, and if not, what the appropriate assessments should be.
Plaintiff, Chevron U.S.A., Inc., (Chevron) seeks a reduction in its local property tax assessments for the tax years of 1984 and 1985, while defendant, Perth Amboy, maintains, through its proofs and the self-executing operation of N.J.S.A. 54:51A-6 (chapter 123), that the assessments should be increased for 1984 and sustained for 1985.
The subject of this proceeding involves 20 separately assessed parcels in Perth Amboy which in combination form the 340-acre site known as the Chevron-Perth Amboy refinery. This refinery property was assessed in the aggregate for the tax years in question as follows:
Land $11,364,700
Improvements 67,725,700
Total $79,090,400
At the outset of the valuation hearing, the parties stipulated that, for the tax years of 1984 and 1985, the appropriate assessment ratios would be in accordance with the ratios set forth by the Director of the Division of Taxation pursuant to chapter 123, N.J.S.A. 54:51A-6, 54:l-35a. The average ratio certified by the Director for Perth Amboy for tax year 1984 was 59.60% and the common-level range, also certified, had an upper limit of 68.54% and a lower limit of 50.66%. With regard to tax year 1985 the average ratio for Perth Amboy was 49.70%, while the common-level range had an upper limit of 57.16% and a lower limit of 42.24%.
Chevron contends that the fair market value of the subject property, in the aggregate, as of the assessment dates, of October 1, 1983 for tax year 1984 and October 1, 1984 for tax year 1985, was approximately $96,365,100.2 It was further Chevron’s position that, with the application of the appropriate *118ratios, the fair assessable value of its property was $57,433,600 (rounded) for 1984 and $47,893,500 (rounded) for 1985.
Perth Amboy contends that the fair market value of the Chevron refinery as of the respective assessment dates was $168,375,000. This value would automatically require an application of the chapter 123 average ratio for 1984 and would produce an increased aggregate assessment for that year in the amount of $100,351,500. Because, however, of the decrease in Perth Amboy’s average ratio for 1985, the asserted value of $168,375,000 would produce an assessment-value ratio within the common-level range for that year, and thus, require this court to sustain the existing aggregate assessment for 1985.
I.
Description.
As part of the initial findings of fact in Chevron I, supra, this court found that:
Prior to June 1,1983, the Perth Amboy refinery was a fully integrated crude oil refinery with a rated capacity of 168,000 barrels per day. It processed crude oils ranging from light, low sulfur to heavy, high sulfur types into finished products that ran the gamut from light products such as gasoline to heavy products such as asphalt. The process units essential to the manufacture of these petroleum products at the Chevron site consisted of three crude distillation units (nos. 3, 4 and 5), the asphalt processing plant, the asphalt air blowing plant, two catalytic reformers (nos. 2 and 3 designated as rheniformers by Chevron), a distillate/vacuum gas oil desulfurizer (VGO Isomax), the Houdry catalytic cracking unit, a hexane plant, a middle distillate hydrotreater, two sour gas treating plants (nos. 1 and 2 amine plants), sulfur recovery units and tail gas plants (nos. 3 and 4 SRU), and a light products terminal. Other support facilities at the Perth Amboy refinery included crude oil and product storage tankage and facilities, off-site piping and transfer pumps, barge and ship loading and unloading facilities, truck loading facilities, steam and electric power generation and distribution systems, wastewater collection and treatment equipment, seawater supply and distribution for cooling purposes, offices, shops, warehouses, control rooms and other related facilities.
Since June 1, 1983, however, Chevron has become a 60,000-barrel per day refinery of heavy crudes producing one finished product, namely, asphalt. The only process units being run since June 1983 are the no. 5 crude unit, the asphalt terminal (asphalt plant) and a light products terminal. The light products terminal was closed on December 31, 1985. [9 N.J.Tax at 209-210]
*119As further noted in Chevron I, Chevron or its predecessor originally acquired the land in the mid-1940’s and early 1950’s from the Barber Asphalt Company and built most of the existing plant facilities during that time period. The improvements then consisted of a crude oil refinery including a number of buildings, process units, related support facilities, instrumentation, piping, and storage tanks. In 1975-1976 a major modernization program was undertaken and items such as the crude distillation unit no. 5, the no. 3 rheniformer and the distillate/vacuum gas oil desulfurizer (VGO Isomax) were constructed. In 1977 an asphalt air blowing facility was added, and in 1981, the last major construction took place when Chevron added a sulfur recovery facility, a sulfur gas unit and an SO2 sulfur recovery unit.
As of both assessment dates Chevron’s appraisal expert indicated that there were 593 buildings of various sizes which in the aggregate comprised 366,383 square feet of space. Due to the reduction in Chevron’s operations, from a 168,000-barrel per day fully integrated crude oil refinery to a 60,000-barrel per day asphalt plant, however, a number of these buildings were only partially used, while others were not utilized at all.
The Chevron refinery also contains 125 oil storage tanks, many of which are between 100,000- and 200,000-barrel (42 gallons to a barrel) capacity. These tanks, having a total capacity of 7,988,000 barrels, also, were not fully utilized by Chevron as of the assessment dates. As a matter of fact, 54 of the tanks were used as part of the refinery asphalt operation, while 14 were used for terminaling operations (bringing in gasoline refined elsewhere and selling it). Additionally, 57 of these tanks were not used at all. As of December 31, 1985 the *120terminaling operation ceased, supposedly because it was uneconomical for Chevron to continue it.
The site improvements at the refinery generally consist of 1,376,000 square feet of paving, 42,700 lineal feet of chain-link fencing, 59,400 line'ál feet of dikes, 8,800 lineal feet of railroad track, below-ground piping, sewer systems and three piers or docks.
II. Valuation.
Here, both parties are in agreement that a cost approach methodology would be most appropriate under the facts of this case.4 There is a fundamental dispute, however, as to whether the value of the improvements should be measured by a replacement cost procedure as recommended by Chevron’s appraisal expert or a replacement cost method as set forth by Perth Amboy’s experts. Before a determination is made in that regard, however, the value of the land must be considered.
A.
Land Value.
The first step in any cost approach procedure is the determination of land value. The most relevant method to determine this value, employed by the appraisers in this case, is a straightforward direct sales comparison approach. American *121Institute of Real Estate Appraisers, The Appraisal of Real Estate (9 ed. 1987) at 801.
With this method, sales of similar parcels of land are analyzed, compared and adjusted to provide a value indication for the land being appraised. [Ibid.}
The amount of land to be valued at the Chevron refinery site is approximately 340 acres as of the assessment dates for each of the tax years of 1984 and 1985. Chevron’s land valuation expert, Jerome J. Gall, valued the land at approximately $15,000 an acre for both tax years, producing a total land value of $5,129,455, while Perth Amboy’s expert, Michael Hiller, estimated the total land value to be $25,500,000 or approximately $75,000 an acre.
Gall reviewed and considered ten sales of allegedly comparable properties but admitted that he relied heavily upon only five of those sales. The five sales upon which he relied primarily occurred during the period of 1976 to 1981 and produced an unadjusted sales price range of $14,115 an acre to $15,675 an acre. He was of the opinion that adjustments to reflect the differences between the subject property and the comparable sales properties for time, location and size were unnecessary because the sales prices of the comparable properties demonstrated a relatively narrow price range regardless of the date of sale, location, or size, of the property sold. Therefore, the unadjusted sales price range of $14,115 an acre to $15,675 an acre remained the same after his consideration of adjustments.
Gall selected $15,000 an acre as the value of subject for both tax years at issue and generally assigned this value to the approximately 340-acre refinery site except for four minor exceptions which required an adjustment to his overall $15,000 an acre value. -These exceptions, which in the aggregate comprised 3.87 acres, consisted of two parcels of 25,000 square feet valued at $25,000 and $20,000 respectively, a third parcel, one acre in size, valued at $20,000 and a fourth parcel of six-tenths of one acre valued at $15,000.
Gall did not estimate the value of Chevron’s riparian rights, and thus, Chevron conceded the validity of the riparian land *122assessments.5 His land value in the aggregate for both tax years was $5,129,455 (336.63 acres at $15,000 an acre and 3.37 acres valued at a total of $80,000).
