dissenting:
The Columbia River stretches more than 1,200 miles from its headwaters in Canada to the Pacific Ocean. It drains an enormous swath of the northwestern United States (258,000 square miles). This great *1146river empties more water into the Pacific Ocean than any other river in North or South America. Description: Columbia River Basin, United States Geological Survey at http://www.vulcan.wr.usgs.gov/Vol-eanoes/ Washington/ColumbiaRiver/de-scription_cohimbiamdver.html (last visited June 26, 2006). Shipping and trade are just one dimension of the Columbia River’s importance to the Northwest. The Columbia River is central to the life of man, beast, and fish in the Northwest; it is a complex system defined by interdependen-cies.
Perhaps the most iconic of these inter-dependencies is that among the river, salmon, and fishermen. A healthy river is vital to endangered salmon and steelhead trout and to the fishermen who rely on these fish. Virtual World, Columbia River, The National Geographic Society at http://www.national geographic.com/earth-pulse/columbia/index_flash.html (last visited June 26, 2006). Salmon and steelhead are anadromous — they spend the first two years of their lives in fresh water, like the cool waters of the upper reaches of the Columbia. These fish then swim to sea, resting and adjusting in the critical estuary at the conjunction of fresh and salt water. After a few years in the ocean, the adult salmon and steelhead return to the fresh waters of the Columbia to reproduce, swimming upstream to the places where they were hatched. The Columbia River is vital salmon and steelhead habitat and was once teeming with these fish. Sadly, salmon and steelhead are in serious decline, decimated by obstruction of passage because of dams and destruction of habitat. In fact, this year sport-fishing on the Columbia was closed until very recently because the numbers of returning fish are perilously low. Salmon Fishing Halted on Columbia River: Concern About a Ten-fold Decrease in Chinook Populations Spurs Action, Jeff Barnard, Associated Press (Apr. 22, 2006). Major efforts to reverse their decline by easing the passage through dams and other measures are ongoing.
The Army Corps of Engineers proposes deepening the shipping channel that runs from the mouth of the river to its ports, including particularly Portland, from forty-feet to forty-three-feet deep. We are presented with Northwest Environmental Advocates’ challenge to the Army Corps’ compliance with National Environmental Policy Act (NEPA). Northwest Environmental Advocates (NWEA) argues, and I agree, that the Army Corps has failed to satisfy its obligations under NEPA to adequately study either the direct, the indirect, or the cumulative effects of channel deepening, particularly the impacts of dredging upon coastal erosion, release of toxicity, and estuary salinity, that affect both humans and salmon. In addition, the Corps’ economic analysis is deeply flawed — adopting a methodology for measuring the costs and benefits associated with channel deepening that enables it to avoid consideration of economic reality and to present a far rosier picture of the economics of channel deepening than is realistic. The Corps has certainly produced an enormous record during this NEPA process; however, simply producing a record of the size presented to us in this case does not mean that the agency has answered the correct questions, undertaken relevant studies, or adopted appropriate methodologies. The Army Corps has not met its obligations to take a “hard look” at the direct, indirect, and cumulative impacts of deepening the Columbia River channel, and its economic analysis is obviously flawed. A serious look at the economics of the project suggests that it is like Kodiak’s proposed bridge to nowhere.
My concerns in this important case are heightened by the enormity of the consequences of a wrong decision. Adequate *1147fact-finding and thorough analysis, which are at the core of NEPA, are all-important to make sure that the Army Corps’ decision in the end is the right one. The Columbia River is too important a resource for all of us to allow its misuse. I, therefore, respectfully dissent.
In light of these serious problems, we should reverse and remand for the agency to undertake a more thorough study of the impacts of channel deepening and the associated ocean disposal site on coastal erosion; a scientifically-sound analysis of changes to toxicity and salinity in the river and estuary; and a more rigorous look at the economics of the proposed project. Too much is at stake to permit the Corps to move forward with such incomplete and inadequate information. NEPA forbids it.
I. THE CORPS’ ECONOMIC ANALYSIS
I have much to criticize about the Corps’ failure to consider adequately the environmental consequences of its proposed action. I will get to that. But I choose to first talk about its disgraceful attempt to justify the dollars and cents of its proposed project. This should not come as a surprise to any of us. The Corps is wont to undervalue costs and overvalue benefits so that it can get on with its mission— constructing water projects.1
Here, instead of objectively looking at whether the proposed project is a net economic benefit to the American economy, it has chosen a method of analysis that reaches its desired result — admitted even by it as perhaps not the best methodology — but a methodology that justifies the project. It ignores salient facts: Northwest ports (Puget Sound and Grays Harbor) can handle the freight and offer safer shorter transit — the river trip is longer (one-hundred miles from the bar to Portland) and fraught with hazards (the shifting bar at the mouth and shifting sand bars up river). Two-thirds of the container ship business has already left the river. The Portland Port that is up the Willamette River cannot be deepened because of pollution. No wonder the Corps declined to consider displacement of traffic from one American port to another. It insisted on including supposed benefit to foreign ships as part of the benefit. Should American taxpayers take comfort that foreign bottoms are the principal beneficiaries of any economies in the project?
