OPINION
ROBERT T. BENTON, II, Hearing Officer.This matter came on for hearing before the Division of Administrative Hearings by its duly designated Hearing Officer, Robert T. Benton II, on November 14, 1984, and concluded on December 7, 1984, the sixth day of hearing. Except for proceedings on December 3, 1984, in Punta Gorda, the hearing took place in Tallahasse, Florida.
At hearing, counsel .took a dismissal as to John Adler, who had been an original petitioner in Case No. 84-3572R. A motion on behalf of Shirley E. Dutton and Elizabeth D. Bronstein for leave to intervene made ore terms at hearing was denied as untimely.
The Department of Natural Resources (DNR) has proposed an amendment to Rule 16B-26.06, Florida Administrative Code, which would move the coastal construction control line in Charlotte County further inland. Both cases challenging the proposed amendment have been consolidated. Petitioners and intervenors contend that the proposed amendment is an invalid exercise of delegated legislative authority on grounds that DNR lacks authority to move the existing coastal construction control line, that the economic impact statement is defective and incomplete, that the proposed relocation is arbitrary and *183capricious and not in keeping with the intent of the statute purportedly implemented, and that the criteria or methodology by which the proposed relocation was decided amounts to an unpromulgated administrative rule, unenforceable against petitioners and intervenors. With respect to all these contentions, issue has been joined.
FINDINGS OF FACT
Charlotte County lies on the eastern shore of the Gulf of Mexico. The mainland is protected by a series of low lying barrier islands running more or less north and south. Manasota Key, Don Pedro Island and Gasparilla Island are the consolidated remnant of seven or more smaller islands. In all, Charlotte County has about 14 miles of sandy beach on the Gulf.
The location of the shoreline is not static. Along the stretch of beach between Stump Pass and the Sarasota County line, for example, the shoreline moved gulfward between 1883 and 1975, while the shoreline south of the pass moved landward between 1883 and 1939, then gulfward between 1939 and 1975. In very broad geological terms, the tendency of barrier islands is to migrate toward the mainland, but accretion is also ongoing. In general, the Charlotte County islands have moved further into the Gulf during the last century. Annual variation is typical: accretion in summer and fall follows erosion in winter and early spring.
Respondent DNR has placed reference monuments along the Charlotte County beaches every 1,000 feet or so, 68 in all. In May of 1974, DNR surveyed a profile of the beach at each station and also made a record of the bottom profile. Olfshore profiles were done at every third range to a depth of 30 feet, and, at the other ranges, out to a wading depth (four feet below mean sea level). Using this information, DNR promulgated Rule 16B-26.06, Florida Administrative Code, which established the existing coastal construction coastal line (the 1977 line).
STANDING
By law, DNR’s Division of Beaches and Shores has permitting authority over certain activities, notably building construction, in the area between the mean high water line and the coastal construction control line. The proposed rule amendment under challenge here would establish a new coastal construction control line (the proposed line) for Charlotte County that would lie landward of the 1977 along most, but not all, of its length. Except for Lisa Noden, the petitioners and intervenors in these consolidated cases own property in Charlotte County on the Gulf of Mexico, including property lying between the 1977 line and the proposed line.
*184Intervenor Boca Grande Club, Inc. owns Gulf frontage on Gasparilla Island including land lying between the 1977 line and the proposed line. On November 30, 1984, Boca Grande Club, Inc. had “the present intention to apply within the next six months for the necessary construction permits for a structure to be located on its real property,” Intervenors’ Exhibit No. 9, landward of the 1977 line and seaward of the proposed line. Respondent stipulated to the intervenors’ standing.
Dean L. Beckstead, president of Charlotte Harbor Land Company, has overseen the construction of 70 to 75 houses on that parcel of Don Pedro Island extending from Stump Pass 8000 feet south and from the Gulf of Mexico to Lemon Bay. With respect to some of these houses, construction is ongoing. The plan is to build additional housing, but no more than 450 residential units in all. In keeping with past practice, new construction would be well landward of the 1977 line, because of Mr. Beckstead’s great respect for the ocean, but might be seaward of the proposed line.
William McCrabb of Sarasota is an officer of a corporation, Nabob of Florida, Inc., that owns Gulf-front property on Manasota Key. He is also a general partner in a partnership that owns adjacent Gulf frontage. A 17-unit condominium has been completed on one parcel and plans exist for a 125-room hotel on the other. The only element of the hotel project seaward of the 1977 line is a planned dune overwalk. A larger portion of the hotel project would be seaward of the proposed line. Petitioner Charles Guy Bátsel owns a house that sits on Gulf-front property in Charlotte County. Some 5,000 square feet of this parcel lie between the 1977 line and the proposed line.
Even when obtaining a coastal construction permit does not result in changes in a project that have an adverse economic effect on a landowner, the costs associated with the permitting process itself may be substantial. The testimony of Randall Craig Norden, a developer, that he spent approximately $100,000, or at least 2.5 percent of the total cost of Phase Three of Colony Don Pedro, on attorney’s fees, engineering fees, travel to Tallahassee and other expenses associated with obtaining a coastal construction permit, went unrebutted.