Perth Amboy’s expert, Hiller, relied upon 12 sales of allegedly comparable property to support his value estimate of $75,000 an acre for both tax years in dispute. Only five of the sale properties were located in Perth Amboy and those five had an unadjusted sales price range of $24,584 an acre, for a 1.57-acre parcel, to $90,900 an acre, for a 1.1-acre parcel. The sale properties located outside of Perth Amboy produced an unadjusted sales price range of $20,900, for a 187.93-acre parcel, to $232,957 an acre, for a 24.79-acre parcel.
After a thorough review of the record in this case, and for reasons to be hereinafter stated, I find by a fair preponderance of the believable evidence that Chevron’s expert, Gall, presented a sounder and more persuasive analysis, involving more convincing market data, than did Hiller, and therefore, adopt Gall’s analysis with one exception with regard to his lack of a time adjustment. I find that land values, based on two sales submitted by Hiller, were appreciating at least at the rate of approximately 3% a year, and thus, with that one adjustment, I accept the values estimated by Gall. The reason that necessitates a time adjustment is the fact that Gall did not present any sales after August 1981, and therefore, offered no market data to support his lack of a time adjustment for the two-year and three-year periods between his latest sale and the respective assessment dates in this case.
*123My analysis of the sales of the comparable properties offered by Gall demonstrates that, of the five sales upon which he primarily relied, two were sales of not only land but improvements in one case and improvements and machinery and equipment in the other. In each instance, however, without thorough research of the circumstances of the sales, he relied upon the allocations made by the parties to the transactions as to the price paid for the land.
Clearly, the allocation made by the parties as reflected in a deed or closing statement cannot be ignored. It was significant to the parties for some reason. On the basis of the record in this case, however, I cannot ascertain what the purpose of the allocation was or whether a different allocation could have been made. It could have an arbitrary allocation not related to fair market value, but rather related to a number of nonvalue factors, primarily federal tax consequences. See, in this regard, the discussion of nonmarket “conditions of sale” in The Appraisal of Real Estate, supra at 322.
No competent proof was submitted with respect to the value of the nonland portions involved in each of the two improved sales. Furthermore, no competent proof was offered with respect to the intention of the parties to either of the improved sale transactions. Thus, as to those improved sales, I am not satisfied, by a fair preponderance of the evidence, that the allocated price represents a negotiated price solely for the land, motivated by the real property value considerations of the parties. Cf. N.J.A.C. 18:12-1.1(a)23 (sales of industrial real property which include machinery and equipment are not usable in an equalization context when the value of the items are indeterminable).
This leaves remaining for consideration three of the five sales to which Gall gave predominant weight in his analysis. These were: (1) a sale by Asarco, Inc. to Neuberne H. Brown, Jr. in August 1979 of 82.18 acres of vacant industrial land for $14,115 an acre, (2) a sale by Lehigh Valley Railroad to Neuberne H. Brown, Jr. in April 1976 of 23.779 acres of vacant industrial *124land for $14,800 an acre and (3) a sale by N.L. Industries to Custom Distribution Services, Inc. (a corporation controlled by Neuberne H. Brown, Jr.) in August 1981 of 4.2 acres of vacant industrial land for $15,160 an acre. These three sale properties, occurring between April 1976 and August 1981, located in close proximity to the subject refinery site, clearly support Gall’s value estimate. Although each of the comparable properties was located in an M2 heavy industrial zone, while the subject is primarily located in an M3 heavy industrial zone, Perth Amboy was unable to demonstrate any significant value difference between these heavy industrial zones.
On cross-examination of Gall it was noted that the M3 zone permits storage tanks of 250,000 barrel capacity by means of a special exception application to the zoning board of adjustment, while the M2 zone special exception procedure only permits tanks of 3,000 barrel capacity. Gall, however, explained, without contradiction by Perth Amboy, that the zoning board has liberally granted variances for large storage tanks in the M2 zone.
Perth Amboy also attempted to demonstrate, on cross-examination of Gall, that perhaps some premium in value should be given to the Chevron refinery land, because it has all of the necessary governmental approvals to operate a refinery, and that with today’s environmental concerns such approvals may not be readily forthcoming. Perth Amboy’s land valuation expert offered no market data in support of this thesis and a review of the record fails to reveal any support for such a premium-in-value concept.
The last point Perth Amboy attempted to make, in the cross-examination of Gall, was with regard to a premium for assemblage or plottage. In general, Perth Amboy asserts that the subject 340-acre parcel enjoys a plottage or assemblage value because smaller parcels have less utility. Gall indicated that he considered the concept of assemblage value but found that the narrow price range of the sales upon which he relied demon*125strated that it was inapplicable. I agree. It is the marketplace, not appraisal theory, which reflects value.
In this regard, Perth Amboy’s expert, Hiller, insisted that his ninth sale, involving a 172.44-acre parcel in Mahwah, New Jersey, demonstrated market evidence of assemblage value. He indicated that the sale reflected a price of $118,800 an acre, at a time when one could have purchased two- or three-acre tracts in Mahwah for $100,000 an acre or less. I am not persuaded by this for several reasons. First, there is nothing in the record showing sales of two- or three-acre parcels in Mahwah for $100,000 an acre or less. Second, one month after the 172.44-acre sale took place, the purchaser sold 65 acres of the 172.44 acres for $198,000 an acre. Without more, the second sale shows that smaller parcels sell for greater unit or per-acre prices than do larger parcels.
In any event, these two sales, in close proximity in time, require a detailed inquiry by the appraiser as to the motivations of the parties. The record does not show that Hiller made any inquiry as to the conditions and circumstances of the Mahwah sales. It is conceivable that the purchaser of the 172.44-acre tract paid $118,800 an acre, only because, he had a purchaser for 65 acres of that larger parcel waiting to pay $193,000 an acre for the 65 acres. That does not reflect plottage or assemblage value.
Moreover, the location of the sale property and the subject is different. Mahwah property may experience plottage while Perth Amboy property may not. Only the market will indicate this. Lastly, the zoning, even though both are industrial, may also be different. Hiller was not fully aware of the zoning uses relative to the Mahwah property, thus any plottage conclusion as to the subject property is suspect.
In its main case, Perth Amboy attempted to demonstrate that Gall’s reliance on the sales to Neuberne H. Brown, Jr. was misplaced simply because those sales were to the same person or to an entity controlled by the same person. There is no explanation by Hiller, however, as to why the purchases by the *126same individual had any effect on either the price paid or the fair market value of the property purchased.
Also detracting from the credibility of Perth Amboy’s land appraisal expert was the fact that the relevant sales, heavily relied upon by Gall, were not included in Hiller’s land value analysis. Yet, nothing was offered by Hiller to explain the absence of these pertinent comparable sales.
Moreover, the market support offered by Hiller, specifically his 12 sales, was not nearly as persuasive as the sales relied upon by Gall. Despite all the appraisal principles and rules, an appraiser must ultimately defer to the marketplace. Values are derived from supportable market data. In this regard, the sales analysis offered by Hiller did not present convincing proof of his estimated land value.
I find that Hiller’s sales analysis is unreliable based on the following considerations. The first sale upon which he relied was a sale that included 400,000 square feet of buildings. Hiller used the entire sales price as attributable solely to the land even though the purchaser had subsequently leased 100,-000 square feet of the improvements. He did not appraise the buildings involved in the sale nor was he aware of the conditions of the sale relative to the allocation by the parties as to land and buildings.
The second sale involved a 1.57-acre parcel purchased by a contiguous landowner. A sale to an adjacent landowner may reflect atypical motivation or undue compulsion on the part of the purchaser to acquire an adjoining parcel in conjunction with an existing use. See The Appraisal of Real Estate, supra at 322. Yet, Hiller did not inquire into the circumstances of the sale to determine purchaser motivation. Additionally, Hiller, based on his plottage, theory, made no adjustment for size even though the comparable was only 1.57 acres.
