OPINION

CARPENETI, Justice.

I. INTRODUCTION

This case involves a challenge by citizens and environmental groups (collectively, Kachemak Bay) to the Department of Natural Resources' (DNR) decision that an oil and gas lease sale concerning state lands in Cook Inlet was (1) in the best interests of the state and (2) consistent with the state's coastal management program.

The challenge has two major aspects. First, Kachemak Bay contends that DNR impermissibly "phased" its review of the proposed sale. Second, Kachemak Bay challenges the substantive basis for DNR's decisions.

Because we find that Kachemak Bay's challenge fails in all respects, we affirm the decision of the superior court.

II. FACTS AND PROCEEDINGS

In early 1996 the state offered over one million acres of state-owned on-shore and off-shore land for lease for potential petroleum exploration and development. This land is almost entirely within the Kenai Peninsula Borough, and is predominantly either under the waters or on the coastline of Cook Inlet.

DNR evaluated the proposed sale. It determined that (1) the sale was in the best interests of the state, and (2) the sale was consistent with the Alaska Coastal Management Plan. DNR issued two documents — a best interests finding and conclusive consistency determination, respectively — to that effect in September 1996.

Kachemak Bay challenged these determinations. It first requested reconsideration of both decisions, which DNR denied in October 1996. Kachemak Bay then appealed to the superior court.

In November, Kachemak Bay unsuccessfully moved for an injunction against the lease sale; the sale took place as scheduled in December. Extraction rights to over 173,000 acres were leased, generating over $3 million in state revenue. The superior court still had before it Kachemak Bay's underlying appeal of DNR's findings regarding the lease sale.

After our decision in Ninilchik Traditional Council v. Noah was issued on December 26, 1996, DNR asked the superior court to remand the appeal to allow DNR to reconsider its conclusive consistency determination. DNR issued the revised conclusive consistency determination in November 1997.

The superior court upheld DNR's best interests finding and conclusive consistency determination in all respects in January 1998. This appeal followed.

III. STANDARDS OF REVIEW

Superior Court's Decision

Here the superior court sat as a court of appeal reviewing DNR's decision. In such cases, we independently review the merits of the administrative determination; we do not defer to the superior court's decision.

DNR's Decision

We have held that "to what extent the . . . code allows phasing . . . . is a question of statutory interpretation which does not involve agency expertise. Thus, this court will use its independent judgment."

Once we have determined whether and to what extent the relevant law allows phasing, both DNR's best interests determination and its determination that a project is consistent with the Alaska Coastal Management Plan's habitat standard are subject to a deferential reasonable basis review. This standard properly reflects the fact that in these cases, DNR's determination "is almost entirely a policy decision, involving complex issues that are beyond this court's ability to decide. . . . This court has neither the authority nor competence to decide whether the public interest is `best served' by a proposed disposition of land for offshore oil and gas exploration and development."

However, while deferential, this is not a toothless standard of review. On the contrary, we have stated that our duty is to ensure that DNR has taken a "hard look at the salient problems and has genuinely engaged in reasoned decision making." Further, we have held that such decisions "will be regarded as arbitrary where an agency fails to consider an important factor in making its decision."

DISCUSSION

The Best Interests Finding

DNR's obligation to consider the "best interests of the state" and to issue written findings when it proposes to alienate state land or an interest in state land can be traced to the Alaska Constitution. Article VIII, Section 1 proclaims that "[i]t is the policy of the State to encourage the . . . development of its resources by making them available for maximum use consistent with the public interest." Section 2 further provides that "[t]he legislature shall provide for the utilization, development, and conservation of all natural resources belonging to the State, including land and waters, for the maximum benefit of its people." Finally, Section 8 endows the Legislature with the power to "provide for the leasing of, and the issuance of permits for the exploration of, any part of the public domain or interest therein, subject to reasonable concurrent uses."

In Title 38, Chapter 5 of the Alaska Statutes, the legislature delegated to DNR much of its authority to ensure that such leasing of state land or interests in land is consistent with the public interest. Alaska Statute 38.05.035(e) provides, in part: "Upon a written finding that the interests of the state will be best served, [DNR] may . . . approve contracts for the sale, lease, or other disposal of available land, resources, property, or interests in them." Kachemak Bay asserts that DNR failed to fulfill this obligation.

"Phasing" is permitted when DNR undertakes best interests findings.

Kachemak Bay argues that DNR "impermissibly `phased' its best interest finding." Thus, a threshold question is whether DNR may phase its best interests finding, an issue addressed in AS 38.05.035(e).

DNR maintains that subsection (1)(C), enacted in 1994, explicitly permits the type of phased review it undertook here. The subsection allows phasing, subject to several conditions:

[F]or a specific proposed disposal of available land, resources, or property, or of an interest in them, [DNR] in the written finding,. . . . may, if the project for which the proposed disposal is sought is a multiphased development, limit the scope of an administrative review and finding for the proposed disposal to the applicable statutes and regulations, facts, and issues identified in (B)(i)-(iii) of this paragraph that pertain solely to a discrete phase of the project when

(i) the only uses to be authorized by the proposed disposal are part of that discrete phase;

(ii) [DNR]'s approval is required before the next phase of the project may proceed; and

(iii) [DNR] describes its reasons for a decision to phase and conditions its approval to ensure that any additional uses or activities proposed for that or any later phase of the project will serve the best interests of the state.

This subsection was enacted by the legislature in reaction to a string of decisions by this court concerning what we characterized as DNR's "phasing" of its review of various mining and gas and oil projects. This line of cases was summed up effectively and concisely in Thane Neighborhood Association v. City and Borough of Juneau. In Thane, citizens' groups challenged the Juneau Planning Commission's approval of a mining company's application to reopen the AJ Mine, arguing that the Borough had "impermissibly used a `phased' approach in approving the permit." We laid out the relevant case law in this way:

Three of our recent cases provide considerable guidance as to what sorts of permit approval "phasing" techniques are appropriate and what kinds are not: Trustees for Alaska v. Gorsuch, 835 P.2d 1239 (Alaska 1992); [ Camden Bay II, and Kuitsarak].

. . . .

We can draw three general, guiding principles concerning when and in what manner "phasing" or "segmentation" is permissible from Gorsuch, Camden Bay II, and Kuitsarak. First, unless a specific statute or regulation allows phasing, phasing is disfavored. Where a statute is silent or ambiguous, phasing should generally not be allowed.