REVISITING THE 1977 LINE
After it came to the attention of DNR staff that erosion along parts of the Gulf shoreline in Charlotte County had resulted in the 1977 line’s approaching the water’s edge in several places, staff recommended that the line be reexamined. The Governor and Cabinet, in approving DNR’s annual work program in 1983 and in voting to enter into a contract with Florida State University for, e.g., “Studies to *185Reestablish Control Lines”, Petitioners’ Exhibit No. 3, on July 1, 1983, ordered a comprehensive review. Even before the Governor and Cabinet took these actions, DNR staff performed a survey in Charlotte County in 1982 to determine beach and bottom profiles at the same points at which they had been measured in 1974, although in two or three instances, the monuments had washed away.
On Manasota Key between ranges 1 (the northernmost in Charlotte County) and 5, the mean seal level line receded an average of 15 feet between May of 1974 and December of 1982. The mean sea level moved further toward the Gulf on average between ranges 6 and 11, but receded an average of 20 feet between ranges 12 and 18. Between ranges 16 and 24, no relocation of the coastal construction control line has been proposed.
Displacement of sand when Stump Pass was dredged may have affected the shoreline south of the pass, although shorelines in the vicinity of inlets are ordinarily unstable. On Don Pedro Island, just south of Stump Pass, there has been accretion. Between ranges 27 and 39, which lie still further south of Stump Pass, the mean seal level line receded an average of 81 feet between May of 1974 and December of 1982. Between ranges 45 ad 49 the line has moved an average of 32 feet landward while there has been accretion, on average, between ranges 50 and 55.
At range 60, the northern end of Gasparilla Island, the mean sea line had receded 100 feet between May of 1974 and December of 1982 and another 10 feet by September 14, 1983. At range 61, the mean sea level line had moved seaward by 100 feet between May of 1974 and December of 1982, while at range 62 there was a seaward shift of 75 feet over the same period. At range 63, there was accretion between May of 1974 and December of 1982 but erosion brought the mean sea level line landward of its May 1974 location by September of 1983. Between ranges 64 and 67, the average recession of the mean sea level line was 76 feet.
By one calculation, the county as a whole lost about 59,000 cubic yards of beach material between May of 1974 and December of 1982. Measurements made shortly before the “No Name” tropical storm occurred, in the summer of 1982, suggest that the storm did not significantly affect these measurements of long term trends.
At no time did riparian landowners or officials of Charlotte County or any affected municipality make any written request that the coastal construction control line be moved, although Franz H. Ross, one of Charlotte County’s county commissioners, testified at the hearing that *186the 1977 line needed replacement. He did not endorse DNR’s proposed line.
After the 1977 line was drawn, the enabling legislation was twice amended. The first reference to a 100-year storm surge appeared in 1978. Ch. 78-257, Section 5, Laws of Florida (1978). More recently, storm waves as well as storm surge became a statutory criterion. Ch. 83-247, Section 2, Laws of Florida (1983). Not only have new laws and a new beach emerged since the 1977 line was established, but there have also been advances in scientific analysis and prediction of the behavior of storm waves, notably with reference to surf beats or “dynamic wave set up.”
ECONOMIC IMPACT STATEMENT
DNR prepared a 27-page economic impact statement in which it estimated the costs of the proposed line to the agency, and costs and benefits to persons directly affected by the proposed rule; and made a detailed statement of the data and method used in making these estimates. With respect to the impact of the proposed rule on competition and the open market for employment, the economic impact statement noted that construction costs would increase under the proposed rule, causing a “market adjustment period” during which “builders would have to absorb the cost increase themselves or delay construction while prices rise sufficiently. . . Postponing of construction would tend to reduce employment . . . temporar[il]y. . . .” Petitioners’ Exhibit No. 5, p. 13.
Although portions of the economic impact statement were originally drafted for Dade and Broward County control lines, they have obvious application to Charlotte County, as well. The thrust of the cost-benefit analysis was that construction costs would increase in the area between the 1977 line and the proposed line, but that enhanced preservation of the beaches, and decreases in flood insurance premiums and storm damage potential would more than offset these increased costs. The increase in construction costs was attributed to the expense and delay of obtaining a coastal construction permit, the additional labor and materials necessary to elevate the structure above the predicted level of storm waves in a 100-year return storm, and the relatively insignificant cost of installing stronger connections (hurricane clips for the roof and bolted metal straps over joists to secure them to supporting piles) so that the structure could withstand wind loads of 140 miles per hour. Petitioners did not disprove the reasonableness of the permitting cost assumptions in the economic impact statement, the evidence of Colony Don Pedro’s experience notwithstanding. Fire escapes and access for *187handicapped persons were not taken into account, but the evidence did not show that differential costs for those items would affect the conclusions of the economic impact statement.