What about the cost side? No analysis is made of the costs of potential devastating erosion that may result to the Washington and Oregon coastlines and the jetties.
The gravamen of Appellant’s claims, with which I agree, is that the Army Corp’s benefit-cost ratio of $1.66 to $1 is inaccurate because the agency has failed to consider several factors, resulting in an overstatement of the project’s economic benefits and the understatement of its costs. Appellant argues: (1) that the agency understated the costs associated with the project by failing to consider environmental externalities associated with channel deepening and disposal of dredged material and (2) that the agency overstated the benefits associated with channel deepening by not distinguishing benefits to the U.S. economy from those to foreign parties, not considering declining contain*1148er-ship traffic into the Columbia River, and not undertaking a multi-port analysis.
The majority agrees that “[ijnaecurate economic information may defeat the purpose of an EIS by ‘impairing the agency’s consideration of the adverse environmental effects’ and by ‘skewing the public’s evaluation’ of the proposed agency action.” Natural Res. Def. Council v. U.S. Forest Serv., 421 F.3d 797, 811 (9th Cir.2005) (quoting Hughes River Watershed Conservancy v. Glickman, 81 F.3d 437, 446(4th Cir.1996)). In Natural Resources Defense Council, we held that the Forest Service’s inflated assessment of market demand for timber from the Tongass National Forest “subverted NEPA’s purpose of providing decision makers and the public with an accurate assessment of the information relevant to evaluate” the agency’s proposed action. Id. at 812. A similar finding is required here: the agency’s economic analysis does not provide an accurate assessment of the costs and benefits of the planned channel-deepening project because of its failure to undertake analysis that could decrease the benefit-cost ratio.
A. Costs of Deep-Water Disposal, Coastal Erosion & Mouth of the Columbia River Jetty Erosion
It is apparent that the agency’s analysis of the economic costs associated with this project was insufficient. First off, because the agency did not consider the impacts from coastal erosion resulting from removing sand from the littoral system and the impacts of channel deepening on the MCR jetties, it follows that the economic analysis is similarly incomplete for failing to consider the potential costs associated with jetty and coastal erosion. In this regard, the agency’s analysis of costs suffers from the same incompleteness as its analysis of the cumulative effects of deep-water disposal.
The majority notes that the FSEIS nowhere finds that the jetties at the Mouth of the Columbia River will be undermined and will require repair as a result of the channel-deepening project. That’s because it did not consider the probability of erosion that would result from disposal outside the littoral system. This is the very type of issue for which Niemi’s declaration should have been admitted since it pointed out factors which the agency did not consider but should have. The Corps, without analysis of facts, says that channel deepening will not undermine the jetties, but Niemi presents credible evidence to the contrary, evidence that the Corps should have considered. The correct path is to remand to the agency for a complete examination of these potential costs.
As the majority concedes, the erosion damage from the Corps’ past practices has been considerable. The Corps tells us it has changed its ways to minimize erosion from the MCR dredging and the construction of the jetties. But it tells us nowhere what steps it would take to control devastating erosion that would occur if considerable dredge material is put in the deep-sea site it has prepared to receive it. Nor has it suggested the costs that would be involved.
B. Benefits to U.S. Economy and Lack of Multi-Port Analysis
In addition to the fact that the Corps has understated costs associated with the channel deepening project, the Corps appears to have overstated the economic benefits of the channel deepening by (a) failing to adequately account for displacement of shipping and trade from one American port to another; (b) failing to consider what if any of the benefits associated with channel deepening would accrue to the U.S. economy; and (c) failing to undertake a multi-port analysis that might well show *1149no net benefit to the U.S. economy from the channel deepening. The Corps’ analysis of the economic benefits of the project, Exhibit M, explains the shipping of various commodities (wheat, corn, barley, and soybeans) from various Columbia River ports as well as the containerized shipping trade. Appellant’s claim that the Corps has overstated these benefits, requires us to look closely at the Corps’ analysis — are there economic factors that the Corps should have considered under NEPA? The record indicates that the answer to this question is “yes.” The Corps neglected important economic realities that undermine the benefit-cost ratio it presents.
While the majority opinion is correct that Exhibit M of the FSEIS has assumed some loss of containerized trade to the ports of Puget Sound, Maj. Op. at 1143, there is evidence that this loss may be far greater than the FSEIS estimated. The Corps estimates that the channel deepening project will generate benefits of $18.8 million annually. Two-thirds of this total, $11.7 million, comes from the Corps’ estimate of gain from the containerized shipping trade. However, two of the three shipping lines that carry containerized cargo to the Port of Portland have decided that they will no longer come to Portland, reducing by 2/3 the amount of containerized cargo coming through Portland.