Second, phasing is prohibited if it can result in disregard of the cumulative potential environmental impacts of a project. The more interlinked the components of a project are and the greater the danger that phasing will lead to insufficient consideration of cumulative impacts, the greater the need to bar phasing.

Third, conditions and stipulations may be used to address unforeseen occurrences or unforeseen situations that may arise during exploration or development, but permit conditions may not serve as a substitute for an initial pre-permitting analysis that can be conducted with reasonably obtainable information.

Thus, phasing through the use of conditions is prohibited where it is feasible to obtain the information necessary to determine whether environmental standards will be satisfied before granting an initial permit, but allowed where it is impractical or impossible to create detailed development plans without conducting additional physical exploration.

Applying these principles, we held in Thane that "the Commission should not have granted the AJ Mine permit while excepting major portions of the project . . . [which are] significantly interlinked to other parts of the project. . . . Phasing the approval of those components could . . . cause the cumulative impacts of the mining project to be inadequately considered."

We have not yet had occasion to analyze the effect of the 1994 amendment discussed above on the line of cases culminating in Thane. It is clear that by enacting the amendment the legislature was seeking to allow DNR to phase its approval of projects. However, the legislature did not explicitly overrule any of the cases mentioned above. Thus, we must discern to what extent, if at all, the principles we enunciated in Thane and its predecessors survived the 1994 amendment.

First, the first principle enunciated in Thane, — that phasing is "disfavored" and not allowed unless specifically authorized by statute — did not survive the 1994 amendment. The amendment added AS 38.05.035(e)(1)(C), which speaks in terms of "phases" of a project, and expressly allows DNR to limit its review to the "applicable statutes and regulations, facts and issues . . . that pertain solely to a discrete phase of [a] project" under certain circumstances. Given this new language, it cannot be said that phasing is "disfavored" under Alaska law; on the contrary, the amendment affirmatively empowered DNR to phase its best interests findings if it meets the criteria.

However, the second Thane principle appears to have survived and, indeed, to have been reaffirmed by the 1994 amendment. In Thane, we held that "phasing is prohibited if it can result in disregard of the cumulative potential environmental impacts of a project." The "Legislative Findings" section of the 1994 amendment provides that "consideration of a disposal as a phase of a development project is not intended to artificially divide or segment a proposed development project to avoid thorough review of the project or to avoid consideration of potential future environmental, sociological, or economic effects." Further, subsection (g)(1)(B)(vi), which was substantively unaffected by the 1994 amendment, provides that "when the director prepares a written finding . . . for an oil and gas lease . . . the director shall consider and discuss . . . the reasonably foreseeable cumulative effects of oil and gas exploration, development, production, and transportation on the sale area."

However, the third principle we announced in Thane did not survive the 1994 amendment, or at least did not survive unscathed. Thane held that "phasing through the use of conditions is prohibited where it is feasible to obtain the information necessary to determine whether environmental standards will be satisfied before granting an initial permit." We drew that principle, in part, from Kuitsarak, where we noted that "DNR's argument that it was difficult to obtain the information necessary to perform a proper evaluation of the impacts of mining in the region was undermined by evidence of federal studies similar to the studies which DNR needed to do."

The 1994 amendment seems to have severely limited, if not nullified, this principle. The legislature did so in two ways. First, it added subsection (e)(1)(B)(ii), which provides that DNR "may limit the scope of an administrative review and finding for a proposed disposal to . . . the facts pertaining to the land . . . that the director finds are material to the determination and that are known to the director or knowledge of which is made available to the director during the administrative review". Second, it stated in its findings section that "analyses comparable to those generally required by [the federal National Environmental Policy Act] for the preparation of an environmental impact statement . . . are not required by the state for support of best interest findings."

Thus, under the revised statute, DNR may limit its review to only those facts which (1) it knows at the time it conducts the review and (2) others bring to DNR's attention during the review. Its determinations may not be overturned, as we did in Kuitsarak, for failure to affirmatively seek out additional evidence. Further, a failure on DNR's part to gather the same level of information as a federal agency would in an environmental impact statement is an impermissible ground for finding DNR's determination inadequate — again, this is a repudiation of Kuitsarak.

Finally, the 1994 amendment created an entirely new subsection (h). That subsection provides that DNR may not be required to "speculate about possible future effects subject to future permitting that cannot reasonably be determined until the project or proposed use for which a best interests finding is required is more specifically defined." The subsection goes on to list three specific types of "speculation" in which DNR may not be required to engage.

In sum, the 1994 amendment, at least in a general sense, empowers DNR to phase its best interests findings under AS 38.05.035(g). It is undisputed that DNR did, in fact, phase its review of Sale 85A. The next step in the analysis, then, is to examine whether it was proper, under the new AS 38.05.035(e)(1)(C), for DNR to phase its analysis of this particular project.

2. It was proper for DNR to "phase" its best interests finding regarding Sale 85A.

a. The only uses to be authorized by the lease were part of the lease sale phase.

As this court has noted, "the mere decision to lease does not in itself bring about great risks to the environment." This is so because the lease is no more than an interest in land, and does not in itself authorize any actual "use" of the land. As DNR puts it, "[w]hile the lease gives the lessee the right to conduct these activities, the lease sale itself does not authorize any exploration or development activities by the lessee on leased tracts." (Emphasis added.) Thus, the first part of the AS 38.05.035(e)(1)(C) test is satisfied.

b. DNR's approval was required before the next phase of the project was allowed to proceed.

As DNR discusses in the introduction to its best interests finding, "state approval is required before the next phase (exploration) may proceed." The Sale 85A lease form provides, under the heading PLAN OF OPERATIONS, that "before any operations may be undertaken on the leased area, the lessee shall comply with the applicable statutes and regulations. . . ."

The relevant regulation is the requirement, codified in 11 Alaska Administrative Code 83.158(a), that a lessee must file a DNR-approved "plan of operations" before any operations may be undertaken on the leased area. This plan is required to contain "sufficient information, based on data reasonably available at the time the plan is submitted for approval, for [DNR] to determine the surface use requirements and impacts directly associated with the proposed operations."

Because the lessees must submit these plans before conducting exploration, construction, or production activities on the land, the second part of the AS 38.05.035(e)(1)(C) test is satisfied.

c. DNR described its reasons for its decision to phase and conditioned its approval to ensure that any additional uses or activities proposed for the lease phase or any other phase will serve the best interests of the state.

