Opinion for the Court filed by Circuit Judge MIKVA.
Opinion concurring in part and dissenting in part filed by Senior Circuit Judge MacKINNON.
MIKVA, Circuit Judge:At issue in this case is the Federal Energy Regulatory Commission’s conditional approval of a proposal to construct a liquified natural gas (“LNG”) terminal near Point Conception, California. FERC approved the project in 1979, but this court remanded the matter to the Commission in 1980 for further consideration of seismic data. In 1983, following new evidentiary hearings, FERC again found the site “seismically suitable.” Petitioners, all of whom live or own property near Point Conception, challenge the merits of that determination and also contend that the 1979 and 1983 orders should be vacated as moot because of alleged abandonment of the proposal by its original backers.
We agree that the 1979 order must be vacated as moot, and we vacate the 1983 order on different grounds.
I.
In the late 1970’s, a group of companies applied to FERC and to the California Public Utilities Commission (“CPUC”) for permission to construct and operate an LNG terminal at Little Cojo Bay, near Point Conception, California. The terminal was to receive and regassify LNG shipped from Alaska and Indonesia.
Concerns were raised about seismic risks at the proposed site from the very beginning. In April 1978, a geologist hired by the Hollister Ranch Owners’ Association (“Hollister”) found an active earthquake fault on the site, and several other on-site active faults have been discovered since then. In 1978, the CPUC nonetheless approved the project but conditioned its approval on further seismic studies to be conducted by the applicants. Adopting the recommendations of its Administrative Law Judge (“ALJ”), FERC did the same in 1979, despite the position of the Commission’s staff that no LNG terminal should be approved on any site containing an active earthquake fault. The staff adhered to this position throughout the later proceedings on remand. See Pacific Alaska LNG Co., 19 FERC Rep. (CCH) ¶ 63,086, at 65,270 (June 23, 1982).
The staff reasoned that “[t]here is such uncertainty in the phenomenon of surface faulting that the current state-of-the-art is inadequate to allow the safe design of an LNG facility near an active fault.” Pacific *174Alaska LNG Co., 8 FERC Rep. (CCH) ¶ 63,032, at 65,365 (Aug. 13, 1979). The ALJ rejected this “per se or absolutist” position as unacceptable, based in part on his impression of the need for an LNG terminal in California. He found that there was “a pressing need” for the gas volumes at issue, and that there were “no available alternatives” to the LNG proposal. Id. at 65,319. In connection with the staff’s argument that the mere presence of an on-site earthquake fault should disqualify Little Cojo Bay as an acceptable site for an LNG facility, the ALJ noted:
... Staff’s position is based on very remote possibilities and a refusal to consider the geoseismic particulars of a given site location. This is especially disturbing since the entire coast of California is seismically active and Staff’s per se approach threatens to impose on the state of California a no-growth energy policy respecting LNG facilities directly contrary to the interest of the citizens of that state as expressed in the LNG Terminal Act.
Id. at 65,346.
Nevertheless, the AU conditioned his approval of the project on the applicants’ completion of further seismic investigations at the site. He also conditioned his approval on prompt completion of the project: gas sales from the facility had to commence by February 15, 1983. See id. at 65,391. FERC affirmed the ALJ’s conditional approval and adopted his decision without relevant changes. See Pacific Alaska LNG Co., 9 FERC Rep. (CCH) ¶ 61,041 (Oct. 12, 1979).
Several parties, including the current petitioners, sought review in this court of FERC’s 1979 approval of the Point Conception proposal. While the case was pending, additional information became available on the seismic condition of the Little Cojo site. In April 1980, the court granted Hollister’s motion to remand the matter to FERC to allow the Commission to consider the new information in the first instance.