Consideration of this displacement of shipping from one American port to another is missing in the agency’s analysis and would decrease the net benefits associated with the channel deepening. In response, Federal Appellees respond that the FSEIS’ economic analysis, Exhibit M, assumed a loss of market share to Puget Sound and therefore that this issue has been accounted for.2
Contrary to the finding of the majority opinion, the Corps’ failure to measure the net benefit to the U.S. economy from the project is wrong. The Corps asserts that it is not required to separate the benefits to foreign vessels from those to the domestic economy. However, the Guidelines, that are mandatory on the Corps, provide that “the procedure for measuring the beneficial contributions to national economic development (NED) associated with the deep draft navigation features of water resources plans and projects.” Economic and Environmental Principles and Guidelines for Water and Related Land Resources Implementation Studies, March 10, 1983, available at http;// www.iwr.usace.army.mil/iwr/pdf/p & g.pdf (last visited April 13, 2006) at 58 (emphasis added). Appellees’ argument that such an analysis is not required is inconsistent with the language of these Guidelines and with the purposes of NEPA.
*1150Appellant also argues that the Corps should have undertaken a multi-port analysis as part of its economic analysis. Such an analysis would provide insight into whether the ships that the channel-deepening project is intended to bring into Columbia River ports are already docking in another American port. If this is the case, bringing those ships to the Columbia provides no net benefit to the national economy. Such a conclusion would undermine the Corps’ stated benefit-cost ratio of $1.66 benefit for each dollar spent. The NED Guidelines discussed above indicate that the agency should, as a first step, “determine the economic study area.” This includes consideration of “diversion from or to adjacent competitive harbors as well as distribution via competing modes of transport .... there should be adequate discussion of the trade area relative to adjacent ports and any commonality that might exist.” Id. at 62.
The Army Corps argues that its choice of methodology was a reasoned one, that choosing a methodology that did not employ a multi-port analysis did not change the outcome of the economic analysis, and therefore that not doing one cannot be arbitrary and capricious. See Comments on FSEIS from Colonel Richard Hober-nicht, Commander, Army Corps; Robert E. Willis, Army Corps; and Judy Grigg, Port of Longview at 23 (Feb. 28, 2003) (responding to criticism that the agency did not measure whether there was a benefit passed through to the U.S. economy and did not do a multi-port analysis), id. at 34-35(explaining its choice of methodology: “The Corps selected an analytical methodology that did not include a multi-port analysis because (1) it was not required by the Principles and Guidelines; (2) the Corps concluded that such an analysis would likely result in higher project benefits than the method selected and, as a result, the substantial expense of a multi-port analysis was not fiscally justifiable; and (3) conducting such an analysis would require assumptions likely to render the outcome overly speculative.”). These assertions are unsupported anywhere in the record. Although the Corps’ Review Panel recommended that a multi-port analysis would have been better methodology, the Panel found that such an analysis would “be unlikely to tip the cost-benefit scales.” Original Review Comments and Benefit Review Team Opinions on Responses at 1 (Jan. 10, 2003). That Review Panel endorsed the Corps’ economic analysis without the multi-port analysis. The Corps insists that the court must defer to such a methodological choice. Salmon River Concerned Citizens v. Robertson, 32 F.3d 1346, 1359 (9th Cir.1994). This argument for deference would allow the agency to end-run NEPA’s obligations to do an accurate economic analysis by allowing the agency to choose a methodology that is clearly not the more appropriate and then to hide behind that choice of methodology by calling for deference to agency decision-making. This is not what NEPA envisioned. While we certainly must respect the agency’s technical decisions, where those decisions enable the agency to ignore reality, we need not acquiesce.
C. Failure to Consider Niemi Declaration
Further, the district court erred in not fully informing itself as to what critical information the Corps did not consider. The district court abused its discretion in striking Ernest Niemi’s declaration as impermissible extra-record material. Only by considering it could it know what the Corps omitted. The majority, as did the district court, relies on Asarco, Inc. v. U.S. Envt’l. Protection Agency, 616 F.2d 1153 (9th Cir.1980), as the basis for its rejection. Both used the “rule” of Asarco as far too blunt an instrument. The rule expressed *1151in Asarco and subsequent cases is much more nuanced.
The Asarco court considered and weighed the several rules that govern the “judicial review of agency action.” 616 F.2d at 1159. “[Ajgency action must be examined by scrutinizing the administrative record at the time the agency made its decision.” Id. However, the court recognized that:
[I]t is both unrealistic and unwise to straitjacket the reviewing court with the administrative record. It will often be impossible, especially when highly technical matters are involved, for the court to determine whether the agency took into consideration all relevant factors unless it looks outside the record to determine what matters the agency should have considered but did not. The court cannot adequately discharge its duty to engage in a substantial inquiry if it is required to take the agency’s word that it considered all relevant matters.
Id. (internal quotation marks omitted).
To reconcile these tensions, the court held the following:
If the reviewing court finds it necessary to go outside the administrative record, it should consider evidence relevant to the substantive merits of the agency'action only for background information ... or for the limited purposes of ascertaining whether the agency considered all the relevant factors or fully explicated its course of conduct or grounds of decision.... Consideration of the evidence to determine the correctness or wisdom of the agency’s decision is not permitted, even if the court has also examined the administrative record. If the court determines that the agency’s course of inquiry was insufficient or inadequate, it should remand the matter to the agency for further consideration and not compensate for the agency’s dereliction by undertaking its own inquiry into the merits.