Hiller’s third sale was a purchase by Neuberne H. Brown, Jr. of 3.864 acres of industrial land for $24,584 an acre. Gall considered this sale but gave it little weight because of the buyer’s atypical motivation in purchasing it. As Gall explained, *127the purchaser had a tenant, who had selected the site and was willing to negotiate with the purchaser, if the purchaser could obtain the land and construct a building for the tenant. Gall was of the opinion that the buyer was impelled to pay a price in excess of fair market value because of this factor. In contrast, Hiller had not inquired as to the conditions of the sale nor was he familiar with the motivations of the parties.
Hiller’s fourth sale involved a purchase for $100,000 by plaintiff, Chevron, of a 1.1-acre parcel which was paved, fenced and directly across the street from Chevron’s administration building. Hiller did not, however, make any inquiries as to whether Chevron was induced to purchase this property because it needed a parking lot for its employees.
Hiller made a computational error in his fifth sale by failing to include 1.6754 acres of riparian land in arriving at his unit value. The sale involved 6.1588 acres and not 4.483 acres as set forth by Hiller in his appraisal.
With respect to his sixth and seventh sales, the record reflects that Hiller failed to inspect either site. It is difficult to understand how an expert can make an appropriate comparison without viewing the property used for comparison. It is also interesting to note that, without a personal inspection by the appraiser, these comparable sites required the largest adjustment. As a rule of general application it is well settled in appraisal practice that the greater the dissimilarity between the subject and the comparable properties, the greater the potential for distortion.
Regarding his eighth and ninth sales in Mahwah, Hiller was not sure of the zoning designation other than the fact that these were industrial parcels. He was also not aware of whether the sales were contingent on the obtaining of zoning variances for the construction of offices and/or a hotel. Additionally, Hiller did not demonstrate that these sale properties were competitive with the subject, i.e., that the buyer of the comparable sale properties would have seriously considered the subject property in his plans as a suitable substitute. Hiller *128also admitted he did not know the zoning of the property which was the subject of his tenth sale. How Hiller could make appropriate adjustments in arriving at his value estimates without this information escapes this court.
The subject of his eleventh sale was zoned ROL (research, office and laboratory). Without a detailed comparison of the uses permitted in an ROL zone, which was absent in this case, with an M3 heavy industrial zone, I fail to see how an appropriate comparison can be made.
Hiller was not sure of the zoning designation for the twelfth sale property except to note that it was an industrial location. It cannot be overemphasized that “[zjoning is often the most basic criterion in selecting comparables.” The Appraisal of Real Estate, supra at 302. If comparables are not zoned the same as the subject, sales available from similar zoning categories can be used if appropriate adjustments are made. Ibid. However, detailed knowledge of zoning uses is necessary in order to make appropriate adjustments.
Lastly, there was one additional deficiency noted in Hiller’s land value analysis. He utilized a multiplication or factoring system of adjustment that is no longer included in modern appraisal texts, because it is generally not currently considered a valid adjustment technique. See Eaton, Real Estate Valuation in Litigation (1982) at 153-154; see also Boyce and Kinnard, Appraising Real Property (1984) at 196-197 (“Cumulative percentages are valid only if a causal interrelationship is shown to exist and this is not often the case”) and Ring, The Valuation of Real Estate (2 ed. 1970) at 140-141.
Therefore, for the reasons previously expressed, I accept the land values as set forth by Chevron’s expert, Gall, with an adjustment of 3% a year for time. This time adjustment is demonstrated by Hiller’s second and fifth sales in Perth Amboy which reflect a rate of progression of approximately 3% a year.6
*129Thus, the land value, with exceptions hereinafter noted, is determined to be $15,900 ($15,000 X 1.06 = $15,900) an acre for 1984 and $16,350 ($15,000 x 1.09 = $16,350) an acre for 1985. This is based on the fact that Gall’s latest comparable sale (August 1981) was approximately two years before the first assessment date and three years before the second assessment date involved in this matter. This produces an aggregate value of $5,352,400 (rounded) for 1984 ($15,900 x 336.63 acres = $5,352,400 rounded) and $5,503,900 for 1985 ($16,350 X 336.63 acres = $5,503,900).
The same rate of increase will be applied to the land values estimated by Gall relative to Block 339, Lot 1A, Block 460, Lots 68-77, Block 460, Lots 93-102 and Block 469, Lot 2, which comprise 3.37 acres. In the aggregate Gall valued these four parcels at $80,000. Therefore, I determine the aggregate value for these parcels to be $84,800 for 1984 ($80,000 X 1.06 = $84,800) and $87,200 for 1985 ($80,000 X 1.09 = $87,200).
Since neither party offered any proofs with respect to the value of riparian rights or lands, I have accepted the aggregate assessment for those rights of $185,500, adjusted by the appropriate chapter 123 ratio, as sufficient evidence of fair market value. See n. 5 supra. Those aggregate values are $311,200 for 1984 and $373,200 for 1985.
B.
Improvement Value.
(1) Chevron’s Cost Approach.
The value of the improvements at the Chevron refinery must be added to the land value to complete the replacement *130cost approach. The experts for both Chevron and Perth Amboy developed replacement cost improvement values, but by completely different methodologies. The results were dramatically different because each proceeded upon a different premise.
Chevron’s expert, Dennis C. Neilson, proceeded upon the proposition that the maximum value of the improvements at the Chevron refinery would be their value for the use that existed at the assessment dates and, as well as he could forecast, would exist in the reasonably foreseeable future.
It was Neilson’s opinion that any prospective prudent purchaser of the Chevron refinery would be interested in buying the refinery only for what he conceived to be its highest and best use: as an asphalt plant with a capacity of 60,000 barrels per day of crude oil which would produce 33,000 barrels per day of asphalt. He assumed this because it was dictated by economic trends in the petroleum industry and was evidenced by the manner in which Chevron had been operating since June 1983.
To begin with, Neilson examined the property, as he explained, as would an informed rational purchaser. He assumed that the subject property was vacant land and that a prudent purchaser, seeking a continuation of the existing business enterprise or use, i.e., a 60,000-barrel per day asphalt facility, would hypothetically construct a modern facility that would have the same desirability and utility. Thus, a prudent investor, if he were to consider purchasing the Chevron facility would pay no more for the subject than he would for an equally satisfactory modern substitute with the same utility.
The replacement cost of this substitute facility of equal utility would thus set the upper limit of value7 because the *131hypothetical improvements would be in a new and unused condition. Once having estimated replacement cost, Neilson indicated that the next step was to estimate and deduct accrued depreciation. His depreciation estimates involved primarily physical deterioration and were essentially a matter of judgment based on his experience, the observed condition of the improvements, an age-life technique8 and depreciation tables from cost manuals, where appropriate. Neilson made no allowance for economic or external obsolescence apparently because such obsolescence was reflected in his selection, as dictated by external considerations, of the subject property’s highest and best use, i.e., a 60,000-barrel per day asphalt plant and not a fully integrated 168,000-barrel per day oil refinery.
Following the replacement-cost, equal-utility concept throughout his appraisal estimate, Neilson considered the improvements in five segments: (1) the buildings, (2) the site improvements, (3) the storage tanks, (4) the machinery and equipment and (5) the pollution control facilities.9
With respect to the 59 buildings, Chevron’s appraiser, using a nationally computerized valuation procedure called the “Boeckh Interactive System,” calculated the cost of replacement for each building that would be required in a replacement facility.
*132In doing so, a number of adjustments were made. One was to eliminate a measure of functional obsolescence due to excess construction by hypothetically replacing a number of smaller buildings with one building with the same total floor area which had the effect of a cost saving due to economy of scale. Second, he considered the fact that, since Chevron was not operating at full capacity, there were a number of excess buildings and space. With respect to those buildings that were fully in use as of the assessment dates, he calculated replacement costs of modern buildings of equal space and utility.
There were, however, a number of buildings that were being used only partially and many not at all. In regard to those buildings that were partially used, he determined the portion in use, and necessary for the present business enterprise, and estimated the cost of modern replacement buildings that would accommodate the same degree of use as the existing buildings. For example, in the 127,500 square-foot shops building, only 42,000 square feet of space was used and needed for Chevron’s existing asphalt operation. Hence, he hypothetically replaced the shops building with a modern 42,000 square-foot building to provide the same utility.