In response to the remand, FERC ordered new evidentiary hearings on the seismic issues. By this point in the proceedings, however, some participants had dropped out of the project, and the applicants had been forced to give up their rights to a dedicated supply of Indonesian gas. The project sponsors nonetheless “remain[ed] committed” to the project, and the Commission therefore decided to proceed with consideration of Little Cojo’s seismic suitability:
The seismic issues at the Little Cojo site ... fall exclusively within the Commission’s jurisdiction, are ripe for decision now, and will entail a significant measure of time for hearing and analysis. Accordingly, ... the Commission will defer consideration of nonseismic issues at this time, pending resolution of the seismic issues by our Commission and the CPUC....
Pacific Alaska LNG Co., Nos. CP75-140 et al., slip op. at 4 (March 2, 1981).
After hearing new evidence, the ALJ determined that “the remand record does not warrant a change in the conclusion of the prior Initial Decision that the potential seismic hazards at the site do not present unacceptable risks to public safety or project reliability.” Pacific Alaska LNG Co., 19 FERC Rep. (CCH) ¶ 63,086, at 65,272 (June 23, 1982). He therefore affirmed the 1979 Initial Decision “to the effect that the Little Cojo Bay site is seismically suitable for the safe and reliable construction and operation of an LNG receiving terminal.” Id. at 65,288.
The Commission affirmed the ALJ’s decision. Pacific Alaska LNG Co., 25 FERC Rep. ¶ 61,005 (Oct. 4, 1983). In response to Hollister’s argument that the proceedings should be dismissed as moot because the industrial sponsors were no longer in a position to go forward, FERC noted that the sponsors might choose to proceed at some time in the future, and concluded that a resolution of the seismic issues could therefore prove useful:
If we find Little Cojo to be seismically suitable, they can proceed with the projects when they are needed in the future; if we find the site to be unsuita*175ble, further efforts to preserve the projects would not be taken____ [A]ll we can say with any certainty now is that the future is uncertain. At some point, LNG may be needed to meet the requirements of gas consumers in California.
Id. at 61,034.
II.
Intervening in this case, the People of the State of California and the California Public Utilities Commission contend that FERC’s 1979 and 1983 orders are not yet ripe for review because they are merely intermediate steps in an ongoing administrative process. In essence, California argues that petitioners should have to wait until a viable . LNG project is actually approved before challenging the 1979 and 1983 orders.
Even assuming that the collateral challenge envisioned by California would not be time-barred, we do not agree that review at this time would be premature. Cf. Baltimore Gas & Electric Co. v. NRDC, 462 U.S. 87, 103 S.Ct. 2246, 76 L.Ed.2d 437 (1983) (reviewing NRC regulations setting environmental cost estimates to be used in nuclear power plant licensing). The 1983 order “terminated” the proceedings; it appears that, as far as FERC is concerned, the ball is now in the project applicants’ court. If the applicants never decide to go forward — a possibility the record suggests is hardly remote — California’s view of ripeness would prevent the petitioners from ever being able to challenge the 1983 order. Petitioners would still be aggrieved by the order, though; they claim credibly that the value of their property is diminished by FERC’s official, purportedly final declaration that a neighboring site is “seismically suitable” for an LNG terminal.
Our conclusion that petitioners’ challenges are now ripe for judicial consideration is buttressed by FERC’s conspicuous failure to raise a ripeness or exhaustion objection. The judicial policy against unwarranted and premature interference with continuing agency decision-making — the basis for California’s ripeness argument— is more typically expressed in terms of the exhaustion requirement; indeed, California itself argues that what is missing in this case is “[ejxhaustion of the administrative process.” Brief for Intervenors the People of the State of California and the Public Utilities Commission of the State of California at 13. This court has observed in the past that an agency’s failure to make such an argument for itself may indicate that the review sought would not in fact unduly disrupt agency decisionmaking:
The agency’s failure to insist upon exhaustion ... suggests that, in [the agency’s] view, the costs — in terms of interfering with agency proceedings, creating an incentive to ignore agency processes in the future, and depriving the agency of a greater opportunity to apply its expertise or of a chance to “correct its own errors” — do not outweigh the benefits of this court’s immediate consideration of [the] petition.