Id. (emphasis added).
Reaffirming Asarco in Love v. Thomas, 858 F.2d 1347(9th Cir.1988), we stated: “The court may find it necessary to review additional material to explain the basis of the agency’s action and the factors the agency considered,” and “the court may consider, particularly in highly technical areas, substantive evidence going to the merits of the agency’s action where such evidence is necessary as background to determine the sufficiency of the agency’s consideration.” Love, 858 F.2d at 1356.
In this case where our task is to determine whether the agency undertook adequate fact-finding and analysis that should have been part of its NEPA compliance, the question is whether Niemi’s declaration constitutes either (a) background, explanatory information or (b) substantive evidence needed to “determine the sufficiency of the agency’s consideration.” Id. If it falls into either or both of these categories, it should have been considered. Given the concern that the agency did not undertake analysis of relevant considerations, the information provided in Niemi’s declaration fits quite cleanly into the second category. It is substantive evidence that demonstrates the inadequacy of the agency’s inquiry, pointing out factors that the agency should have considered and did not. Had these factors been considered, substantial revisions to the agency’s benefit-cost ratio of $1.66 in benefit to $1 cost well could be in order. As long as it is not employed to substitute the court’s substantive determination for that of the agency, the Niemi declaration is exceedingly useful to a thorough review of the agency’s economic analysis. In a highly technical area, it sheds light on the sufficiency of the analysis undertaken by the agency.
*1152The district court appropriately cited to the Asarco rule but decided that the declaration had been offered to “determine whether or not the FSEIS provides a misleading description of the project’s potential impacts,” citing Mr. Niemi for that impermissible purpose. Northwest Environmental Advocates v. Nat’l Marine Fisheries Serv., No. C04-0666RSM, at 11-12, 2005 WL 1427696 (June 15, 2005). What matters in applying Asarco’s rule is not how the declarant views his extra-record testimony, but rather, for what purpose the court should make use of it. The district court was obligated to undertake a detailed review of the agency’s analysis. Where the concern is whether the agency considered all relevant factors in its NEPA analysis, the failure to look outside of the record to Niemi’s declaration in my view was a failure by the district court to engage in the detailed inquiry required by the Administrative Procedure Act and therefore was an abuse of discretion.
We should admit Niemi’s declaration to the extent it contains evidence of the omissions and inaccuracies in the agency’s analysis and find that the agency’s failure to consider likely costs and its probable overstatement of the project’s benefits constitutes “[ijnaccurate economic information” within the meaning of Natural Resources Defense Council. 421 F.3d at 811.
II. CUMULATIVE EFFECTS ON COASTAL EROSION OF DEEP-WATER DISPOSAL OF SPOILS FROM THE MOUTH OF THE COLUMBIA RIVER DREDGING
A. The Majority’s Flawed Analysis of Cumulative Impacts
The majority concludes that the Corps has adequately considered the cumulative impacts on coastal erosion of deep-water disposal of sediment dredged as part of both the Mouth of the Columbia River (MCR) Project and the channel-deepening project at issue here. It bases its determination on (1) the Corps’ assurance that it has considered those cumulative impacts; (2) the Corps’ treatment of deep-water disposal as allowable, although the least preferred option; (3) changes made to disposal practices for the Mouth of the Columbia River Project to maximize the amount of sediment retained in the littoral cell; and (4) plans to monitor and manage sediment disposal in the future to prevent coastal erosion. Maj. Op. at 1135-1138. The documents cited by the majority reveal that the Corps has proposed a deep-water disposal site large enough to receive all of the dredged material from both projects although the Corps alleges it would prefer to keep material in the littoral cell; that the agency plans not to employ the deep-water disposal site for a large volume of dredge disposal unless necessary. As the majority tells us, the record explains the volume the Corps expects to dispose in deep water by reference to the expected capacity of Site E and the North Jetty Site, both disposal sites within the littoral system, and the dispersion of material from those two sites. That is, the volume of dredged material that the Corps plans to place in deep-water disposal depends upon how much Site E and the North Jetty Site can accommodate in any given year. If Site E and the North Jetty Site do not disperse sediment as quickly as anticipated, a clear possibility that the Corps acknowledges, the volume of material that would be disposed in the deep-water site would increase accordingly. This demonstrates that the volume of material to be disposed in the deep-water site is uncertain, dependent upon several variables. Despite this uncertainty and the probability that the deep-water disposal scenario will not play out as it hopes, the Corps has failed to undertake broader cumulative-effects analysis that would result from dumping significant volumes of sand *1153into a deep-water site outside the littoral cell.