The balance of the space of the existing building was assigned no value because, in his opinion, neither Chevron nor any other refinery owner or major operator would lease the additional space in a building that they would be partially using themselves. Neilson indicated that this procedure was dictated by the theory of substitution.10 That is, in valuing a building *133by the replacement cost approach one eliminates excess construction by hypothetically replacing the existing facility with one of equal utility but without the superadequacies.
Relative to those buildings that were not in use and not needed for the Chevron asphalt plant, Neilson took the position that there was the possibility that these had a rental potential and thus, he ascribed a nominal value of approximately $2 a square foot or less. This nominal value was assigned to all of the vacant buildings even though they may have been in the middle of the Chevron complex and, as a practical matter, had little potential for leasing to another user.
After estimating the cost of replacing the existing buildings with modern equivalents, he then estimated and deducted the physical depreciation of the existing buildings on the basis previously noted. This produced a total estimated fair market value for all the buildings of $2,531,000.
Next, Neilson carried the replacement concept to his value estimate of the site improvements. For example, there is presently 1,376,000 square feet of paving at the Chevron site but only 40% or 555,000 square feet of paving (termed “operating paving” by Neilson)11 would be necessary for the present asphalt plant operation. The cost of replacement for the operating paving was based on the 1984 R.S. Means construction cost data manual at .85<p a square foot and depreciation was then deducted based again on observed condition and an age-life concept. For the unnecessary or nonoperating paving, i.e., the paving that would not be replaced in a substitute facility, *134Neilson attributed a value of .10$ a square foot or a cost new of .85$ depreciated approximately 90%.
Following the same procedure, he determined that only 18,-500 lineal feet of fencing, out of the existing 42,700 lineal feet of fencing, would be necessary for the replacement asphalt facility. The cost of the necessary or operating fencing was derived again from the R.S. Means cost manual and depreciated based on observed condition and the age-life formula. The unnecessary or nonoperating fencing was given a nominal value of one-tenth of the cost of the operating fencing, or stated differently, depreciation was deducted at a rate of 90% from cost new. In like fashion Neilson concluded values for dikes and railroad trackage.
With regard to the piers that were present at the assessment dates, Neilson determined that only one modern -pier or dock was necessary to replace the utility of the piers in existence. This was due to the fact that one of the existing piers, in his opinion, was totally useless, and further, that only one pier would be necessary, in any hypothetical replacement of the existing facility, to provide equal utility for an asphalt facility. The replacement cost was based on a pier that was in conformance with today’s industrial standards for a typical dock. Specifically, it was patterned after a modern pier that Chevron had recently built at another refinery. His estimated fair market value of all of the site improvements was $1,742,000.
The next phase of Neilson’s appraisal was estimating the value of the 125 storage tanks at the refinery site. After determining the size, diameter, roof type, insulation and capacity of each tank, he determined the utility a particular tank had, as of the assessment dates, and in the foreseeable future. In this regard the tanks were divided into three categories: (1) in use for the asphalt operation, (2) in use for Chevron’s terminaling operation (which ended December 31, 1985) and (3) not in use or out-of-service or, as Neilson characterized this last class, not feasible for use for the refinery or for terminaling operations.
*135He estimated the replacement cost, which in this case equaled the reproduction cost, of each tank in use based on the Richardson Engineering construction cost manual, the Kenneth Guthrie pricing manual as a check, along with other pricing guides that were used for other refineries. The depreciation deducted from cost new was based on the observed condition of the tanks, a review of Chevron's maintenance records, discussion with Chevron’s refinery employees and employment of the age-life concept.
Those storage tanks that were in the out-of-service category were priced at .10$ on the dollar or, as with the nonoperating site improvements, were depreciated at a rate of 90%. The reason that Neilson assigned some nominal value to the unnecessary tanks was based on the fact that the tanks were at the refinery site and there was a possibility that one or more tanks might be periodically pressed into service. He could not determine which, therefore, he assigned a nominal value or potential value of the tanks, in place, to all out-of-service storage tanks. On cross-examination, Neilson noted, without contradiction by Perth Amboy, that the nominal value he ascribed to the out-of-service tanks would exceed the value of the tanks if Chevron were to sell them to another user who would have to dismantle, move and reconstruct the tanks at another location. His estimated fair market value for all tanks and tank foundations was $19,622,000.
Next, Neilson considered the fair market value of the machinery and equipment at the refinery site. See Chevron I, supra, for a detailed description of a number of the enormous process units at the Chevron refinery. Neilson also divided the machinery and equipment into operating and nonoperating assets. Since the operating assets were considered to be special purpose in nature Neilson concluded that the cost approach was the most appropriate technique for valuing them.
The nonoperating assets, consisting of the machinery and equipment that was idle and had no use as of the assessment dates in the continued operation as a 60,000-barrel per day *136asphalt refinery, were divided into three categories: (1) those process units, such as the asphalt air blowing plant, that had a potential use value in the asphalt operation, and thus, would also be most appropriately valued by a cost approach, (2) those units, such as the no. 3 rheniformer and the VGO Isomax which, in Neilson’s opinion, would never be operated again at the Chevron refinery, but were relatively new and technologically current, and thus, could be sold to another user at a different location, and therefore, could be valued by a market data approach for their salvage value, and (3) those process units, most of which were built in the 1950’s, which were old, in very poor physical condition and were technologically outdated, and thus, could be valued by a market data approach solely for their scrap value.
The operating assets consisted of those units, such as the crude distillation unit no. 5, the asphalt plant and power plant, which would be necessary to operate Chevron’s asphalt facility. Neilson noted that the crude distillation unit, which was constructed in 1975, had a rated capacity of 80,000 barrels per day. The unit had been operating, however, at a reduced capacity of 60,000 barrels per day since 1982 and it was Neilson’s opinion that the unit would never be operated at full capacity again because of economic conditions. Therefore, his replacement cost was for a modern crude unit capable of processing 60,000 barrels per day of crude oil. The cost figures were derived from Chevron’s original costs which Neilson trended to derive costs new as of the assessment dates. Depreciation was then calculated based on a physical inspection and a consideration of an age-life formula.
The remaining operating units were either valued on the replacement cost of modem units of equal capacity and utility, or cost of reproduction “in kind” if the units were of modern utility and technology, based either on trended original costs or pricing manuals such as Richardson Engineering, Chemical Engineering or Kenneth Guthrie’s manual. These units were then depreciated based on observed physical condition in conjunction with Neilson’s effective-age, remaining-life relationship *137formulation. The operating assets were estimated to have an aggregate fair market value of $36,794,000.
With respect to the three categories of nonoperating assets, i.e., the machinery and equipment that was idle and had no use as of the assessment dates in the continued operation of Chevron’s 60,000-barrel per day asphalt facility, Neilson employed the replacement cost approach if the unit had potential use value in the existing asphalt operation. Specifically, he concluded that, as of the assessment dates, the asphalt air blowing plant12 had such potential, and therefore, he estimated its fair market value “as a complete operational facility installed in-place and ready to operate.” The plant was valued as if it were in the same category as the operating assets, and in the same fashion, at $5,912,000.
The second category of nonoperating assets were those process units, such as the no. 3 rheniformer and the VGO Isomax, even though of relatively new construction (1974-1975), which Neilson concluded would never be operated again at the Chevron-Perth Amboy refinery due to the economics of the oil industry. These units were thus valued on the basis of a market approach. That is, based on a hypothetical sale to another user with the buyer assuming the cost to dismantle, remove and relocate the units. Neilson's procedure to estimate this “salvage value” was to first determine cost new uninstalled and unconnected (based on trended original costs)13 After deducting for accrued depreciation, he applied a 50% salvage-value factor to his cost new less depreciation for the cost to remove and relocate. This factor for salvage value was from his consultations with Ventech, a group of companies that *138specializes in the removal and the reapplication of surplus refinery process units and equipment, see Chevron I, supra, and a salvage and scrap dealer by the name of Arnold Landisberg. Neilson estimated the fair market value or salvage value14 of these process units at $7,150,000.