The economic impact statement assumes that buildings would have to be elevated off grade even without the additional coastal construction control requirements, and that foundations would be designed by engineers, in any case. Neither assumption was proven false. The assumption that costs increase in direct proportion with elevation yields only a very rough estimate of differential costs.
The differential cost analysis did not take fully into account the criterion that applies in coastal construction permitting that relates to a structure’s two-dimensional “footprint.” DNR permitting staff may recommend denial of a permit even though a proposed building meets all structural integrity requirements whenever, on a site-specific basis, the area to be covered by the building fails to “minimize any expected adverse impact on the beach system.” Rule 16B-33.07(2), Florida Administrative Code. In such circumstances, one resolution may be to place the proposed structure at a more landward site, and the economic impact statement does address the economic consequences of removing structures landward, but other resolutions, such as scaling down the project or decreasing floor size and adding floor(s) are not considered. In this connection, there is no mention of Charlotte County’s three-story (35 feet) height limitation nor, in general, does the economic impact statement identify what Charlotte County ordinances now require for coastal construction.
On the benefit side, some of the flood insurance rate comparisons are inappropriate because Charlotte County will not allow floor elevations several feet below base flood elevations set by the Federal Emergency Management Agency, as the economic impact statement assumed for comparative purposes. The rate differentials also apparently ignore the fact that the same structure at a higher elevation will be worth more and have a greater insurable value. From Petitioners’ Exhibit No. 23, moreover, it appears that federally subsidized flood insurance may no longer be available in certain parts of Charlotte County. But the Sheaffer and Roland study, Respondent’s Exhibit No. 14, to which the economic impact statement refers, shows that elevating a structure to the wave crest level instead of to the storm surge level of a 100-year return storm creates additional benefits in the form of lessened storm damage potential (without regard to insurance premiums) that exceed the additional costs. With respect to benefits as well as costs, the economic impact statement suffers from a failure to explicate existing requirements of local law governing building construction.
*188Without this base line, differential costs and benefits were not and cannot be quantified precisely for the specific case of Charlotte County. But see page 12 of Petitioners’ Exhibit No. 5. The evidence adduced at hearing failed, however, to discredit the general conclusion of the economic impact statement that requiring sufficient elevation and sufficiently sturdy connections to withstand a 100-year storm event, including the wave crests it would generate, was cost effective, assuming the structure is to have an engineered foundation off grade, in any event. Precise quantification of the economic impact of the site coverage criterion DNR will extend to the area between the 1977 line and the proposed line, if the latter takes effect, may be impossible. The benefits will accrue to the beach and to adjacent landowners as much as to the owner of the structure and both costs and benefits will vary from parcel to parcel with changing topography.
THE SANDS OF TIME
Under conditions that have recently obtained in Charlotte County, sloping sand beaches climb from the water’s edge to the toe of a more or less pronounced primary sand dune, behind which other dunes undulate in succession across the barrier islands to Lemon Bay or Gasparilla Sound, from which they are occasionally insulated by mangrove swamp. Vegetation over much of the islands, which vary from 200 to 2000 feet in width, attests to their present stability.
But chances are that a hurricane will in time strike, flattening the dunes, spreading the sand well inland everywhere, all the way across the islands in some places, and leaving a wide beach face without, in many places, any discernible dunes. Such a reconfiguration will ineluctably result from the major hurricane identified as the 100-year return storm.
Thereafter, under more favorable weather conditions, dunes will grow and reemerge, comprised of sand the Gulf gives back as well as the sand strewn across the island by the storm, unless surface impediments prevent. The cycle complete, dunes will again stand their erosion-damping vigil against the sea, a buffer protecting the mainland, as well as insular upland.
In establishing coastal construction control lines, DNR is charged by statute with the job of “defining] that portion of the beach-dune system which is subject to severe fluctuations based on a 100-year storm surge, storm waves, or other predictable weather conditions.” Section 161.053(1), Florida Statutes (1983). DNR naturally looks to the beach dune system in the configuration it is predicted to assume after a 100-year return storm in defining “that portion of the beach-*189dune system which is subject to severe fluctuations based on a 100-year storm...” The folly of limiting consideration to the landward toe of a primary dune as it existed in a period of fair weather as well illustrated by the photograph that came in evidence, as Respondent’s Exhibit No. 12, of a monument placed behind a dune in St. Johns County only recently that is now well down on the beach.
DNR METHODOLOGY
In drawing the proposed line, DNR followed the procedure it employed in establishing coastal construction control lines in Nassau, Franklin and perhaps other counties. The line proposed for Martin County was drawn in the same fashion.
By statistical inference from historical data, five characteristics of a predicted 100-year return storm are identified: central pressure deficit, radius to maximum winds, speed of hurricane system translation, hurricane direction (track angle), and landfall location or some other geographical reference. In order to assess the likelihood of various combinations of storm attributes actually observed, cumulative probability curves are generated and predictions of storms and their characteristics over several hypothetical 500-year time periods are made. Of each 500-year suite of storms, the fifth most severe is chosen as the 100-year return storm. A date between June 1 and November 30, 1982, is chosen at random and astronomical tides on that day are assumed to coincide with the 100-year return storm. Using the average characteristics of the 100-year return storm, associated wind velocities and storm surge are predicted.