(1) Reasons to phase

DNR stated in the introduction to the best interests finding that "[i]n oil and gas leasing, it cannot be determined with any specificity or definition at the leasing stage if, where, when, how, or what kind of production might ultimately occur as the result of leasing. . . . A discussion of the possible specific effects of unknown future exploration, development and production activities is not within the scope of this finding." Thus, DNR has described its reasons for phasing its review.

(2) Conditioned approval to ensure that the project continues to serve the state's best interests

DNR states that its best interests finding meets this part of the AS 38.05.035(e)(1)(C) test because DNR "is conditioning this best interests determination and any leases ultimately issued with a number of mitigation measures designed to ensure that any future activities in the exploration, . . . development and production phases will serve the best interests of the state." These measures are found in Chapter 7 of the best interests finding. All Sale 85A lessees must comply with these measures as a condition of the lease.

The measures deal with many subjects: oil and hazardous substance pollution control, use of explosives, road building, water removal, facilities and structures, local hiring, training, public access to the leased land, stream preservation, waste disposal, gravel mining, and wildlife preservation. This detailed list of mitigation measures suffices to ensure that the development of Sale 85A remains in the public interest, as determined by DNR.

d. DNR, by phasing its review, would not avoid thorough review of the project or avoid consideration of potential future environmental, sociological, or economic effects.

While DNR has met all of the factors enumerated in AS 38.05.035(e)(1)(C), our analysis is not finished. As discussed above, in addition to these factors, the statute contains an overarching factor that DNR must consider. That is, whether DNR, by phasing its review, would "avoid thorough review of the project or . . . avoid consideration of potential future environmental, sociological, or economic effects"; or, as we stated in Thane, whether phasing would "result in disregard of the cumulative potential environmental impacts of a project."

Kachemak Bay vigorously asserts that it would, stating:

The threshold decision that the expansion of oil industry infrastructure to the Southern Kenai Peninsula is acceptable is made at the lease sale stage. It is at that stage, therefore, that DNR must consider all relevant factors bearing on that determination. Otherwise, the impacts associated with that decision will be masked until later stages of the process, when it is, as a practical matter, too late to prevent the development.

DNR's best interests finding contains a thirty-page chapter entitled "reasonably foreseeable cumulative effects of Sale 85A." One section contains descriptions of each of the "three phases of industrial activity: exploration, development, and production." The next section "focuses on how these later phases would impact communities and municipalities in the sale area, and the expected distribution of fiscal benefits to the state and local areas." The final section "focuses on physical and non-physical effects on habitats, fish and wildlife, subsistence and historic and cultural resources of the sale region." Given these detailed and comprehensive findings, it does not appear that DNR's decision to phase its review of Sale 85A has caused the impacts of development to be masked.

e. Conclusion

Sale 85A meets the requirements for a phased development under AS 38.05.035(e)(1)(C), and phasing review of Sale 85A does not allow DNR to avoid thorough review of the project or avoid consideration of potential future environmental, sociological, or economic effects. Therefore, DNR's decision to phase was proper.

3. DNR fulfilled its obligation to consider and discuss the relevant factors in its best interests finding.

Having determined that DNR was permitted to phase its review of Sale 85A, the next step is to examine the challenged aspects of DNR's best interests finding and determine whether they meet the AS 38.05.035(g) standards.

a. Transportation methods and risks

Kachemak Bay's first contention is that "DNR failed to discuss the likely oil and gas transportation methods and risks." Analogizing to Demarcation Point and Camden Bay I, Kachemak Bay asserts that "[i]n its [best interests finding] for Sale 85A, DNR discussed only the methods currently used to transport oil and gas in Alaska generally. . . . Thus, DNR again failed to address the methods and risks of transporting oil from the sale area to market." Kachemak Bay cites an environmental impact statement prepared by the federal Minerals Management Service for a 1995 federal lease sale in Cook Inlet to support its claim that it is "simply untrue that DNR lacks the ability to estimate the likely oil transportation modes and impacts until the specific location and nature of oil and gas deposits [are] determined."

As previously discussed, however, the 1994 amendment explicitly rejected the federal environmental impact statement standard as the standard for state best interests findings. Thus, Kachemak Bay's argument — that DNR's failure to conduct an analysis of the likely impacts of oil and gas transportation similar to an environmental impact statement renders the best interests finding insufficient — is unavailing. In any event, the environmental impact statement cited by Kachemak Bay is, in many ways, less complete than the best interests finding submitted by DNR.

Kachemak Bay's more general argument — that DNR did not adequately consider the methods and risks of transporting oil from the sale area to market — is not as easily resolved. Alaska Statute 38.05.035(g)(1)(B)(viii) provides the relevant standard, requiring DNR to "consider and discuss . . . the method or methods most likely to be used to transport oil or gas from the lease sale area, and the advantages, disadvantages, and relative risks of each."

DNR devotes some three pages of its best interests finding to a section entitled "transportation of crude oil and natural gas." DNR first discusses the existing Cook Inlet oil and gas transportation infrastructure generally, and lists the methods used. These methods, according to DNR, include pipelines, both onshore and offshore; marine terminals with offshore loading platforms; and tank vessels. The best interests finding then proceeds to discuss each of these methods in greater detail.

The "advantages and risks" portion of the finding is reserved mainly for the next section of the best interests finding, entitled "oil spill risk, prevention and response." This analysis takes up some three pages of the best interests finding. In that section, DNR discusses in general terms how spills might occur, as well as the oil spill history of Cook Inlet. DNR then discusses in more detail spills that occur as a result of exploration and production, as well as spills that occur as a result of each transportation method and measures to prevent such spills.

Regarding exploration and production, DNR gives statistics for platform spill history and discusses various types of spills that might occur — it also gives examples of several individual Cook Inlet spills and discusses the "worst case discharge" scenario. DNR then discusses pipelines — it reviews the current condition of those in Cook Inlet, as well as the potential amount of oil that might be spilled from a pipeline, how such a spill might occur, and the history of Cook Inlet pipeline spills. Next, DNR discusses marine terminals, their safety records generally, and the volume of oil they handle. In this discussion, DNR briefly addresses three incidents where oil spills occurred at marine terminals in Cook Inlet. Finally, DNR discusses tanker vessels. It states that tankers "present the largest potential for oil pollution" and explains why, as well as giving world-wide oil spill statistics. Finally, DNR discusses a 1987 Cook Inlet spill as well as the infamous 1989 EXXON VALDEZ disaster in Prince William Sound.