The majority attempts to distinguish this case from our decision in Klamath-Siskiyou Wildlands Center. v. Bureau of Land Management, 387 F.3d 989 (9th Cir.2004) that the Bureau of Land Management did not sufficiently consider the cumulative impacts of several timber sales. The majority finds central to the holding in that case the fact that a single project was subdivided and those sub-parts were analyzed separately from one another. Maj. Op. at 1138-1139. In focusing on this factor, the majority overlooks an important commonality between Klamath-Siskiyou and this case: the Forest Service’s cumulative effects analysis looked only at the effects of the particular project at issue, failing to account for the “combined effects that can be expected as a result of undertaking [multiple timber sales] and other foreseeable projects, in addition to [the project at issue.]” Klamath-Siskiyou, 387 F.3d at 996. Although the majority opinion finds that the Corps’ analysis did look at the cumulative impacts of deep-sea disposal of all material dredged from the Mouth of the Columbia River Project, what it concluded was that it would be as devastating as indeed it would. What it doesn’t do is detail what the erosion would cause. Apparently, it thinks nothing could mitigate it because the record is devoid of any such analysis.
B. The Army Corps’ Analysis and Appellant’s Concerns
The Dredged Material Management Plan/Supplemental Environmental Impact Statement (DMMP/SEIS), which constitutes the no-action alternative in the Final Supplemental Environmental Impact Statement (“FSEIS” or “Final Environmental Impact Statement”), called for establishing a small deep-water disposal site for 7.7 mcy (million cubic yards) over 20 years. This plan was “intended to reduce rehandling of material that currently erodes back into the navigation channel.” Columbia River Dredged Material Management Plan and Supplemental Environmental Impact Statement at 32 (June 1998). Deep-water disposal would occur outside the zone where currents would bring dredged material back into the navigation channels from whence it would ultimately have to be dredged once again.
The 1999 Final Environmental Impact Statement expanded the DMMP/SEIS proposal for deep-water disposal of dredge spoils with the continued dredging associated with the Mouth of the Columbia River Project in mind. “Continued maintenance of the MCR project is a necessary component for the viability of not only the existing 40-foot navigation channel but also to any proposed channel improvements.” Vol. I: Main Report and Exhibits, Integrated Feasibility Report for Channel Improvements and Environmental Impact Statement (FEIS) at 1-2 (Aug.1999). The four existing ocean disposal sites for the Mouth of the Columbia River Project were determined to have “inadequate capacity.” Id. “The timing of this long-term site selection process [for new ocean disposal sites] and the need to identify suitable ocean disposal sites for construction and long-term maintenance of proposed channel improvements further established the need to address the combined ocean disposal requirements in this document [the FSEIS].” Id. This language in the Final Environmental Impact Statement indicates that ocean disposal will be necessary for both the disposal of materials dredged for the Mouth of the Columbia River Project and either the maintenance of the 40-foot depth of the channel or the deepening of the channel to 43 feet “as existing estuarine disposal sites reach capacity.” FEIS, Exh. H at H-3.
*1154To deepen the channel to 43 feet, over the next thirty years, “total Project dredging quantity” -will be “about 190 mcy [million cubic yards] for the Project.” Biological Opinion at 12 (May 20, 2002). Without the deepening, simply maintaining the channel at 40 feet (the no-action alternative spelled out in the DMMP/SEIS) over the same time period, approximately 160 mcy would have to be dredged. Id. In its proposal for additional ocean disposal sites, the Corps estimated volumes of ocean disposal for both the Mouth of the Columbia River Project and the channel-deepening or channel-maintenance dredging.
MCR O & M3
Annual 50 Years
4.5 mcy 225 mcy
40-foot O & M Total 40-foot Channel
1-20 Years 21-50 Years
8 mey 12 mcy 20 mey
Construction 43-foot O & M Total 43-foot
1-20 Years 21-50 Years Channel
7 mey_9 mcy_21 mcy_37 mey
Although deep-water disposal, according to the Corps, is not the first or most preferable option, see FSEIS, Exh. H at H-6 (“Dredged material placed at [disposal sites closer to shore] is expected to move out of the site and feed the littoral system.”), these ocean disposal sites were selected because the mouth of the Columbia River is “so dynamic.” Id. The Corps’ preferred outcome is dependent (1) upon dredge volumes to maintain both the Mouth of the Columbia River Project and the channel deepening not exceeding expectations in any given year and (2) the dredged materials put into near-shore sites eroding at the rate anticipated. If either of these expectations plays out at a different rate,4 there will be no room for new dredge spoils to be dumped in littoral sites, id., and deep-water disposal would be necessary. Another way of making this point is that, in any given year, the agency’s preferred outcome, littoral disposal of dredge spoils, is dependent upon at least two factors, the uncertainty as to both of which is acknowledged by the Corps. A “single Deep Water Site was located and sized to accommodate almost a 5-year disposal capacity (225 mcy).” Id. at H-7.
The Army Corps simply has failed to consider cumulative impacts of the permanent removal of sand from the littoral system from both the Mouth of the Columbia River and channel-deepening or maintenance that is inevitable if there is deep-water disposal. The agency should have considered the cumulative impacts of removing at least 262 mcy (the total amount of material to be dredged from both channel deepening and ongoing MCR dredging) to the deep-water site, not just 70 mcy of sediment from channel deepening.