Astronomical tides, barometric pressure, wind stress, and the Coriolis effect all contribute to the height of the still water storm surge, “the water level that you could measure at a point due to a hurricane’s passage if you could turn off the waves.” Dean’s deposition at 31. The storm surge prediction model also takes storm waves into account: Maximum wave height is 78 percent of water depth. As storm waves approaching a beach reach the break point, their height falls by five percent, but the waves attain and exceed their former height by the time they reach shore. Waves have momentum which, as they break, is transferred, at least in part, “to the water column in the form of a wave setup.” Dean’s deposition at 32.
For many years, “static wave setup” have been observed in wave tanks where waves of the same size have been set in motion to break against a wall of the tank. For six years or so, Drs. Dean and Chiu and others have been convinced that an additional allowance should be made for surf beats or “dynamic wave setup”, to reflect the fact that *190waves in nature do not occur in uniform sizes at regular intervals. Their magnitude oscillates, in the case of breaking storm waves, around the still water surge elevations. To allow for dynamic wave setup, the static wave setup component of the predicted surge elevation is increased by half.
The 100-year storm surge height is then used to predict, taking observed beach profiles into account, the landward penetration of waves which will have degenerated to heights of three feet and, with the aid of a mathematical model, the extent to which stormwater transporting sand offshore will cause erosion. At each range, unless a three foot wave is predicted to go further landward, the point to which erosion by offshore transport is predicted to occur is chosen as the endpoint for a segment of the coastal construction control line. Where penetration of a three foot wave farther landward is predicted, the coastal construction control line is drawn on that basis, in light of topography on either side of the range involved. In predicting the landward penetration of a three foot wave, aerial photographs or surveyor’s field notes are consulted and, if there is vegetation along the range involved, a coefficient of friction is applied that simulates the existence of trees a foot in diameter with centers five feet apart.
At the hearing, the use of a three foot wave horizontal penetration criterion was called into question, and there is an apparent difference of opinion between two of DNR’s experts, Dean and Chiu, as to the significance a three foot wave has for the beach dune system.
The three foot wave is notorious. This unassuming natural phenomenon has become laden with engineering and legal significance, ever since 1962 studies the Army Corps of Engineers performed in Galveston, Texas, showed that a three foot wave had enough energy to demolish a frame structure build on grade. If it has as much energy as that, Dr. Chiu reasons convincingly, it also has enough energy to transport significant amounts of sand and to damage vegetation. With its sand-holding properties, vegetation plays a critical role in the beach-dune system. Evidence that a three-foot wave does not rearrange substantial quantities of sand, if any was adduced, has not been credited. Even Dr. Dean reported seeing three foot waves moving substantial quantities of sediment.
The mathematical erosion model, known for the inventor as the modified Kriebel model, assumes relatively higher sand dunes that are eaten away by waves transporting sand offshore. The erosion model does not take into account lateral movement of sand or the effects of waves overtopping a dune and carrying overwashed sand inland. The *191model predicts what distance inland an assumed storm surge will move various contours. By comparing the model’s predictions to the effects of actual storms, calibration has been possible. After Hurricane Eloise hit Walton County, erosion along 25 miles of shoreline was observed and compared to the model’s predictions. In order to draw a line landward of 98 percent of the points to which Eloise eroded the five foot contour, it was necessary to multiply the erosion model’s prediction of the landward movement of the five foot contour by 2.5. With the 2.5 factor the model overpredicts for most of the coast affected by a storm, but underpredicts erosion for the point where the storm does its worst. The most severe erosion eats two and a half times further inland than average erosion along the affected coastline.
AS APPLIED
Underlying both approaches to drawing the proposed line was the prediction of the water height a 100-year return storm would produce. DNR calculated three storm surges, one for the northern, one for the central, and one for the southern Charlotte County coast, and predicted maximum water heights during a 100-year storm ranging from 13.1 to 12.7 feet above NGYD at the mean sea level line. Identifying the 100-year return storm for a particular locale is an elaborate exercise in probability theory that begins with the collection and analysis of historical data. At hearing, various criticisms of this process were advanced.
Among them was the way certain data were assigned to categories or “bins”. On rebuttal, the data were treated as discrete points and the result was the prediction of a storm surge .8 feet higher than the “bin” prediction method had yielded. Another criticism was the number of hurricanes selected as pertinent over the 80-year period studied. To the same effect was a criticism of the length of the period chosen. Reducing the number of storms from 28 to 20 causes the predicted storm surge to fall half a foot. Respondent’s Exhibit No. 18. In their proposed recommended order, intervenors contend
Had DNR used . . . correct data in its development of the predicted 100-year storm surge level for Charlotte County, it would have simulated only 137 hurricanes for a 500-year period in Charlotte County. By simulating 182 hurricanes using the storm surge computer model, DNR simulated 45, or 33%, more hurricanes than justified . . . . P. 13 (footnotes omitted)
The 33% is a red herring, since, as intervenors later note, the difference between simulating 182 hurricanes and simulating 125 hurricanes produces a difference of only .7 feet (from 12.9 to 12.2 feet) in *192the predicted storm surge, a difference of five or six percent. Nor was DNR’s approach shown to be “incorrect”.