Asarco, Inc. v. EPA, 578 F.2d 319, 321 n. 1 (D.C.Cir.1978). In this case, FERC may well have decided not to argue ripeness or exhaustion for the sensible reason that requiring judicial review to be sought later instead of now would preserve much of the uncertainty the order attempted to remove.
III.
In United States v. Munsingwear, 340 U.S. 36, 71 S.Ct. 104, 95 L.Ed. 36 (1950), the Supreme Court held that lower court decisions should be vacated if they become moot pending appellate review.
That procedure clears the path for future relitigation of the issue between the parties and eliminates a judgment, review of which was prevented through happenstance. When that procedure is followed, the rights of all parties are preserved; none is prejudiced by a decision which in the statutory scheme was only preliminary.
*176340 U.S. at 40, 71 S.Ct. at 107. The Court later made clear that the same procedure should be followed when mootness prevents review of administrative orders. See A.L. Mechling Barge Lines, Inc. v. United States, 368 U.S. 324, 82 S.Ct. 337, 7 L.Ed.2d 317 (1961).
These cases require us to vacate FERC’s 1979 approval of the Point Conception proposal. When the applicants failed to meet the Commission’s February 1983 deadline for the completion of the project, the 1979 authorization expired according to its own terms. Such expiration is perhaps the most classic way in which a challenge to an order can become moot. See, e.g., Tennessee Gas Pipeline Co. v. FPC, 606 F.2d 1373 (D.C.Cir.1979); Alton & So. Ry. Co. v. International Ass’n of Machinists, 463 F.2d 872, 877-78 (D.C.Cir.1972). Even FERC concedes that “the conditional certificates issued in 1979 are no longer valid.” Brief for Respondent at 20. It follows that the 1979 order is now unreviewable and thus fits squarely within the spirit as well as the letter of the Munsingwear rule.
IV.
Unlike the 1979 order, FERC’s 1983 order has not expired, but neither does it grant authority that may be used by private parties in the future. The 1983 order does not give anyone permission to do anything; it purports to be merely declaratory. Since the 1983 order is still in effect, petitioner’s challenge to the order is not precluded by mootness, and the Munsingwear rule appears not to apply. Nonetheless, we conclude that the 1983 decision must be vacated for reasons closely tied to the attenuated and abstract nature of the question FERC posed for itself.
Assessment of the action FERC took in 1983 is made difficult by the ambiguity of the Commission’s order. FERC found that “Little Cojo is seismically suitable for the safe and reliable construction and operation of an LNG receiving terminal.” Pacific Alaska LNG Co., 25 FERC Rep. (CCH) ¶ 61,005, at 61,034 (Oct. 4, 1983). That language could be read to mean simply that Little Cojo Bay is not the crater of Mauna Loa; i.e., that FERC refuses to rule out the Point Conception site on seismic considerations alone, although it reserves judgment on whether the benefits of the project outweigh all of its costs, including the seismic risks. Petitioners, however, read the order to conclude entirely FERC’s considerations of seismic risks at Point Conception and hence to preclude, absent new seismic evidence, any weighing of these risks in future proceedings on the LNG project.
FERC itself apparently shares petitioners’ view of the order. The text of the order suggests that the Commission intends it to have broad preclusive effect: the Commission acted “to resolve the seismic issues” and deferred consideration only of “non-seismic issues.” Id. at 61,034-35. Similarly, the ALJ’s decision, affirmed by the Commission, states that “[i]t is clear that the seismic conditions, risks and suitability of the site in no way depend upon or vary with changes in project need, financial sponsorship, gas supply or the other non-seismic factors mentioned above.” 19 FERC Rep. (CCH) ¶ 63,086, at 65,287. At oral argument, moreover, counsel for the government described the 1983 order as conclusively settling all seismic questions regarding the Point Conception site, subject only to reconsideration based on new evidence.