The Corps insists that it has considered the cumulative impacts on coastal erosion of disposing of dredged material from both the Mouth of the Columbia River Project and the channel-deepening project. In evaluating the Corps’ contention, the majority focuses on two dynamics that potentially drive coastal erosion: (1) the impacts associated with disposal of sediment dredged from the MCR Project and the channel-deepening project and (2) the potential changes in sediment transport and river hydraulics caused by these two projects. The Corps and majority opinion focus on Exhibit J to the FSEIS and the *11552003 FSEIS, as well as Exhibit H to the FEIS. Neither addresses NWEA’s concern that the specific impacts on coastal erosion of disposing of dredged material from both the Mouth of the Columbia River Project and channel deepening in the deep-water disposal site remain unquantified and essentially unknown.
Exhibit J is the agencies’ supplemental analysis in response to Oregon and Washington’s initial refusal to certify the channel deepening project over concerns about sediment transport and coastal erosion. Exhibit J is simply not adequate. The Corps’ analysis in Exhibit J proceeds as follows. First, the Corps considered “historic trends and changes in sediment transport in the Columbia River System” and attributed decreased sediment transport down the Columbia River to the dams upstream. This analysis appears to be sound. It is what follows that is questionable. Next, the Corps “examined whether the channel deepening project itself would affect sediment transport.” Fed.App. Br. at 25. The Corps determined, first, that dredging the channel will not reduce the sand available for transport because the surface area to be dredged is a relatively small percentage of the entire riverbed (3.5%) and, second, that because the riverbed has sand 100-400 feet deep, the amount being dredged also represents an insignificant percentage. Then the Corps determined that the channel-deepening project’s impacts on the Columbia River’s hydraulics are “too small to measurably alter sand transport or erosion/accretion in the river or estuary.” FSEIS, Exh. J at 2-3. The Corps’ conclusion: “Because the Project will neither reduce the sand available for transport nor alter the river’s capacity to move sand, the Project will in turn have no impact on the erosion or accretion of coastal beaches.” Fed.App. Br. at 27.
Lastly, the Corps weighed whether consideration of other actions’ impacts would alter its analysis of the impacts of the channel deepening Project. This is the crucial part of the analysis. In its consideration of the cumulative impacts associated with the Mouth of the Columbia River Project’s ongoing dredging, the FSEIS relies on Exhibit J’s finding that “the reduction in the Columbia River’s net sand discharge to the MCR since the early 1900s is related to lower Columbia River flood discharges and not the navigation channel or the MCR jetties.” FSEIS at 6-72. The FSEIS finds that reduced accretion of sand in the estuary is the result of the decreased sand flow from the River, the result of upstream dams and concludes that “[ejxcluding the historic effect of the MCR jetties, navigation channel development and maintenance, including maintenance of the MCR project, has not altered the estuary’s overall accretion/erosion or bedload transport patterns.” Id. at 6-73. So far, so good. The agency’s analysis might be fine, except that its consideration of coastal erosion and sediment transport issues stops here.
Appellees point out that Exhibit J explains how the Columbia River and the currents at its mouth historically moved sand downstream and distributed that sand along Washington and Oregon’s coastlines. It also describes the give-and-take of sand between the estuary and the mouth of the Columbia River. It analyzes how the construction of dams upstream, jetties downstream, and dredging the Columbia River changed these patterns. This may all be true and important to the analysis as a whole, but it does not answer the critical question. The difference between the analysis the Corps is obligated to provide and that provided in Exhibit J is as follows:
*1156As Exhibit J explains, historically, for the most part, disposal of dredged sand largely occurred within the littoral system. Although transport patterns were being altered, that is, the way in which the sand was being moved around was changing, and although less sand was being sent downstream because of dams, once downstream the amount of sand in the system remained fixed. By contrast, the record indicates that near-shore and beach disposal are not a permanent solution and that ocean disposal (outside of the littoral system) will increasingly be employed. Increased ocean disposal means that, unlike in the past, the amount of sand being moved around near the mouth of the river will not remain fixed. Over 50 years, 262 mcy may be removed from the littoral system. This permanent removal of material from the littoral system is not the same as moving sand from one part of the littoral system to another. Exhibit J does not address the effect of the removal of sand from the littoral system. The agency nowhere analyzes what happens when 262 mcy is removed from the littoral system entirely, despite the fact that this is an anticipated possibility, verging on a probability.
The majority opinion accuses NWEA of relying on the 2003 FSEIS to establish its argument, when it is the 1999 FEIS that contains the relevant analysis of cumulative impacts. According to the majority, the FEIS, particularly Exhibit H, demonstrates that the Corps was “aware of’ the serious impacts on coastal erosion of disposing of large quantities or all material dredged from both the Mouth of the Columbia River Project and the channel-deepening project in the planned ocean disposal site. Maj. Op. at 1134 -1135. To establish this, the majority references an alarming statement from the 1999 FEIS:
If the deepwater site is used as intended5 (4.5 mcy of MCR sand placed per year for 50 years), the implications on the littoral sediment budget at MCR and adjacent coastal areas could be profound .... The result of such a mass removal of littoral sand would likely be adverse: Local and possible regional coastal erosion may result. The stability of MCR jetties may be reduced....