Nevertheless, Kachemak Bay argues that "DNR failed to estimate the risk and impacts of a substantial oil spill in the sale area." Again, Kachemak Bay cites the federal Minerals Management Service's environmental impact statement, which contained a detailed analysis of the "likely impacts of a spill in Cook Inlet, taking into account the weather patterns, likely oil fate and transport, and the nature and location of the substantial natural resources at risk in the Inlet." Kachemak Bay asserts that because this analysis was available at the lease sale stage, DNR "may not hide behind a professed lack of ability" to similarly estimate the probability of an oil spill in the context of Sale 85A.

DNR counters that its analysis shows that it "considered the available and known facts [and] used them in its own consideration of the relative risks and foreseeable effects of the likely transportation methods." While DNR concedes that it is "required to consider the relative risks of the likely methods of transporting oil or gas from the sale area," it argues that "[n]either statute nor case law . . . requires DNR to quantify the probability of oil spills in the sale area." DNR further claims that Kachemak Bay "is completely incorrect when it characterizes DNR's consideration of the issue as `hiding behind a professed lack of ability to estimate an oil spill.'" Rather, DNR states that it was "skeptical of [the Minerals Management Service's] spill assessment methodology and, purposefully, decided not to adopt it."

The record shows DNR's original decision-making process regarding the Minerals Management Service's spill analysis in more detail. In a response to a public comment, DNR characterized the analysis as "speculative." DNR asserted that the Minerals Management Service "generates its spill estimate using a computer model which manipulates large numbers of variables, most of which are estimates themselves. Resource estimates are one of the variables in the spill model and are highly speculative." Likewise, in an October 15, 1996 memorandum, a DNR Natural Resource Manager stated that "[a]lthough [the Minerals Management Service] includes an estimate of oil spill risk and impact in [its] [environmental impact statement] for [outer continental shelf] areas, I have serious reservations about the reliability of their estimates." The manager went on to make four specific comments about the speculative nature of the Minerals Management Service oil spill "guess."

DNR's decision to question and ultimately discard as "unhelpful" Minerals Management Service's methodology is entirely proper. DNR is correct in its assertion that it need not "use the MMS methodology to estimate oil spills." The 1994 amendments explicitly exempted DNR from being required to conduct analyses similar to environmental impact statements.

Kachemak Bay is correct, however, that DNR does not discuss "how any oil or gas will likely be transported from the remote areas of the sale area." In fact, DNR expressly declined to discuss exactly what "transportation strategies" would be used in the area:

A discussion of specific transportation alternatives for oil from the Sale 85A area is not possible at this time because strategies used to transport potential petroleum reserves depend on many factors, most of which are unique to an individual tract and discovery. The location and nature of oil and gas deposits determine the type and extent of facilities necessary to develop and transport the resource. No oil or gas may be transported from Sale 85A leases until the operator has obtained the necessary permits and authorizations. . . . [DNR] and other state, federal, and local agencies will review the specific transportation system when it is actually proposed.

Kachemak Bay argues that by failing to address "the key question of how any oil or gas will likely be transported from the remote portions of the sale area," DNR falls short of the subsection (g) standard relating to transportation.

Again, Kachemak Bay analogizes to Camden Bay I and Demarcation Point, where we remanded best interests findings that failed to discuss and weigh the risks of the likely transportation methods to be used. Kachemak Bay asserts that

[t]he forty to sixty mile distance between some portions of Sale 85A and any existing oil and gas industry infrastructure is reminiscent of the fifty to eighty mile distance in the Demarcation Point and Camden Bay I cases. As in those cases, the transportation of oil from remote areas devoid of existing infrastructure — e.g., via a long-distance pipeline on the ocean bottom or across land — entails potentially significant risks and impacts which DNR must consider in its best interest finding.

DNR correctly points out that both of these cases are distinguishable from the instant case. Camden Bay I involved a potential oil field off Alaska's northern coast, west of Kaktovik and north of the Arctic National Wildlife Refuge (ANWR). At the time DNR issued the best interests finding, Congress was deciding whether to open ANWR to oil and gas leasing and associated uses. We remanded the best interests finding for a supplemental written finding regarding transportation issues because DNR had failed to "consider the unique risks presented by the oil transportation methods that would be necessary if the legal status of ANWR remains unchanged." Demarcation Point involved a nearly identical situation. The potential oil field was in the same geographic region, abutting ANWR, though further east than Camden Bay. Relying on Camden Bay I, we held that the best interests finding was deficient for failure to address the issue of transportation if ANWR did not become available. We noted that the best interests finding in that case was "remarkably similar to the . . . Finding that was rejected by this court in Camden Bay [I]. . . . [It] deal[t] with transportation issues in a similar cursory manner." Thus, we remanded the best interests finding for a supplemental finding on the issue. Camden Bay I and Demarcation Point are distinguishable from this case in that the best interests findings in those cases did not even address how oil would be transported if ANWR's status was not changed. DNR's transportation analysis here is not similarly lacking. As discussed above, DNR analyzes the various methods of transportation and their risks in some detail, and no such major policy change, beyond DNR's control, is necessary before oil and gas can be transported in Cook Inlet.

Kachemak Bay argues for a broader application of Camden Bay I and Demarcation Point; it would have us hold that "[w]hile DNR may flesh out the size, nature and location of the specific facilities at a later stage, it must identify at the lease sale stage the most likely methods of moving oil from the sale area to market, and the associated impacts on the sale area and affected communities." Specifically, Kachemak Bay urges us to remand the best interests finding for an alleged failure to address transportation of oil "from remote areas [of the sale area] devoid of existing infrastructure."

However, statutory and case law do not support this approach. As mentioned above, the 1994 amendment added AS 38.05.035(h)(1), which provides that in best interests findings DNR "may not be required to speculate about . . . the exact location and size of an ultimate use and related facilities." Further, as we held in Camden Bay II, "[u]ntil exploration is proposed and, in all likelihood, until and unless a commercially exploitable discovery is made, there will be no occasion for siting, designing or constructing transportation and utility routes." Thus, DNR's relatively general approach to the issue of the methods and risks of transporting oil and gas in Cook Inlet satisfies the now-more-generous AS 38.05.035(g) standard. Kachemak Bay's assertion that DNR needs to specify the means and risks of transporting oil and gas from the most remote parts of the Sale 85A area is asking DNR to engage in what the legislature has deemed improper "speculation."