The third category of nonoperating assets were those non-operating process units, most of which had been built during the 1950’s, that were old, of outdated technology and in very poor physical condition. Neilson concluded that based on the condition of these units, such as the Houdry catalytic cracking unit, nos. 3 and 4 crude distillation units and the middle distillate hydrotreater, and considering the state of the economy of the petroleum industry, that this category of nonoperating assets would never again be used, and therefore, had no salvage value, but rather the value of these units would be limited to their value as scrap. Again, he used a market approach and, based on his consultations with Ventech and Arnold Landisberg and a review of the costs reflected in Chemical Engineering and the Oil and Gas Journal, he concluded a scrap value of ,10<t on the dollar or an aggregate scrap value of $5,172,000.15
Lastly, at the supplemental hearing, Neilson testified as to his estimate of the value of the pollution control facilities that had not been previously appraised because it had been erroneously assumed that these assets were tax-exempt. See Chevron II, supra. Specifically, those facilities consist of sulfur recovery units nos. 1, 2, 3 and 4, the waste gas flares, the effluent treatment plant and two buildings associated with the sulfur recovery units.
In estimating the value of the pollution control facilities Neilson followed the same replacement cost methodology that *139he utilized at the initial valuation hearing. Since the two buildings were not in use and not necessary for Chevron’s asphalt operation, he assigned a nominal value of $2 a square foot on the theory that because they were at the refinery site they could possibly be pressed into service even though that was not likely.
He ascribed no value to sulfur recovery unit no. 1 because it was old, completely obsolete and in very poor condition. This was evidenced by the fact that Chevron had stopped using this unit in 1980 and simply allowed it to rust. He was of the further opinion that a scrap dealer would not pay for the unit, but rather would have to be paid to remove it.
With respect to sulfur recovery unit no. 2, Neilson estimated the value based on the operating and nonoperating portions. The operating portions were valued based on reproduction and installation costs derived from the Kenneth Guthrie manual and depreciation based on age and observed condition, while the nonoperating portions were determined to have scrap value only.
Sulfur recovery units nos. 3 and 4, although constructed in 1980 and still technologically current, were shut down by Chevron in June 1983, allegedly, for economic reasons. Neilson indicated that these units are not needed and will not be operated by Chevron in their current location. Since, however, the units are considered “state-of-the-art,” they have value for another user at another location, and therefore, Neilson ascribed salvage value much in the same fashion as he did with the no. 3 rheniformer and VGO Isomax units.16
*140Finally, Neilson estimated the value of the effluent treatment plant and the waste gas flares. He determined the capacity of these units as originally constructed and the capacity which would be required for Chevron’s present needs as an ongoing asphalt facility and then estimated the cost to replace only the presently required capacity. He then deducted depreciation based on experience, observed condition, discussions with industry personnel and an age-life formula to derive his estimate of fair market value. In the aggregate, Neilson was of the opinion that the fair market value of the pollution control equipment and related facilities was $12,001,400.
The overall estimated fair market value advanced by Neilson for all of the refinery improvements was $90,924,400.
(2) Perth Amboy’s Cost Approach.
The cost approach, as previously noted, was also employed by Perth Amboy. The fee appraisal division of Day & Zimmerman, Inc. was selected by Perth Amboy to perform a cost analysis because this firm had previously provided appraisals for Perth Amboy for use in its assessment of the Chevron refinery for the 1967 tax year and the major modernization project undertaken by Chevron in 1974-1975.
Martin Sigel, the manager of real estate fee appraisals for Day & Zimmerman, Inc. was assigned the responsibility for estimating the value of the buildings, storage tanks and miscellaneous land or site developments, while Richard M. Daniel, a chemical engineer, estimated the depreciated replacement cost of the process machinery and equipment. Sigel estimated the value of his portion of the Chevron refinery to be $35,306,000 and Daniel estimated the depreciated replacement cost of his portion to be $107,569,000 which in the aggregate produced an overall improvements value of $142,875,000.
In estimating the values of the respective portions, both experts worked under the constraints of the same premise. That is, that the valuation assignment was to estimate the value or depreciated replacement cost of the improvements at a 168,000-barrel per day fully integrated oil refinery. This prem*141ise was in direct conflict with the assumption upon which Neilson had predicated his replacement cost valuation.
Sigel utilized the Marshall Valuation Service, a nationally known building cost manual published by the Marshall and Swift Publication Company for his building values. He applied the manual’s calculator cost method for each building. This method essentially requires inspecting the building to be appraised and placing it in one of the building classifications set forth in the manual. The selected classification reveals, after necessary adjustments and mathematical computations, the approximate cost to replace17 the improvement as of a particular time.
In Sigel’s replacement procedure he estimated the cost to replace all of the space in each existing building whether it was fully used by Chevron in its present 60,000-barrel per day asphalt operation or not. He explained that each and every building at the refinery site was required for a 168,000-barrel per day fully integrated oil refinery operation and he was appraising a fully integrated oil refinery — not an asphalt plant.
After estimating the replacement cost new of each building improvement, the next step was to estimate accrued depreciation in all forms. Sigel based his allowances for physical deterioration and functional obsolescence on a review of the depreciation tables in the Marshall cost manual, observation of the existing facilities and his experience and judgment. He recognized the concept of functional obsolescence due to excess construction (though not to the extent advocated by Neilson)18 *142and applied a percentage of depreciation for that item whenever he felt it was applicable. He considered external or economic obsolescence, but concluded that none existed at the Chevron site, hence he made no deductions for this form of depreciation.
Relying on factual data supplied to him by Chevron and his exterior inspection, Sigel then estimated the values of the 125 storage tanks at the refinery site. Chicago Bridge & Iron Co., fabricators and installers of storage tanks, provided Sigel, „by means of telephone conservations, with price ranges for various types of tanks, which he adjusted depending on the size and type of tank at the Chevron site. Depreciation estimates for the various tanks were derived from telephone discussions with representatives of Chicago Bridge & Iron and depreciation tables, available in the Day & Zimmerman files, compiled from other appraisals. What is most important, however, is that he assumed that each and every tank was essential to Chevron’s operation — not as an asphalt refinery — but as a fully integrated 168,000-barrel per day oil refinery operation.
Finally, Sigel, apparently based on Day & Zimmerman’s library of cost compilations, estimated the depreciated replacement costs of the miscellaneous site developments such as yard fencing, paving, piers and bulkheads, again, on the basis that Chevron was conducting a fully integrated oil refinery operation and had a need for all of these site improvements. However, since Sigel was told by Chevron representatives that the railroad tracks were not in use, had not been used for years and that Chevron had no future use for the trackage he assigned a “scrap and salvage” value for that item.
The last portion of the Chevron improvements, and most significant in value, were the process units and associated facilities. As previously noted, Richard M. Daniel, a chemical engineer, with a good deal of experience in cost estimating of *143refinery process units, was assigned this valuation task. Daniel was not asked to estimate fair market value of the process equipment but was specifically instructed to estimate replacement costs on a depreciated basis of the Chevron process units.19 Daniel defined the function of replacement cost not as the cost of a replication in kind, but rather to estimate the cost of a process unit with the most current technology that would replace the capacity and function of an existing process unit. Therefore, Daniel’s assignment was predicated on estimating the cost to replace process units and related facilities that had the original capacity to function as a fully integrated 168,000-barrel per day oil refinery.
Daniel divided his replacement cost appraisal of Chevron machinery and equipment into two parts. The first involved those units and related facilities that were in existence in 1950 which Daniel referred to as “the old refinery” and the second considered the units and related facilities that were constructed in Chevron’s modernization program in 1974-1975 or constructed thereafter, as were sulfur removal units numbers 3 and 4. Daniel referred to this latter category as “the new refinery.”
With regard to the old refinery Daniel concluded that many of the units, such as the Houdry catalytic cracker, crude distillation units nos. 3 and 4, no. 2 rheniformer, an alkylation unit, hydrotreater, sulfur recovery unit no. I,20 etc., had exceeded their useful lives and thus, had value only for scrap pur*144poses. Some of the units, such as the CO combustion unit for the Houdry catalytic cracker, in his opinion, had value for continued use in another location. Therefore, he ascribed salvage value.