A certain amount of time at hearing was devoted to the categorization of hurricanes as alongshore (or bypassing), landfalling, or exiting. Different sources of data may be a source of confusion, if compiled with reference to different points of geographical reference. A hurricane that makes landfall in Miami may exit the peninsula in Charlotte County, while a hurricane that makes landfall in Tampa may bypass Charlotte County en route. Some confusion seems to have attended the integration of data taken from a National Weather Service publication that was used joinly with data from a NOAA source. Respondent’s Exhibit No. 18 demonstrates that no significant distortion resulted, however. Among the historical storms omitted by DNR’s consultants was the infamous Labor Day Hurricane of 1935, which generated storm tides of 18 feet at Marathon in the Florida Keys. Using only the 1973 NOAA data for Charlotte County yields predictions of storm surges a foot higher than DNR’s consultants predicted.
On Manasota Key, it was the erosion model that determined placement of the proposed line. Since the probability of the 100-year return storm hitting at any particular point on the Charlotte County coastline is virtually the same as for any other point on the Charlotte County coastline, the 2.5 factor is appropriate. The average amount of erosion over the whole of the affected coastline is of theoretical interest only.
Between ranges 1 and 10 on Manasota Key, the dunes are 12 to 14 feet high, as compared to an average elevation for all three islands of slightly above five feet. The high dunes on Manasota Key, where the erosion model was used to set the proposed line, resemble the walls of a wave tank more closely than the lower dunes on , Don Pedro and Gasparilla Islands, where stormwater is predicted to cross the islands and keep going. Dr. Chiu also predicted overtopping of Manasota Key. Dr. Dean’s testimony was to the effect that overtopping might mean a diminution in wave height attributable to diminished dynamic wave setup of .3 to .4 feet. To this should logically be added a corresponding diminution attributable to diminished static wave set up, viz., .6 to .8 feet, for a total of up to 1.2 feet.
The controversial testimony was that the storm surge model ignores altogether a documented phenomenon known variously as “initial rise”, “forerunner”, and “presurge anomaly”, that adds 1 to 3 feet to surge elevations, and that this factor would offset any diminished wave setup almost entirely.
*193Dr. Chiu’s opinion that a three foot wave would cause overwash and damage to vegetation on Don Pedro and Gasparilla Islands, severely damaging the beach dune system, has been accepted. The contention that these phenomena are immaterial since they were characterized as “severe impact on” instead of “severe fluctuations of’ the beach dune system must be dismissed as a semantic quibble.
Penetration of the three foot wave was the controlling criterion from Range 25 south to the Lee County line, although the average prediction of wave length at points where ranges intersect the proposed line is between 3.3 and 3.4 feet. These wave height predictions ignore, moreover, the erosion of the profile which is bound to occur. Witnesses on all sides agreed that a 100-year return storm would submerge Don Pedro and Gasparilla Islands. Waves three feet high and higher would travel across the barrier islands and Lemon Bay onto the mainland.
In Charlotte County, use of the coefficient of friction was very conservative, inasmuch as vegetation on none of the coastal barrier islands approaches the density of trees one foot in diameter, five feet apart on centers. Trees a foot in diameter 15 or 20 feet apart cause only a four percent reduction in wave height every 100 feet as compared to the 20 percent reduction every 100 feet assumed for vegetated ranges in Charlotte County.
PROOF IN PUDDING
The conservatism of DNR’s methodology is shown by damage done landward of the coastal construction control line in Franklin County, where a road upland from the line washed out during a storm of less than 100-year return magnitude and in Martin County, where as the result of a 10-year return storm severe topographical fluctuations were seen landward of the coastal construction control line proposed there.
The storm surge model has been calibrated against storms of record, which has demonstrated its reliability, whatever the merits of its theoretical underpinnings. Even assuming some problems with the categorization of storm parameters put into the model for Charlotte County, the evidence adduced in this case does not support the conclusion that stormwaters in Charlotte County will reach an elevation of less than 11 to 12 feet, in the event of a 100-year return storm. The evidence overwhelmingly establishes that waves would reach at least that height.
Respondent’s Exhibit No. 2 shows what a storm surge of 11 to 12 feet did to a beach not dissimilar to Charlotte County’s when Hurricane Frederic hit Gulf Shores, Alabama. Radical changes in topogra*194phy occurred well landward of the point at which the proposed line for Charlotte County has been placed. This exhibit also shows why Mr. Tackney’s opinion to the effect that a modest reduction in storm surge would permit sand dunes of a certain height to block the water’s progress landward must be rejected. When a hurricane hits, the dunes are so dramatically eroded that their former height is not determinative. Although DNR ignored this factor in calculating the horizontal penetration of a three foot wave, in order to make the proposed line conservative, erosion of the profiles is inevitable.