In short, DNR has met the subsection (g) standard. Kachemak Bay's argument that DNR "failed to identify and discuss the methods and risks of transporting oil and gas from the Sale 85A area to market" is simply not supported by the record.

b. Sociological impacts on sale area and affected communities

Kachemak Bay argues briefly that "the addition of industry infrastructure requires a labor pool, necessary support companies and services, and associated demands on public services, which will impact nearby communities. . . . DNR failed to meaningfully respond to public comments and to discuss the sale's reasonably foreseeable impacts on the sale area and affected communities."

Kachemak Bay is correct that DNR is obliged to "consider and discuss . . . the reasonably foreseeable effects of oil and gas exploration, development, production, and transportation on municipalities and communities within or adjacent to the lease sale area" in its best interests finding. DNR's discussion of these issues is found in a three-page subsection of the best interests finding entitled "effects on municipalities and communities."

This subsection first lists the affected communities and refers broadly to possible changes in "education level, household or per capita income, occupancy or rental rates, population density or expansion of a particular age bracket," but states that "due to intervening factors, such as overall population growth in the sale area, such changes are not measurable or predictable." DNR goes on to state that "[l]ong[-]term effects of Sale 85A could mean a change in employment opportunity, influx of cash, or increase in demand for services, like sanitation, police protection, or road maintenance."

In a similar fashion, DNR lists several possible "[p]hysical effects of industry activities," including "vegetation loss, siltation, sedimentation, water quality changes, noise, increases in human congestion, or disturbances to wildlife." Again, however, DNR does not make specific predictions as to the exact location or degree of these effects:

Effects related to physical alterations of the environment would be directly related to the number of exploration, delineation, and production drill sites, and those numbers depend on the size, extent, location, and recoverability of discovered reserves. The siting of possible drill sites depends on factors including ecology, costs, the presence of an existing road system, and land ownership and management. The extent of effects would also be related to the proximity of development sites to important habitats. . . .

DNR then refers the reader to chapter six of the best interests finding for a discussion of water quality. In that discussion, DNR describes current instances of groundwater contamination, methods of disposal of contaminated water, and the potential for drawdown of local groundwater tables. DNR also points the reader to chapter six for its discussion of oil spill "risk, preparedness, response and clean-up."

DNR goes on to discuss the possibility that oil and gas projects will require "new or expanded utilit[ies]," including electricity and water. DNR also mentions the potential for "increased use of transportation systems, such as air charter services, air strips, docks or roads." Once again, however, DNR disavows its ability to predict more precisely the probability or extent of these needs and uses, as they would "depend on the specific project proposed, its location, and the existing supply and demand for the service."

DNR next addresses land use, beginning with a discussion of the communities likely to be affected by development. DNR then mentions the possibility that uses such as hunting, fishing, and trapping might be adversely affected by development, and that increased "human presence" on private property made more accessible by development might have "negative impacts on traditional and recreational use."

Finally, DNR addresses the issue of employment. It discusses some of the extant gas fields and describes their impact on employment as "minimal." DNR states that Sale 85A "may create new employment opportunities," but adds that the "long-term employment benefits of this sale on the [Kenai Peninsula Borough, the Municipality of Anchorage,] and local communities will depend on the subsequent production of commercial quantities of petroleum." DNR claims that "[a]n influx of workers from outside Alaska is unlikely," because "[a]s existing Cook Inlet fields decline, more and more of the current resident labor pool and service industry will be in need of employment." DNR states that for those positions which the local labor pool is unable to fill, contractors will be encouraged to employ Alaskans.

As part of the best interests finding, DNR also catalogues and responds to the voluminous public comment that its announcement of the proposed lease sale generated.

As this summary shows, DNR is essentially correct when it says that Kachemak Bay "may not agree with DNR's conclusions, but it has simply not established that DNR failed to consider and discuss material information or that DNR failed to take a hard look at the reasonably foreseeable effects of post-lease sale activities on communities." DNR did consider these impacts in a manner that meets the relatively forgiving standard of AS 38.05.035(g)(1)(B)(x). Moreover, pursuant to AS 38.05.035(h), added by the 1994 amendment, DNR need not engage in "speculation" regarding "possible future effects subject to future permitting that cannot be reasonably determined until the project . . . is more specifically defined." Kachemak Bay has not borne its burden of showing that DNR's decision was arbitrary, unreasonable, or an abuse of discretion.

c. Conclusion

The best interests finding is affirmed in all respects.

B. The Consistency Determination

Besides being obliged under Title 38 to determine whether lease sales are in the state's best interests, DNR is also required, under Title 46, Chapter 40 (the Alaska Coastal Mamanagement Plan), as implemented by Title 6, Chapter 80 of the Alaska Administrative Code, to determine that the lease sale is consistent with (1) the Alaska Coastal Management Plan and (2) district coastal management plans. Kachemak Bay asserts that DNR's consistency determination regarding Sale 85A was erroneous in both respects, and that DNR improperly phased its consistency determination regarding the latter. As discussed below, however, these arguments fail.

1. Consistency with the habitats standard

a. DNR's determination that Sale 85A was consistent with the Alaska Coastal Management Plan habitats standard was not erroneous.

We discussed the Alaska Coastal Management Plan's habitats standard in Camden Bay II, notings its stringent requirements:

The ACMP has, among its objectives, protecting numerous environmental and cultural values in Alaska's coastal zone. As we have elsewhere had occasion to note, the ACMP's standards are extremely protective of the environment. Offshore areas are among the habitats subject to the Alaska coastal management program. Such habitats "must be managed so as to maintain or enhance the biological, physical, and chemical characteristics of the habitat which contribute to its capacity to support living resources." 6 AAC 80.130(b) (emphasis added). Uses or activities that fail to maintain or enhance the habitat's capacity to support living resources may be authorized only if several stringent additional conditions are met.

Kachemak Bay asserts that DNR's consistency determination violates 6 AAC 80.130, commonly known as the "habitats standard," because Sale 85A neither (1) maintains or enhances the relevant habitat nor (2) meets the three "stringent additional conditions" that the habitats standard specifies.