With regard to those facilities that he believed had value for continued use, in place, as part of the old refinery, such as the dehexanizer, the asphalt plant, electrical distribution system, yard piping and the power plant, he relied on the replacement values that he had estimated for Perth Amboy for its 1967 tax year in a previous valuation assignment. Using these previously determined replacement values, he applied additional depreciation, essentially physical, and then trended the depreciated value estimate to the current assessment dates with a cost conversion or inflation factor derived from Marshall Valuation Service.
Insofar as the new refinery was concerned, Daniel relied upon the actual costs given to him by Chevron when he performed a subsequent appraisal, at the request of Perth Amboy, relative to the modernization program Chevron undertook in 1974-1975. Utilizing the actual cost figures as his starting point for such items as the crude distillation unit no. 5, no. 3 rheniformer, VGO Isomax and the pollution control equipment, Daniel provided for depreciation, again, essentially physical deterioration, and then trended the depreciated historical cost to the current assessment dates with a cost conversion factor.
Lastly, Daniel followed essentially the same procedure in estimating the depreciated replacement costs of sulfur removal units 2, (built in 1975) and sulfur removal units 3 and 4 (built in 1980). His aggregate depreciated replacement costs for all of the process units and related facilities was $107,551,000, by far, the most significant value portion of Perth Amboy’s total value estimate of the Chevron refinery.
(3) Discussion and Determination of Improvement Values.
Although neither party has specifically framed the issue as a dispute over the highest and best use of the subject property, that concept is at the root of this controversy. Value *145estimates, after all, are based on use. It is the selected highest and best use of a property that provides the foundation on which an estimate of market value rests. The Appraisal of Real Estate, supra at 269-270.
Highest and best use has been defined as:
The reasonably probable and legal use of ... an improved property, which is physically possible, appropriately supported, financially feasible, and that results in the highest value. [Id. at 269]
As noted in The Appraisal of Real Estate, in order for a particular use to be considered the highest and best that use must be both financially feasible and capable of producing the highest return consistent with market risks. Id. at 274-277.
Throughout the testimony of Chevron’s appraiser, Neilson, it was evident that he believed that only Chevron’s present use met these criteria. As a consequence, he asserted that the maximum value of the improvements at the refinery site would be their value for use as an asphalt plant with a 60,000-barrel per day crude oil capacity which would produce 33,000 barrels per day of asphalt.
In contrast, defendant, Perth Amboy, asserted that the maximum value of the improvements (and thus, their highest and best use) would be best measured by viewing the property as a fully integrated 168,000-barrel per day oil refinery producing a host of oil products, ranging from light products such as gasoline and heating oil to heavy products such as asphalt, even though that use had been discontinued on June 1, 1983, four months prior to the first assessment date.
As can readily be seen, it is these different assumptions as to highest and best use that produce the dramatically different estimates of improvement values advanced by the valuation experts for Chevron and Perth Amboy. The valuation conclusion of each party is only as valid as the premise upon which it is based. If the selected use is not warranted by market conditions as of the appropriate assessment dates, the value conclusion based on that use is just as unwarranted.
*146The uncontroverted facts reveal that Chevron’s refinery, when fully operational had a crude oil capacity of 168,000-bar-rels per day of crude oil. Thus, in 1980 when Chevron was operating at full capacity it was producing gasoline, fuel oil, heating oil and asphalt. In June 1983 the refinery operations were reduced to a capacity of 60,000-barrels per day of crude oil producing approximately 33,000-barrels per day of asphalt.
Neilson, Chevron’s appraiser, asserts that this reduction in operations was due to economic conditions in the petroleum industry and, in part, to some obsolete process equipment located at the Chevron site, specifically the Houdry catalytic cracker and the crude distillation units nos. 3 and 4. Neilson was of the further opinion that, as of the assessment dates, based on oil industry market conditions, Chevron, or any other prudent refinery operator for that matter, would not resume full scale operations with the existing improvements, but would continue utilizing the subject solely as an asphalt plant.
There were a number of factors upon which he relied in arriving at his conclusion. These included the high cost of crude oil worldwide, increased conservation of energy, automobiles being made more fuel efficient, road construction has been curtailed because less tax has been collected from reduced gasoline sales, industrial energy users have switched to other sources of energy such as gas because it is less expensive, and the adoption of environmental regulations which virtually eliminate the use of lead additives in gasoline. Because of these factors, Neilson indicated that not only has the demand for oil products decreased, but there is now a worldwide glut of crude oil and an excess of petroleum refining capacity.
In turn, excess capacity has caused greater competition among the major oil companies and the marginally profitable refineries have been shut down because of the economics of the petroleum industry. Neilson noted that since 1981 more than 100 refineries have completely ceased operations.
All of these considerations, which existed in 1983, along with the fact that Chevron as of June 1, 1983, four months prior to *147the first assessment date in this case, reduced its operation to that which still exists today, led Neilson to the firm conclusion that a reasonably prudent purchaser or investor would only view the Chevron site as a 60,000-barrel per day asphalt plant and not a fully integrated 168,000-barrel per day oil refinery.
In contrast, Perth Amboy’s premise that Chevron or a hypothetical purchaser might reinstate a fully integrated 168,000-barrel per day refinery operation is without support in the record. One of Perth Amboy’s valuation experts, Sigel, admittedly not an expert on the economics of the petroleum industry in general and oil refineries in particular, testified that he perceived the Chevron refinery had the potential of going to a fully integrated refinery operation again, and therefore, should be the basis upon which it should be valued. Sigel, however, submitted nothing to support this perception.
To the contrary of Sigel’s unsupported assumption, the record clearly shows, and Perth Amboy has not disputed the fact, that even if the economic outlook of the oil industry changed, and it again became financially feasible for Chevron, or any other refinery operator, to return to full operations that could not be done without a substantial investment, at minimum, of approximately 80 to 100 million dollars.
This is due, in part, to the fact that the only operable crude distillation unit at the refinery site has a rated capacity of only 80,000 barrels per day. In order to return to 168,000-barrel per day capacity, another crude distillation unit would have to be installed at a cost of approximately 40 to 50 million dollars.
Moreover, since the existing Houdry catalytic cracking unit was indisputably nothing more than scrap, Chevron could not return to full production at 168,000-barrel per day capacity without a fluid catalytic cracking unit which represents another capital infusion of approximately 40 to 50 million dollars.
In this regard the record is barren of any proofs adduced by Perth Amboy as to whether the economics of the petroleum industry would ever justify such a substantial capital investment in order for Chevron, or any hypothetical purchaser, to *148resume a fully integrated oil refinery operation at the Chevron refinery site. In other words, there is nothing to show that the required substantial capital investment will provide greater profitability than Chevron’s present use. See The Appraisal of Real Estate, supra at 280-281.
Perth Amboy did, however, seek to justify its premise that the Chevron site should be valued as a fully integrated refinery on the basis that the price of crude oil had diminished since the pivotal assessment dates.21 This assumes that a fully informed prudent business decision to fully activate an oil refinery is simply tied to the fluctuations in the price of crude oil. Without more, I cannot arrive at such a conclusion. If that decision were as unsophisticated as suggested by Perth Amboy, the question arises as to why Chevron, a reasonably successful petroleum company, has not resumed full scale refining operations at the Chevron site in light of the price fluctuations of crude oil. Perth Amboy suggested no appropriate response nor did it provide a basis from which to conclude one.
In short, there is nothing in the record of this case that supports Perth Amboy’s thesis that, as of the assessment dates in question,22 the highest and best use of the subject property was a fully integrated 168,000-barrel per day oil refinery. Based on my review, I am persuaded by a fair preponderance of the evidence that, as of the respective assessment dates, the *149maximum value of the improvements would be reflected by the use, that not only existed at those dates,23 but that use, based on market conditions, that could have been reasonably foreseen at that time, i.e., a 60,000-barrel per day asphalt refinery. Cf. Transcontinental Gas Pipe Line Corp. v. Bernards Tp., 111 N.J. 507, 537, 541-543, 545 A.2d 746 (1988) (“Whatever method of determining replacement cost is utilized, the goal should be to arrive at a value which properly reflects the value of the pipeline that would be required to meet the current demand of consumers of natural gas, ... ”).