The Kriebel erosion model is designed to show what amount of sand stormwater will move offshort when a dune takes the full force of stormwater. When dunes are not high enough to do that, some other criterion for a coastal construction control line is necessary to reflect the different types of erosion that hurricane Frederic and other storms have shown will occur. Overwashed sand deposits stretched 800 feet and more from the water’s edge after Frederic abated.
The parties proposed findings of fact have been considered and have been adopted, in substance, except where unsupported by the weight of the evidence, immaterial, cumulative, or subordinate.
CONCLUSIONS OF LAW
The Division of Administrative Hearings has jurisdiction of petitions to invalidate agency rules pursuant to Section 120.54(4), Florida Statutes (1983).
Petitioners assert that the proposed amendment to Rule 16B-26.06, Florida Administrative Code, is unlawful because, they contend, (a) DNR lacks authority to reset a coastal construction control line in the circumstances of this case, (b) the economic impact statement is defective, (c) the methodology used to establish the line is invalid as an unpromulgated rule, and (d) the proposed line has not been set in a way that defines “that portion of the beach-dune system which is subject to severe fluctuations based on a 100-year storm surge. . .” Section 161.053(1), Florida Statutes (1983), and so lacks statutory authorization. The statute states in full:
(1) The Legislature finds and declares that the beaches in this state and the coastal barrier dunes adjacent to such beaches, by their nature, are subject to frequent and severe fluctuations and represent one of the most valuable natural resources of Florida and that it is in the public interest to preserve and protect them from imprudent construction which can jeopardize the stability of the beach-dune system, accelerate erosion, provide inadequate protection of upland *195structures, and endanger adjacent property and the beach-dune system. In furtherance of these findings, it is the intent of the Legislature to provide that the department establish coastal construction control lines on a county basis along the sánd beaches of the state fronting on the Atlantic Ocean or the Gulf of Mexico. Such lines shall be established so as to define that portion of the beach-dune system which is subject to severe fluctuations based on a 100-year storm surge, storm waves, or other predictable weather conditions. However, the Department may establish a segment or segments of a coastal construction control line further landward than the impact zone of a 100-year storm surge, provided such segment or segments do not extend beyond the landward toe of the coastal barrier dune structure that intercepts the 100-year storm surge. Such segment or segments shall not be established if adequate dune protection is provided by a state-approved dune management plan. Special siting and design considerations shall be necessary seaward of established coastal construction control lines to ensure the protection of the beach-dune system, proposed or existing structures, and adjacent properties.
(2) Coastal construction control lines shall be established by the department only after it has been determined from a comprehensive engineering study and topographic survey that the establishment of such control lines is necessary for the protection of upland properties and the control of beach erosion. No such line shall be set until a public hearing has been held for each area involved. After the department has given consideration to the results of such public hearing, it shall, after considering ground elevations in relation to historical storm and hurricane tides, predicted maximum wave uprush, beach and offshore ground contours, the vegetation line, erosion trends, the dune or bluff line, if any exist, and existing upland development, set and establish a coastal construction control line and cause such line to be duly recorded in the public records of any county and municipality affected and shall furnish the clerk of the circuit court in each county affected a survey of such line with references made to permanently installed monuments at such intervals and locations as may be considered necessary. Upon the establishment, approval, and recordation of such control line or lines, no person, firm, corporation, or governmental agency shall construct any structure whatsoever seaward thereof; make any excavation, remove any beach material, or otherwise alter existing ground elevations; drive any vehicle on, over, or across any sand dune; or damage or cause to be damaged such sand dune or the vegetation *196growing thereon seaward thereof except as hereinafter provided. Control lines established under the provisions of this section shall be subject to review at the discretion of the department after consideration of hydrographic and topographic data which indicates shoreline changes that render established coastal construction control lines to be ineffective for the purposes of this act or at the written request of official of affected counties or municipalities. Any riparian upland owner who feels that such line as established is unduly restrictive or prevents a legitimate use of his property shall be granted a review of the line upon written request. After such review, the department shall decide if a change in the control line as established is justified and shall so notify the person or persons making the request. The decision of the department shall be subject to judicial review as provided in Chapter 120. Section 161.053(1) and (2), Florida Statutes (1983).
Subsection two specifies the circumstances in which DNR is to reexamine placement of a coastal construction control line and, if appropriate, relocate it.
Like challengers of rules already adopted, parties challenging as invalid proposed rules or, as here, proposed rule amendments, have the burden of showing invalidity. The challengers must show that “the agency, if it adopts the rule, would exceed its authority. . . .” Agrico Chemical Co. v. State Department of Environmental Regulation, 365 So.2d 759, 763 (Fla. 1st DCA 1978).