(1) Sale 85A does not "maintain or enhance" the habitat.

It is undisputed that the activities that will and may occur on the leased tracts — i.e., exploration for and production of oil or gas — "do not maintain or enhance the coastal habitat." Therefore, DNR was required to show that the proposal met the three subsection (d) conditions.

(2) DNR showed that Sale 85A meets the subsection (d) requirements.

(a) DNR showed a "significant public need" for Sale 85A.

In its consistency determination, DNR found that there was a significant public need for Sale 85A. DNR listed several aspects of the need, including: (1) "increased revenues from[:] oil and gas lease bonus bids, rentals, and royalties; production taxes; and corporate income taxes"; (2) the "need for a continuous supply of energy"; (3) "the public need for continuing economic stability," based on the state's dependence on revenue and employment derived from oil and gas exploration and development; and (4) the "national need to replace domestic [petroleum] reserves to maintain future production levels and prevent a further deterioration of the country's position in the world market."

Despite this litany, Kachemak Bay argues that DNR "has not demonstrated any significant public need to offer Sale 85A." Kachemak Bay contends that the amount of money that the state will receive from the lease sale itself will "not materially affect total state revenues." Further, Kachemak Bay argues, "DNR did not even estimate potential revenues to be derived from subsequent exploration, development, and production, and its reliance on this phantom figure as a `significant need' is arbitrary and capricious." Kachemak Bay argues that DNR was required to estimate the "Sale 85A-related income," and points to the Minerals Management Service's estimate for Sale 149 as proving that such estimates are feasible. In any event, Kachemak Bay contends, "any revenues derived from Sale 85A are highly unlikely to be `significant' in terms of their impact on existing revenues." Finally, Kachemak Bay makes the broad argument that

[i]f a "need" for any revenue which may flow from an oil and gas lease sale, standing alone, constitutes a "significant public need," then there would be in effect an unwritten exception to the Habitats Standard for oil and gas lease sales, since such sales always [would] be expected to generate revenue. . . . While the estimated revenue from a given lease sale might potentially constitute a significant public need, that is not the case here, where DNR has refused to estimate revenues, and [has] identified no reasonable possibility that the Sale 85A revenues will significantly impact total state oil and gas revenues.

These arguments are unavailing; we considered and rejected similar arguments in Ninilchik. In that case, DNR adverted to benefits that closely resemble those advanced here: namely, revenue for the state, revenue for local governmental bodies, employment opportunities, and the "long-range goal of the State of utilizing the oil and gas lease program to provide the basis for a stable and prosperous economy." Citing our holding in Camden Bay I, we held that DNR's analysis sufficed to show that DNR's determination that there was a significant public need for the sale was not arbitrary or capricious.

The determination that a significant public need for a lease sale exists is exactly the type of "policy decision, involving complex issues that are beyond this court's ability to decide," to which we give considerable deference. Without evidence that this decision was arbitrary or capricious, we cannot negate this policy decision by DNR. We accordingly decline to reverse the consistency determination on these grounds.

(b) DNR showed that there is no feasible and prudent alternative to meet the public need for the proposed use or activity which would maintain or enhance the habitat.

Kachemak Bay argues that "[t]here were many potentially feasible and prudent alternatives to offering Sale 85A as it was offered." The alternatives proposed by Kachemak Bay include deferring or deleting some portions of the sale area, and evaluating the use of alternative energy resources and conservation measures to reduce the United States' national demand for oil.

In its consistency determination, DNR found that there was "no feasible and prudent alternative to offering lands in the coastal zone that would meet the public need for Sale 85A." DNR found that because "[i]ndustry interest lies in the coastal zone," it would not be prudent to offer lands outside the coastal zone. If the state offered lands where there was a lack of industry interest, DNR found, "this state would receive no economic benefit from holding this sale."

Examination of the record reveals that DNR's decision that there were no prudent and feasible alternatives to Sale 85A was not arbitrary nor capricious. In fact, as DNR points out, the sale area was pared down by 40,000 acres in direct response to concerns of residents of Homer. This indicates that DNR took the requisite "hard look" at the sale area before issuing its consistency determination.

Further, deleting part of the sale area would potentially reduce or eliminate all of the benefits listed above. In Ninilchik, we rejected such deletion as a feasible and prudent alternative for just that reason: "The State maintains that while it might be feasible to offer less promising areas for lease, it would not be prudent because potential lessees have expressed an interest in the Sale 78 area." Obviously, forgoing offering the lease sale altogether (as Kachemak Bay apparently proposes in favor of development of alternative energy sources) would result in complete loss of the benefits cited by DNR; Kachemak Bay has not shown that its proposals regarding alternative energy sources would provide those benefits.

b. Conclusion

We affirm DNR's determination that Sale 85A was consistent with the habitats standard.

2. Consistency with the Kenai Peninsula Borough Coastal Management Plan

Kachemak Bay further alleges that DNR failed to fulfill its duty, pursuant to AS 46.40.100(a), to "administer land and water use regulations or controls in conformity with district coastal management programs." The relevant district program here is the Kenai Peninsula Borough Coastal Management Plan — specifically, Policy 2.7.

Policy 2.7 is succinct. It provides: "The cumulative effects of proposed new and existing development on ambient air and water quality and coastal habitats shall be considered in the review or renewal of coastal projects."

a. It was proper for DNR to "phase" its determination that Sale 85A was consistent with the Kenai Peninsula Borough Coastal Management Plan.

DNR's revised consistency determination begins with a caveat:

State leases alone do not authorize any activities that would impact air and water quality. Focused consideration of this requirement is reserved until a specific activity is proposed. Until a development, use or activity which might actually impact air and water quality is proposed, DNR will have no specific information about its impact on future air and water quality.

In short, DNR is indicating that it will phase its determination that Sale 85A conforms with Policy 2.7.

The same line of case law discussed above in the best interests finding context is relevant to the question of whether DNR may phase its consistency determinations; those cases apply to phased permitting generally. The 1994 amendment discussed above is also relevant; in that amendment, the legislature added AS 46.40.094, entitled "Consistency Determinations for Phased Uses and Activities." That section permits DNR to phase its consistency determinations if:

(1) at the time the proposed use or activity is initiated, there is insufficient information to evaluate and render a consistency determination for the entirety of the proposed use or activity;

(2) the proposed use or activity is capable of proceeding in discrete phases based upon developing information obtained in the course of a phase; and

(3) each subsequent phase of the proposed use or activity is subject to discretion to implement alternative decisions based upon the developing information.