(emphasis added).
The majority is correct — this statement does show that the Corps “was fully aware of the potential erosion effects entailed by deepwater disposal.” Maj. Op. at 1135. The agency’s response, or lack thereof, to this possibility is very troubling. The majority insists that Exhibit H contains a detailed analysis of these potentially “profound” and “adverse” effects. Not so. Exhibit H provides a recital of the process that the Corps went through with other groups to narrow the list of potential deep-water disposal sites and to select one that the Corps would use. The majority finds that the lists of possible conflicts (harms) associated with each potential deep-water site (which the Army Corps calls “conflict matrices”) provide thorough analyses of the advantages and disadvantages of each potential site. In fact, they are no more than tables and checklists. Undertaken for each of the ten potential deep-water sites, each matrix, such as it is, notes whether there is a conflict, potential con*1157flict, no conflict, or a beneficial use for 27 specific characteristics of a site based on eleven specific factors for ocean disposal site selection specified in 40 C.F.R. § 228.6 and five general criteria for the selection of ocean disposal sites from 40 C.F.R. § 228.5. FEIS, Exh. H at H-45-55. Among the specific site characteristics being evaluated are unusual topography, physical sediment compatibility, commercial fisheries, critical habitat of threatened or endangered species, and cumulative effects. For example, the matrix for the proposed deep-water disposal site notes a potential conflict with “potential for cumulative effects,” with a check in the “potential conflict” box; a note in the “comments” column indicates that use of this site could have a “[potential affect from crab and other fishing as well as disposal;” and a note as to the relevant regulatory factors (4 and 7 of the eleven factors6 and c and d of the five general criteria7). That’s it. Exhibit H does not dig deeper to quantify or elaborate these effects as to any of the 27 characteristics, including “potential for cumulative effects.” This is the extent of Exhibit H’s analysis of the cumulative impacts on coastal erosion of disposing of large quantities or all dredged material from the Mouth of the Columbia River Project and the channel-deepening project in this deep-water site: “[potential affect from crab and other fishing as well as disposal."
We have rejected this kind of underwhelming specificity before. In Klamath-Siskiyou, we held that tables that did not “provide object quantification of the impacts” where the tables noted only whether a certain factor would be “unchanged,” “improved,” or “degraded,” were inadequate under NEPA. 387 F.3d at 994. Great Basin Mine Watch v. Hankins, 456 F.3d 955 (9th Cir.2006), draws on Kla-mathr-Siskiyou. In that case, the Bureau of Land Management’s (BLM or Bureau) environmental impact statement, noted, in a statement reminiscent of the Corps’ statements, that “[t]here is a potential for cumulative effects from hazardous air pollutants including compounds of arsenic, hydrogen cyanide, manganese, propylene, and acid aerosols.... ” The extent of the Bureau’s analysis was a “generic map” and “three tables that list existing and reasonably foreseeable mines.” We held that the Bureau’s “somewhat alarming statement was nowhere ... supported by data broken down by mine, or even by cumulative data.” Great Basin, 456 F.3d at 973. The same could be said here of the Corps, which has told us that the “intended” use of the deep-water disposal site could result in “profound” or “adverse” coastal erosion impacts and then provides us only with checklists of factors and no data. Exhibit H does not adequately address the cumulative impacts of coastal erosion, despite *1158the Corps’ and the majority’s efforts to argue that it does.
Bottom line: the majority finds that the “hard look” standard has been satisfied by the Corps’ assurance that, although the FEIS and FSEIS both propose authorizing the deep-water site for disposal of all MCR dredge, the Corps “presents this option only for ‘contingency planning purposes’ and as representing a ‘worst-case’ scenario.” Maj. Op. at 1136 (quoting FSEIS, App. H at H-6). I strongly disagree with the majority’s conclusion that the Corps discharged its obligations under NEPA by expressing its preference to dispose of dredged material at Site E and at the North Jetty Site, rather than at the deep-water site, even though substantial evidence suggests that those sites lack the required capacity. In so doing, the majority has held that, because the deep-water disposal of all materials dredged from the MCR project is a “worst-case” scenario and because the agency has revised its disposal plans and developed a management plan to minimize deep-water disposal, the agency need not detail the impacts of that scenario. By admitting it is a “worst-case” outcome, the Corps admits it is a bad solution, but nowhere tells us just how bad or if anything in the way of mitigation is possible.
C. Flaws in the Corps’ Compliance and the Majority Analysis
The majority’s reasoning is inconsistent with the requirements placed on agencies by NEPA and by our case law interpreting NEPA. The “hard look” standard requires, not just that the Corps express its preference not to reach a situation in which all MCR dredge is disposed of in deep water, and develop a management and monitoring plan intended to accomplish this preference, but that the Corps analyze the impacts of such a potential outcome, even if the Corps hopes to avoid it.