Aside from any question of highest and best use, Perth Amboy also argues, however, that Neilson’s replacement cost approach was incorrectly applied and is unrealistic. Perth Amboy points to Neilson’s value analysis of the storage tanks as an example of his “erroneous approach to fair market value.”
As set forth in its brief, Perth Amboy offers the following illustration:
On page 1 of Appendix C-l, [of Neilson’s appraisal report] there are listed tanks # 102, # 103 and # 104. Each of these three tanks was constructed in 1950, is a vertical tank and each has a cone roof. As further indicated in said schedule, the barrel capacity of each tank is almost identical. Likewise the schedule shows that the cost of reproduction of each tank is practically identical. Notwithstanding the cost similarity of the physical characteristics of all three tanks, and despite their identical age, Mr. Neilson has seen fit to appraise tanks # 102 and # 103 at $44,476.00 while tank # 104, practically an identical structure, has been valued by him at $177,491, almost four times as much. The reason assigned by this expert for the vast disparity in his opinion of fair market value between tank # 104 on the one hand and # 102 and # 103 *150on the other, is that # 102 and # 103 are out of service while # 104 is in refinery use for the storage of gas oil. To point up the incongruity of this position, it should be noted that Mr. Neilson readily admitted that if tanks # 102 and # 103 likewise were filled with refinery products, the true value of each of such tanks would necessarily change from $44,476 to $177,491. Inversely, by emptying tank # 104 and classifying it as “out of service”, would so sharply impact upon its market value as to reduce it from $177,491 to $44,476.
Perth Amboy has focused on, and emphasizes, the alleged absurdity in having two completely different values for two practically identical storage tanks. This result, however, is perfectly logical depending upon the premise upon which the valuations are based. Neilson has determined that the maximum value of the Chevron refinery improvements is based on its use as a 60,000-barrel per day refinery. Thus, value of all the improvements is in relation to that premise. See Transcontinental Gas Pipe Line Corp. v. Bernards Tp., supra at 541-543, 545 A.2d 746.
Neilson based his valuation of the storage tanks on what a reasonably prudent purchaser would pay for the existing utility in relation to highest and best use. A rational purchaser seeking an asphalt facility with a crude oil capacity of 60,000 barrels per day, needing only a designated number of storage tanks, will not pay anything extra for those tanks beyond what was necessary for the desired operation. Those extra tanks would be superadequacies or superfluities that would have no value to him unless he might need one or more on a periodic basis or he could sell them (salvage value) to another user at another site. Neilson considered these alternatives and reasonably concluded values with these considerations in mind.
The values he ascribed to the tanks were overall in relation to the need. Thus, allocated tank values were based on Chevron’s overall needs as of the assessment dates. If, as Perth Amboy suggests, one tank that is needed and in service is switched with another that is unneeded and out-of-service, the allocated values as to those particular tanks may change, but the overall value of both tanks remains the same. I find that Neilson’s value conclusions are derived from an appropriate appraisal methodology and reflect economic realty. Ibid.; see The Ap*151praisal of Real Estate, supra at 35-36, 352, 357-358, 393-395; Shenkel, Modern Real Estate Appraisal (1978) at 162; 1 Bonbright, op. cit., supra at 153, 156-157, 172-176; Eaton, op. cit., supra at 104-105; Boyce and Kinnard, op. cit., supra at 256-257; O’Flaherty, “Cost Approach to Value,” Friedman, Encyclopedia of Real Estate Appraising (3 ed. 1978) at 69-70; Hartman and Shapiro, “Depreciation: Incurable Functional Obsolescence and Sequence of Deductions,” The Appraisal Journal (July 1983) at 409.
Overall, I am of the opinion that Neilson’s replacement cost approach premised upon a 60,000-barrel per day asphalt operation was essentially sound and supported.24
In contrast, the valuation approach of Perth Amboy’s experts lacked the appropriate foundation because it was based on a selected highest and best use that was not demonstrated to be financially feasible and maximally profitable consistent with market risks.25
*152Accordingly, for the reasons set forth, I find the replacement cost methodology of Chevron’s expert, Neilson, produces a realistic and reliable indication of the value of the improvements at the Chevron refinery, and therefore, adopt his replacement cost value for the improvements of $90,924,400.
The value of the improvements when added to the previously determined land values produces total fair market values for the tax years in issue as follows:
1984 1985
336.63 acres at $15,900 an acre - $ 5,352,400 336.63 acres at $16,350 an acre - $ 5,503,900
3.37 acres 84,800 3.37 acres 87,200
Riparian lands 311,200 Riparian lands 373,200
Improvements - 90,924,400 Improvements - 90,924,400
1984 value - $96,672,800 1985 value - $96,888,700
The total values, as determined, produce a ratio of assessment to true value of 81.81% for 1984 ($79,090,400 -4- $96,672,-800 = 81.81%) and 81.63% for 1985 ($79,090,400 -4- $96,888,700 = 81.63%). It is clear that an application of the chapter 123 average ratio is in order, since the ratio of the assessed valuation of the subject property to its true value exceeds the upper limit (1984 — 68.54%, 1985 — 57.16%) of the common-level range for each tax year. Therefore, the assessable values, in the aggregate, shall be:
*1531984 1985
Land $ 3,426,000 Land $ 2,964,300
Improvements 54,191,000 Improvements 45,189,400
Total $57,617,000 Total $48,153,700
Since the existing assessments are set forth in 20 different parcels or line items and my conclusion is in the aggregate, the parties are directed, pursuant to R. 8:9-3, to submit agreed upon allocated assessments for the purpose of the entry of judgment. Such allocations shall be submitted within 15 days. If the parties are unable to agree, the provisions of R. 8:9-4 shall apply.
N.J.S.A. 54:4-23 provides the statutory criterion for the determination of value in a local property tax case and directs that value to be the price for which a parcel of real property would sell at a fair and bona fide sale on October 1 of the pretax year.
This sum is approximate because Chevron did not challenge the assessment of its riparian lands. See n. 5 infra.
There were actually 61 buildings. Two buildings were not considered by Chevron’s appraisal expert in his initial appraisal report and testimony because he assumed these were tax-exempt. Subsequent to this court's decision in Chevron II, supra, however, the value of these two buildings was estimated at a supplemental hearing and is considered later in this opinion.
Chevron's appraisal expert, Dennis C. Neilson, also made an analysis of the most recent sales of oil refineries which provided a value range that was not useful to determine a precise value, but could be used as a check of his cost approach value. Neilson did, however, employ a market approach to value those improvements or assets that he considered "nonoperating," ¿a, those assets that were not needed in Chevron’s existing asphalt operation. So did defendant’s valuation experts. Richard M. Daniel, Perth Amboy’s process unit valuation expert, employed a market data approach with regard to certain machinery and equipment whose only value was in salvage or scrap to be obtained in a market sale. Martin Sigel, in a similar fashion, estimated scrap or salvage value for the railroad trackage.
Hiller, Perth Amboy’s expert, did not provide any specific valuation estimate for the riparian rights. Therefore, it is assumed that the value of the riparian land rights are not in issue and this court will accept the assessed value of the riparian rights, adjusted by the chapter 123 ratio, as furnishing "sufficient and competent evidence" of the fair market value. See Glen Wall Associates v. Wall Tp., 99 N.J. 265, 272-274, 491 A.2d 1247 (1985). Since the riparian lands were assessed at $185,500 in the aggregate and the applicable chapter 123 ratios were 59.60% for 1984 and 49.70% for 1985, I will assume a fair market value for the riparian rights of $311,200 for 1984 ($185,500 59.60% = $311,200 rounded) and $373,200 for 1985 ($185,500 49.70% = $373,200 rounded).
This is the result when a correction is made in the unit price reflected by Hiller's fifth sale. As previously noted, Hiller assumed the sale involved 4.483 *129acres whereas 6.1588 acres was the correct figure. With the correction, the two sales reflect a change in unit value of $5,810 an acre ($53,580 — $47,770 = $5,810) over, roughly, a four-year period or approximately 3% a year ($5,810 ~ $47,770 = 12.16% 4- 4 = 3%). Although I questioned the usefulness of Hiller’s second sale as a reflection of value for the subject property because the buyer may have paid more than market value due to that buyer’s need for an adjacent parcel of land, this does not detract from its usefulness as an indication of whether real estate values were increasing.