REVIEW APPROPRIATE
At one time, Section 161.053(2), Florida Statutes, provided that coastal construction control lines “shall .be subject to review by the department at 5-year intervals.” The present wording can be traced to Ch. 80-183, Section 3, Laws of Florida (1983), which began as Senate Bill 841. The senate staff analysis explained the change:
Current law provides that established county coastal construction control lines be reviewed by DNR every 5 years. DNR staff has found that some control lines do not need review this frequently while other control lines require more frequent review. . . . This bill permits DNR to schedule review “as needed”, determined by the evaluation of pertinent hydrographic and topographic data.
More than ten years have now elapsed from the survey in 1974 on which the existing control line is premised.
There have been significant changes in the coastline in that time, but there have been other changes as well. It is the other changes that account for the fact that the proposed line has been placed more *197landward of the 1977 line, even in some places where there has been accretion. Since 1977, the statute that Rule 16B-26.06, Florida Administrative Code, was adopted to implement has itself been amended, first to provide that the “severe fluctuation zone” be defined with reference to a 100-year storm, Chapter 78-257, Section 5, Laws of Florida (1978), then to specify that storm waves be taken into account. Chapter 83-247, Section 2, Laws of Florida (1983). These changes alone may well have indicated to DNR staff that review was “needed”. When the question was reexamined, current scientific knowledge was naturally brought to bear. Whether mere advance in scientific understanding could justify review of an existing coastal construction control line need not be decided in the present case.
ECONOMIC IMPACT STATEMENT
While the economic impact statement addresses a multiplicity of subjects, it speaks largely in general terms, without specifying what requirements now exist under county ordinances, and without quantifying the exact differential impact of the proposed rule. The full consequences of the “footprint” requirements are not explored, although there is much to commend Dr. Nicholas’s observation that since the effect will vary with individual topography, it is difficult to generalize. One of the best features of the economic impact statement is the bibliography.
The precise economic impact depends on the use to which individual owners put their respective parcels. The value of land already fully developed will be enhanced, according to the evidence. Here, as in Brewster Phosphates v. Department of Environmental Regulation, 444 So.2d 483 (Fla. 1st DCA 1984), the intentions of various owners, in addition to the topography of their individual parcels, are “incapable altogether of estimation,” at 487, in any generally applicable way. The court said in the Brewster Phosphates case:
As a matter of law, the economic impact statement was sufficient. As in State Department of Insurance v. Insurance Service Office, 434 So.2d 903 (Fla. 1st DCA 1983), the Department’s statement is a lengthy (ten page) document which complies paragraph by paragraph with the substantive requirements of Section 120.54(2)(a). Admittedly, some of the costs are speculative or incapable altogether of estimation, in view of the fact that appellants have no immediate plans for development of their property as phosphate mining sites. As in State Department of Insurance, however, the Department can hardly be faulted for failing to make estimates on the basis of unknown variables. 444 So.2d at 487.
*198Here, too, the economic impact statement is a “lengthy . . . document”. It runs 27 pages and addresses the statutory criteria in a meaningful, if not an ideal way.
The proposed amendment to Rule 16B-26.06, Florida Administrative Code makes uninspiring reading: It is nothing more than a listing of metes and bounds. The amendment extends DNR’s permitting jurisdiction geographically, but does not effect any change to permitting procedures and criteria, which are the subject of other provisions of Chapter 16B, Florida Administrative Code. On balance, it cannot be said that the shortcomings of the economic impact statement have worked to the material harm of petitioners and intervenors any more than the appellants in Florida-Texas Freight, Inc. v. Hawkins, 379 So.2d 944 (Fla. 1979) suffered material harm.
METHODOLOGY NO RULE
DNR’s proposed amendment to Rule 16B-26.06, Florida Administrative Code, makes no reference to the mathematical models or other criteria DNR’s staff and consultants looked to in drawing the proposed line. Intervenors contend that DNR’s methodology, which the evidence showed to have been employed in drawing other coastal construction control lines, amounts to an administrative rule itself and should have been promulgated as such. But DNR’s methodology was not shown to constitute “agency policy statements of general applicability intended to be applied with the force of a rule of law.” State Department of Administration v. Stevens, 344 So.2d 290, 296 (Fla. 1st DCA 1977). DNR’s methodology is at most a tenative “rule of decision” for promulgation of genuine administrative rules, rules which are “to be applied with the force of a rule of law”.
The methodology does not purport to control any adjudication of substantial interests pursuant to Section 120.57, Florida Statutes. Nor did the evidence show that the methodology was used in promulgating most of the coastal construction control lines now in place. The methodology is subject to reexamination and reformulation in the course of rule proceedings instituted to establish or reestablish any particular coastal construction control line. Until and unless such proceedings eventuate in a specific coastal construction control line, the methodology implicit in the line lacks legal affect.