In essence, these conditions determine whether a given project is eligible for a phased consistency review. The statute further provides that if these conditions are met, DNR may

limit the consistency review to [a] particular phase if, but only if,

(A) the agency or another state agency must carry out a subsequent consistency review and make a consistency determination before a later phase may proceed; and

(B) the agency responsible conditions its consistency determination for that phase on a requirement that a use or activity authorized in a subsequent phase be consistent with the [Alaska Coastal Management Plan].

As is the case with best interests findings, then, it can no longer be said that phasing is disfavored under Alaska law when DNR makes a conclusive consistency determination. Neither can a consistency determination be found wanting for DNR's failure to conduct additional studies; the material DNR must consider is limited to that already at hand and that presented to DNR.

Further, the "findings" section of the amendment applies to consistency determinations as well as best interests findings. Therefore, DNR is not required to undertake analyses comparable to federal environmental impact statements in its consistency determinations. As discussed above, another finding provides that DNR may not avoid thorough review of the project or its potential effects by artificially dividing the project.

Just as with the best interests finding, then, the threshold question here is whether, under the Alaska Coastal Management Plan as amended, DNR may phase its conclusive consistency determination.

(1) Sale 85A is eligible for phasing.

Kachemak Bay asserts that DNR "ignored virtually all of the most relevant and readily available information pertaining to the impacts of existing development in the sale area. Further, as to future impacts, DNR did not provide any meaningful discussion or estimates, and instead asserted its inability to forecast such impacts until a later stage." Kachemak Bay states that DNR may not phase its consideration of the effects of existing development at all, because those impacts are "quantified and accessible." Further, Kachemak Bay asserts, "impacts from new development, including those from activities associated with Sale 85A, are foreseeable and capable of estimation."

Kachemak Bay's characterization of the scope of the review required by Policy 2.7 is somewhat misleading. The policy does not call for piecemeal review of (1) the impacts of existing development, (2) the impacts of proposed new development, and (3) the cumulative effects of (1) and (2). Rather, it calls for DNR to consider the "cumulative effects of proposed new and existing development" — what might be called a holistic approach. This approach comports more closely with the Kenai Peninsula Borough Coastal Management Plan's "goals and objectives" — i.e., "[t]o protect important fish and wildlife habitat areas and environmentally sensitive areas from incompatible development" — than a piecemeal analysis would.

DNR asserts that it "carefully considered, to the extent reasonable and practical at the time of the consistency determination, the cumulative effects of existing and proposed new development on ambient air and water quality." However, DNR stated that it was unable to predict the effects of development on air and water quality. "Focused consideration of this requirement is reserved until a specific activity is proposed. Until a development, use or activity which might actually impact air and water quality is proposed, DNR will have no specific information about its impact on future air and water quality."

Again, the 1994 amendments endorse DNR's approach to this issue. As DNR states, "leases alone do not authorize any activities that would impact ambient air and water." The future activities that would have such impacts — exploration and development — are "subject to independent permitting requirements," as previously discussed. Pursuant to the 1994 amendments, DNR may not be required to "speculate" concerning the effects of these activities. For these reasons, DNR's finding that there was insufficient information to render a determination regarding the entire project was proper.

(2) The proposed petroleum development project was capable of proceeding in discrete phases based upon developing information obtained in the course of a phase.

DNR states that "[e]ach phase of oil and gas activities builds on the previous phase. That is, exploration cannot take place before the lease sale. Depending on the information gathered at the exploration phase, development may or may not take place." This analysis seems correct — in any event, Kachemak Bay does not contest this characterization of oil and gas activities. Accordingly, this element of the AS 46.40.094(a) test is met.

(3) Each subsequent phase of the proposed petroleum development project is subject to discretion to implement alternative decisions based upon the developing information.

DNR states:

At each phase, proposed site-specific and project-specific activities will be analyzed for consistency based on information developed at the previous stages. Further, each of these subsequent phases [-] exploration and development/production [-] is subject to discretion to implement alternative decisions based upon this developing information. During the exploration and development phases, lessees submit proposed plans of operation for permit approval. At that time, agency staff reviews information developed from the previous phases along with new technological developments and site-specific data, and implements any needed alternative mitigation measures when determining whether the permit request complies with the [Alaska Coastal Management Plan].

Assuming that this characterization of DNR's internal policies is accurate — and Kachemak Bay does not dispute it — the third part of the AS 46.40.094(a) test is met as well.

Therefore, Sale 85A is eligible for a phased review under AS 46.40.094.

b. DNR may limit its consistency review to the lease sale stage.

As mentioned above, once a project is found eligible for a phased review, AS 46.40.094(b)(1) contains two more conditions that must be met before DNR is allowed to limit its consistency review to a given phase.

(1) DNR must carry out a subsequent consistency review and make a consistency determination before the next phase may proceed.

DNR states:

Each lessee who proposes to conduct activities related to exploration and development of the proposed Sale 85A area will be required to submit a plan of operations to [DNR] for review and approval . . . [which] must contain sufficient information based on data reasonably available at the time to determine the surface use requirements and impacts directly associated with the proposed operations.

This analysis is supported by 11 AAC 83.158(d), and Kachemak Bay does not suggest otherwise. Accordingly, this element of the AS 46.40.094(b)(1) test is met.

(2) DNR conditioned its consistency determination on a requirement that future uses or activities be consistent with the Alaska Coastal Management Plan.

DNR states that "[a]s a condition for consistency approval of the lease operations, [DNR] will require such modifications as may be necessary to ensure consistency with the Alaska Coastal Management Plan. Measures in addition to . . . the lease sale mitigation measures may be added to address site[-] specific resource values and activities directly associated with the proposed project." Again, Kachemak Bay does not dispute that this is so. Therefore, this element of the AS 46.40.094(b)(1) test is met as well.

c. Conclusion

Under AS 46.40.094, it was proper for DNR to phase its consistency review.

3. DNR's decision that Sale 85A was consistent with the Kenai Peninsula Borough Coastal Management Plan was not erroneous.