An eminent commentator in the field of local property taxation has noted that this statement may be invalid if a prospective purchaser cannot secure a substitute without undue delay, and thus, property which is presently available may be worth more than the cost of a hypothetical substitute. Therefore, instead of setting the upper limit of value, a special premium for immediate availability may have to be added to the cost of replacement in order to *131determine value. See 1 Bonbright, Valuation of Property (1937) at 157. In this case, however, neither of the parties has addressed the issue and without factual underpinnings as to delay and the procedure for computation of a special value premium for undue delay, conclusions cannot be drawn.
The age-life technique is an appraisal method by which accrued depreciation is measured. It is “the ratio of a building’s effective age to its total economic life” which is then “applied to the current cost of the improvements to obtain a lump sum deduction for accrued depreciation.” The Appraisal of Real Estate, supra at 379.
As previously noted, Chevron did not originally consider its pollution control equipment as part of its assessable property at the initial valuation hearing. In Chevron II, supra, this court held that the value of the pollution control equipment had to be considered. At a supplemental hearing, Neilson testified as to his estimate of fair market value of that equipment.
Substitution is:
A valuation principle that states that a prudent purchaser would pay no more for real property than the cost of acquiring an equally desirable substitute on the open market. The Principle of Substitution presumes that the purchaser will consider the alternatives available and will act rationally or prudently on the basis of the information about those alternatives, and that reasonable time is available for the decision. Substitution may assume the form of the purchase of an existing property, with the same utility, or of acquiring an investment which will produce an income stream of the same size with the same risk as that involved in the property *133in question. [Boyce, Real Estate Appraisal Terminology (rev. ed. 1981) at 234]
Neilson defined substitution as:
that cost today, at the appraisal date, of a structure, machine unit, or process unit based on the most current technology and construction materials available that will substitute for an existing unit with like capacity, service, desirability, and utility.
Generally, Neilson distinguished those improvements that were required for Chevron’s present asphalt operation as “operating” and that which was not needed for the asphalt facility as “nonoperating,” and valued them accordingly.
As Neilson explained, the asphalt plant makes asphalt for paving but the asphalt air blowing plant could upgrade the asphalt produced at the Chevron refinery so that it could be used for roofing.
It is interesting to note that the uninstalled cost new of the process unit is only one-third of the total cost of the process unit installed, connected and ready to operate. This was Neilson’s uncontroverted opinion based on the Kenneth Guthrie manual.
NeiIson testified that these units, if not used, would, in three to five years, become nothing more than scrap.
Neilson testified that his initial inclination was to ascribe no value to these old, outdated units — but after his discussions and consultations with Chevron he determined that scrap value would be appropriate.
NeiIson indicated that Chevron has had a few parties, who might be interested purchasers, inspect the units and, as he stated, “kick the tires." He further noted that although cost new of these units not installed or connected is $13,000,000, and while he valued them at $5,200,000, the informal discussions Chevron has had to date relative to a sale have "revolved around $3,000,000 to $4,000,000 for both units.” Neilson further indicated that if these units are not sold in a year or two, because of the rapid deterioration due to nonuse, they will have value for scrap only.
According to Marshall Valuation Service, replacement cost is:
The total cost of construction required to replace the subject building with the substitute of like utility. These costs include labor, materials, supervision, contractor’s profit and overhead, architect’s plans and specifications, sales taxes, and insurance. [Marshall Valuation Service, § 1 at 3]
Sigel’s estimates of excess construction or superadequacies were presumably based on that which was in excess of what he believed was necessary for a fully integrated 168,000-barrel per day oil refinery, while Neilson’s estimates of excess construction were based on that which was in excess of what was *142necessary to operate a 60,000-barrel per day asphalt plant. Thus, the issue is not whether Neilson’s theoretical basis for his replacement cost analysis was appropriate, but whether it was correctly applied in this case.
As a matter of fact, Daniel conceded that he did not have a lot of experience in estimating fair market values nor was he familiar with the extremely important appraisal concept of economic or external obsolescence.
Daniel’s conclusion as to depreciated replacement costs can only have validity in this proceeding if those conclusions can be equated by this court with the concept of fair market value as set forth in N.J.S.A. 54:4-23. In this context, it must be noted that “cost estimating is not appraising,” but is rather only "one step in the appraisal process.” See International Association of Assessing Officers, Property Assessment Valuation (1977) at 136.
In his section of the Day & Zimmerman appraisal report Daniel placed sulfur removal unit no. 1 in his cost estimates for the new refinery even though it was built before 1950 and, based on his own criterion, should have logically been placed in his cost estimates of the old refinery.
As part of its post-trial brief, Perth Amboy has submitted a newspaper article from the New York Times setting forth factual assertions presumably supportive of defendant’s position. Plaintiff has correctly noted that this offer is totally inappropriate, and therefore, the article along with the conclusions drawn by defendant solely from this article, offered as support for defendant’s position, will not be considered.
While market value is normally viewed "as a reflection of market participant’s perceptions of future economic conditions,” these perceptions of future economic conditions are based on market evidence of economic conditions, in a local property context, as of the assessment dates. See The Appraisal of Real Estate, supra at 65. Value reflects economic conditions at a specific time and as Chevron's expert, Neilson, noted, sudden economic changes can dramatically influence value. Id. at 38-39.
As Judge Lario of this court has noted in Owens-Illinois Glass Co. v. Bridgeton, 8 N.J.Tax 495 (Tax Ct.1986), it is still the general rule in local property taxation that property must be valued "in the actual condition in which the owner holds it.” Id. at 501-502; see also Trustees of Stevens Inst. of Technology v. State Bd., 105 N.J.L. 99, 101, 143 A. 356 (Sup.Ct.1928), aff’d 105 N.J.L. 655, 146 A. 919 (E. & A.1929) (property should be valued in the actual condition in which the owner holds it) and Hackensack Water Co. v. Bor. of Old Tappan, 77 N.J. 208, 214, 390 A.2d 122 (1978) (since property valuation must “have some relationship to reality," it would be proper to consider actual use as the highest and best use). This is not to say that actual use will always be a property’s highest and best use, but it is certainly a very relevant consideration.
There was one explanation given by Neilson that was not entirely satisfactory. With respect to his replacement of the existing administration building it was noted on cross-examination that the existing building had an elevator and was air-conditioned. Neilson’s replacement had neither. He explained that neither was necessary because he had overstated the value of the administration building because he replaced the building with a modern facility of the same size and that was more than would be necessary for an asphalt plant. This is not persuasive. However, considering the magnitude of the improvement value (890,924,400), I do not believe that Neilson’s failure to add the depreciated' cost of an elevator and air-conditioning to the administration building diminished his overall value conclusion significantly. His estimated values, in my opinion, still are more convincing than those of Perth Amboy's experts.
In light of my conclusion that Perth Amboy's valuation estimate is erroneous, because it was based on an incorrect premise, I have not detailed all of the discrepancies that exist in its experts' appraisal of the improvements. There were many. Following are only a few examples. Daniel estimated the value of several process units in the “old refinery" by trending his previously estimated values. Cost trending is based on converting a known historical cost, not an appraisal, into a current cost estimate by using a cost service. See The Appraisal of Real Estate, supra at 361-362. Sigel supposedly relied on Daniel for his determination that there was no economic or external obsolescence at the Chevron refinery, yet, Daniel did not know anything about, nor did he *152consider, economic obsolescence. Additionally, it is difficult to understand how Daniel’s estimates of replacement costs of the process units and related facilities can be equated with fair market value without a thorough consideration of economic conditions affecting the petroleum industry in general, and the Chevron refinery, in particular. Without considering the influence of externalities, Daniel's replacement costs would be equivalent to fair market value by coincidence rather than design. Id. at 38. Moreover, Daniel offered no objective basis for his scrap or salvage values. There is simply nothing in the record to support his value estimates for scrap or salvage items.