Perhaps, on an appropriate record, FERC could proceed in this fashion, conclusively ruling out in advance a set of environmental concerns as insignificant; perhaps not. We need not decide that question, because the record in this case clearly does not support the far-reaching conclusion that seismic risks at Little Cojo Bay are so miniscule that they could never tip the balance in a reasoned determination whether to approve plans for an LNG terminal, no matter how slight the need for the project and no matter how worrisome *177its other costs. To begin with, at no point in these proceedings did the Commission articulate, let alone justify, criteria — numerical or otherwise — for identifying risks of no possible importance.
More importantly, the Commission’s unarticulated test for negligible risk was apparently tied closely to its assessment in the late 1970’s of the need for LNG in California — an assessment which everyone agrees is now hopelessly obsolete. As discussed above, the Commission’s staff has argued consistently that scientific understanding of surface faulting is so meager that the presence of active faults on a site should rule it out as a possible location for an LNG terminal. This position was rejected in 1979 in large part because of considerations having nothing to do with the seismic characteristics of Little Cojo Bay: the ALJ feared that the staff’s approach would disqualify much of the California coast, and both. the ALJ and the Commission perceived a pressing need for. an LNG receiving terminal somewhere in California. On remand, neither the ALJ nor the Commission offered any new reason for rejecting the staff’s position; the prior conclusion was reaffirmed solely because, as the staff conceded, no new seismic evidence warranted a change in policy, and because neither the ALJ nor the Commission was willing to consider new non-seismic evidence pertaining to the need for the facility.
The record thus provides patently inadequate support for the 1983 order as construed both by petitioners and by the Commission. The order could perhaps be saved by reading it more narrowly to rule only that seismic risks alone do not permayiently disqualify Little Cojo Bay as a potential site for an LNG plant. Such a reading, however, would deprive the order of any significant legal content. Although in general “there is no law ... against the Commission’s giving comically irrelevant responses,” Transwestern Pipeline Co. v. FERC, 747 F.2d 781, 786 (D.C.Cir.1984), under the circumstances we decline to depart from the Commission’s own interpretation of its 1983 decision solely so that we may affirm the order as vacuous.
Our refusal to transform FERC’s decision into a harmless nullity is based in part on the terms of the Administrative Procedure Act, which authorize agencies to issue declaratory orders only “to terminate a controversy or remove uncertainty,” 5 U.S.C. § 554(e). FERC was faced with no live “controversy” in 1983, since the sponsors of the Point Conception project were by their own admission no longer prepared to go forward. Although a genuinely conclusive evaluation of the seismic risks could conceivably have alleviated some confusion, FERC was not in a position to make such an evaluation, and an order which in reality concluded next to nothing could not in any meaningful way “remove uncertainty.” On the contrary, FERC’s 1983 order has served only to exacerbate uncertainty by placing matters in a “peculiar, uncertain limbo,” and “that uncertainty this court removes by vacating the order.” Tennessee Gas Pipeline Co. v. FPC, 606 F.2d 1373, 1380, 1382 n. 40 (D.C.Cir.1979).
V.
We therefore vacate both of the orders before us. The 1979 order we vacate as moot, A.L. Mechling Barge Lines, Inc. v. United States, 368 U.S. 324, 82 S.Ct. 337, 7 L.Ed.2d 317 (1961), and the 1983 order as without substantial support in the record, Natural Gas Act § 19(b), 15 U.S.C. § 717r(b).
The Commission has compiled a lengthy and instructive record in the proceedings giving rise to the vacated orders and is under no obligation to ignore its results or to retrace its steps. In any future proceedings concerning the Point Conception site, FERC may incorporate by reference any or all of the record developed in these proceedings and may rely to the extent it deems appropriate on the evidence contained in the record. The orders issued in these proceedings, however, may not them*178selves be accorded any preclusive legal effect.
It is so ordered.