THE MERITS
Under a variety of technical headings, petitioners and intervenors contend that the proposed line is “not appropriate to the ends specified in the legislative act”. Agrico Chemical Co. v. State Department of *199Environmental Regulation, 365 So.2d 759, 763 (Fla. 1st DCA 1978). They contend that use of a dynamic wave setup component in the storm surge model was unjustified and a “denial of due process” because theoretical work on this phenomenon is relatively new and “not accepted”; that the data used in the storm surge model were skewed by improper selection and categorization of historical storms; that the penetration of a three-foot wave criterion does not comport with statutory intent; that the penetration of a three foot wave criterion was not consistently applied, in any event; and that “the erosional model with the 2.5 multiplier will over-predict erosion in 98 of 100 given locations on the shore during a 100-year storm event”.
Petitioners and intervenors did not carry their burden to show that DNR’s consultants materially overpredicted the height of storm water reasonably to be anticipated in the event a 100-year return storm hits Charlotte County. Petitioners’ and intervenors’ experts testified that they were unfamiliar with efforts to quantify dynamic wave setup and did not believe such efforts had received widespread acceptance, but they did not dispute the existence of the phenomenon. Intervenors’ expert witness, a coastal engineer who conceded he was no expert either in statistics or in hurricanes, consulted a statistics text and raised various possibilities for different treatment of historical hurricane data, but, a person of obvious intelligence and integrity, he did not testify that altering the analysis in keeping with any of his criticisms would have yielded significantly lower storm surge predictions. He offered reasons why it might, but he did not do the calculations, or pretend to know what they would show. Respondent’s experts showed that implementation of one of intervenors’ expert’s suggestions—discrete point analysis—would increase the predicted evaluation. No attempt was made to rebut respondents’ experts’ assertions that the storm surge model had been calibrated or “checked” against storms of record and its reliability demonstrated. It is conceivable that the storm surge model produces good results for the “wrong reasons”, although, again, there was no proof of any significant defect in the theory or use of the storm surge model.
USE OF THREE-FOOT WAVE
Intervenors and petitioners contend that Section 161.053(1), Florida Statutes (1983) does not authorize use of the horizontal penetration of a three-foot wave as a means of setting a coastal construction control line. One argument to this effect hinges on imputing unlikely significance to the difference between the phrases “that portion of the beach-dune system which is subject to severe fluctuations based on a 100-year *200storm surge”, and “impact zone of a 100-year storm surge”, language which the statute seems to use interchangeably. Testimony throughout the record to the effect that a three foot wave would have a severe impact on the beach dune system, particularly Dr. Chiu’s testimony to that effect, clearly established that severe fluctuations to the beach dune system, in the area defined by penetration of the three foot wave, will occur in a 100-year return storm.
The erosion prediction model assumes dunes high enough to reflect storm waves, and only predicts erosion attributable to sand transported offshore by receding waves. Along most of the Charlotte County coast, in the event of a 100-year return storm, storm waves would roll across the barrier islands, and the principal movement of sand would be inland, not offshore, to be deposited as overwash. Selection of a three-foot wave as one with sufficient energy to move significant amounts of sand inland seems conservative, in light of the fact that such a wave has enough energy to knock a house down (an attribute which is of only incidental significance in connection with establishing coastal construction control lines.)
Intervenors contend that variance between predicted wave heights ¿t the different ranges makes the proposed line arbitrary. Dr. Chiu explained the lower extreme of 2.1 feet at range 68, where the monument was at an elevation surrounded by depressions on either side. In order to place the line with reference to the topography of that stretch of beach, rather than on the basis of one unrepresentative point, he used engineering judgment to select lower wave heights for the range. Over all 68 ranges, the second lowest value used was 2.9 feet. The extreme variation in the other direction was much greater, 5.3 feet, at range 47. At that range, the line was placed well seaward of where it would have been, if surrounding topography had been ignored.
In addition to rises and falls in the terrain, DNR staff also had to contend with the limitations inherent to surveying. DNR adopted the common sense approach of defining the proposed line as a sequence of straight line segments. The evidence showed that no coefficient of friction for shrubs was available to DNR’s consultants; and that the coefficient for trees that was employed worked greatly to the advantage of those wanting to see the proposed line placed as seaward as possible.
EROSION MODEL
Petitioners complain that the erosion model was calibrated, by use of the 2.5 factor, in a way that accounted for 98 percent of erosion an assumed storm would produce. But, as used in the present case, the *201model did not predict the worst erosion that a 100-year return storm could reasonably be expected to cause. If it had, the proposed line would have been placed further landward. The proposed line was set, at each range, on the assumption that the 100-year storm would hit at that point and not on the assumption that the profile being considered would be on the periphery of the storm. Use of the erosion model as calibrated was appropriate to that end.
It is, accordingly,
ORDERED:
1. As to Lisa Noden, the petition to invalidate agency rule is dismissed.
2. The other petitioners’ and intervenors’ prayer, that the proposed ámendment to Rule 16B-26.06, Florida Administrative Code, be determined invalid, is denied.