That DNR may phase its consistency determination does not, however, relieve it of its duty to make its determination for the lease sale phase that Sale 85A is consistent with Policy 2.7. Alaska Statute 46.40.094(b)(2) provides that DNR is to conduct the review based on (1) applicable statutes and regulations; (2) material facts that are known to DNR or are made known during the consistency review; and (3) the "reasonably foreseeable, significant effects of the use or activity for which the consistency determination is sought."

Kachemak Bay challenges DNR's analysis of the second and third of these elements, both as to air and water quality. It argues that DNR's revised conclusive consistency determination fails "to even mention the existing impacts due to current development in the sale area, in blatant contravention of [Kenai Peninsula Borough Coastal Management Plan] Policy 2.7. Further, the revised [conclusive consistency determination] . . . fails to employ available data to estimate future impacts."

a. Air quality

In its conclusive consistency determination, DNR discusses, in general terms, the likely sources of air pollution that would result from exploration, development, and production in the sale area. DNR lists "routine" activities, which it states would have only a temporary effect which would diminish after construction was complete. It then discusses the possibility of gas and oil leaks and spills, but concludes that either would have minimal impact. DNR concludes that the impact of possible emissions during tanker loading operations would be "minimal" as well.

Regarding present development, DNR briefly states that "overall air quality in Cook Inlet is good. . . . There may be specific locations that exceed air quality standards, but the impacted areas are restricted in size due to meteorological air patterns. The Municipality of Anchorage and Tesoro's oil refinery are two examples."

Kachemak Bay argues that DNR should have considered other sources of information in its air quality analysis. For instance, Kachemak Bay argues that DNR should have examined information about air pollution that is "readily available" pursuant to the Clean Air Act and from the federal Environmental Protection Agency's Toxics Release Inventory.

b. Water quality

DNR's findings regarding water quality are more detailed than its findings regarding air quality. In the revised conclusive consistency determination, DNR lists the ways that Sale 85A may affect water quality; namely, discharges of drilling muds and cuttings. It also discusses how discharges from platforms might affect Cook Inlet. DNR mentions a 1997 report that characterizes Cook Inlet as a whole as "healthy," despite some areas that had been disproportionately affected by human pollution.

As mentioned above, DNR also discusses Sale 85A's potential effects on water quality in its best interests finding, which DNR incorporated by reference in the revised conclusive consistency determination. DNR devotes some five pages to the issue and provides a schematic of a Cook Inlet wellbore.

Kachemak Bay makes similar arguments regarding the insufficiency of DNR's water quality findings as it does regarding DNR's air quality findings; these arguments boil down to Kachemak Bay's dissatisfaction with the information that DNR considered. For example, Kachemak Bay discusses a subdivision in Soldotna which was apparently subject to oil field pollution in its drinking water. Kachemak Bay also refers to publicly available information that DNR could have obtained pursuant to the Clean Water Act.

c. Conclusion

DNR's analyses are not exhaustive, nor do they take into account all of the information that Kachemak Bay discusses. However, the analyses do show that DNR "considered" the cumulative effect of the lease sales on the air and water quality in the district. These findings, combined with the fact that the Borough itself concurred with DNR's determination, suffice to show that DNR's determination was not arbitrary or capricious.

Further, as discussed above, the 1994 amendment discharged DNR from a duty to consider information other than that "known to [DNR] or made a part of the record during the consistency review." There is no basis in the record to conclude that the information Kachemak Bay would have DNR consider fits either of these categories. Therefore, DNR's conclusive consistency determination cannot be overturned for failure to consider such information.

C. Additional Issues

1. Intervenors' arguments

At the superior court, several petroleum companies who had successfully bid on tracts offered in Sale 85A moved to intervene in this action; the court granted these motions. Cook Inlet Regional Corporation, Inc. (CIRI), a Native corporation with lands in the vicinity of the Sale 85A area, also successfully moved to intervene. The Alaska Mental Health Trust did likewise; title to some of the land in the Sale 85A area is held by the Trust.

The brief jointly submitted by the oil companies and CIRI mounts a defense of the substance of DNR's best interests finding and conclusive consistency determination — a defense ably mounted by DNR itself. Because it does not raise separate issues, we have considered its arguments when we considered DNR's arguments.

The Trust argues separately that Sale 85A was in the best interests of the Trust. That claim is not challenged by Kachemak Bay. In fact, this issue was not even raised by Kachemak Bay in its points on appeal. Therefore, we need not consider the issue.

2. The "juggernaut" argument

At several points in its brief and with slight variations, Kachemak Bay argues that "the lease sale stage marks the best, and, as a practical matter, the only time to plan for the impacts of the subsequent development." This is so, Kachemak Bay contends, because if the state cancels a lease after it has been sold, it must compensate the lessee. Because of this requirement, Kachemak Bay asserts, "[a] lease cancellation would undoubtably be a painful, disruptive, and expensive proposition, and is patently not a transaction in which any state would care to engage. The lease sale is thus a pivotal threshold event, after which point there is substantial momentum toward developing the leased tracts." In short, Kachemak Bay argues that DNR's failure to evaluate, at the lease-sale stage, all of the potential impacts of the development creates an unstoppable juggernaut of a project. The implication is that the state will be unwilling to cancel the leases, no matter how dire the environmental consequences, because of the financial burden of doing so.

The "sample lease" that DNR generated for Sale 85A does indeed contain a provision requiring the state to reimburse lessees if the state cancels a lease. And Kachemak Bay's argument holds much intuitive appeal. It does seem possible that the state may unwisely decide to continue to allow a project which is causing untoward environmental damage if the state stands to lose a significant amount of revenue by canceling a lease. In fact, we have discussed this problem in several of our "phasing" cases.

However, Kachemak Bay's argument really goes to the question of whether phasing should be allowed at all — a question which was definitively answered by the legislature in the 1994 act amending Title 38 and the Alaska Coastal Management Plan. Within the strictures specified by the legislature, phasing is now expressly allowed. It is not for us to overturn that policy choice.

We note, however, that the legislature's policy choice does not, by any means, relieve DNR of its duty to take a continuing "hard look" at future development on the lease sale lands. To the contrary, DNR is obliged, at each phase of development, to issue a best interests finding and a conclusive consistency determination relating to that phase before the proposed development may proceed.

V. CONCLUSION

Because DNR has not impermissibly phased its review of the proposed lease sale, and because there is a reasonable basis for both DNR's best interests finding and conclusive consistency determination, the agency's actions are AFFIRMED.