Lands Council instructs that cumulative-effects analysis under NEPA “requires the Final Environmental Impact Statement to analyze the impact of a proposed project in light of that project’s interaction with the effects of past, current, and reasonably foreseeable future projects.” Lands Council v. Powell, 395 F.3d 1019, 1027 (9th Cir.2005) (emphasis added) (citing 40 C.F.R. § 1508.7). It is not “appropriate to defer consideration of cumulative impacts to a future date. NEPA requires consideration of the potential impact of an action before the action takes place.” Neighbors of Cuddy Mountain v. United States Forest Serv., 137 F.3d 1372, 1380 (9th Cir.1998) (emphasis in original) (internal quotation marks and citation omitted). That the Army Corps has provided for deep-water disposal of all dredged material from the Mouth of the Columbia River Project (even as a “worst-case” scenario) and has sought authorization of a site with sufficient capacity for this potential outcome demonstrates that such an outcome is “reasonably foreseeable.” The impacts of disposal of all dredged materials from the MCR Project and channel-deepening into the deep-water site should, therefore, have been analyzed under NEPA. The majority’s assertion that because the Corps was aware of potentially adverse impacts and tried to avoid them, it need not study or quantify those impacts, is simply inconsistent with this standard.
The EIS must “include a useful analysis of the cumulative impacts of past, present and future projects. This means a discussion and an analysis in sufficient, depth and detail to assist the decisionmaker in deciding whether, or how, to alter the program to lessen cumulative impacts.” Muckleshoot Indian Tribe v. United States Forest Serv., 177 F.3d 800, 809-10(9th Cir.1999) (internal quotation marks and citations omitted). In Muckleshoot, this court *1159found that the agency’s cumulative effects analysis “f[e]Il short” of the required “useful analysis” where the agency “merely indicated] the amount of land to be exchanged, for example, and whether the land would be subject to commercial harvest, followed by an optimistic conclusion.” Churchill County v. Norton, 276 F.3d 1060, 1080 (9th Cir.2001) (citing Muckle-shoot, 177 F.3d at 811). The Army Corps’ analysis of the cumulative effects of deep-water disposal of MCR Project and channel-deepening dredged material is analogous to the analysis rejected by this court in Muckleshoot. The Corps’ statement of a preference against deep-water disposal is analogous to providing this court with an “optimistic conclusion.” This does not constitute a “useful analysis.”
Although the majority asserts correctly that an agency need not “engage in the most exhaustive environmental analysis theoretically possible”; it just has to take a “hard look.” Maj. Op. at 1139. Nonetheless, what is required to satisfy that standard must be applied in a manner consistent with the purpose of NEPA. Klamath-Siskiyou, 387 F.3d at 993(citing Churchill County, 276 F.3d at 1072). First off, it is important to identify at what the agency must take a hard look. The Army Corps asks us to find that the agency’s dismissal of a highly likely consequence of its proposed project, without any assessment of its consequences, satisfies the “hard look” standard. This is inconsistent with the mandate of NEPA, a statute that requires agencies “to put on the table, for the deciding agency’s and for the public’s view, a sufficiently detailed statement of environmental impacts and alternatives so as to permit informed decision making.” Id. at 1072. Here, the Corps tells us that there won’t be a problem. Although the agency admits the adverse consequences will be profound, it hasn’t studied what will happen if the “worst-case” scenario comes to pass, the Corps asks us to trust it. That is not the standard. The Army Corps has failed to take the required “hard look” at the cumulative impacts on coastal erosion of dumping large quantities of the dredged materials from the Mouth of the Columbia River and those from channel deepening into the proposed deep-water disposal site.
NEPA analysis must include an answer to the question posed by Appellant: "What would be the impact on coastal erosion if all or substantial quantities of the material dredged from the Mouth of the Columbia River Project and the channel-deepening project were disposed of in the deep-water site? NEPA requires the Corps to answer this question because it is a “reasonably foreseeable” outcome. Talking around that possible outcome, explaining why that is not the preferred outcome, planning to minimize this outcome, all the while taking steps that would bring this possibility to fruition, make clear that the Corps has not met its obligations under NEPA. We should remand for the agency to undertake further study. Instead, the majority allows channel deepening and authorization of the deep-water disposal site to proceed, even without an understanding of the cumulative effects on coastal erosion of a “reasonably foreseeable” future project.
III. OTHER SIGNIFICANT IMPACTS OF CHANNEL DEEPENING: TOXICITY AND SALINITY
NEPA requires the Army Corps to evaluate the direct and indirect impacts of channel deepening and disposal of the dredged material, considering all “relevant factors.” 40 C.F.R. §§ 1502.16, 1508.8; Rybachek v. U.S. Envtl. Prot. Agency, 904 F.2d 1276, 1284 (9th Cir.1990).
A. Toxicity
Northwest Environmental Advocates challenges the sufficiency of the Army *1160Corps’ analysis of the effects of channel deepening on the river’s toxicity.8 NWEA asserts that, in its analysis of the impact of channel deepening on levels of toxins in the Columbia River, the Corps failed to draw adequate samples from the sides of the riverbed, which as the channel is deepened may erode into the deepening channel, mobilizing whatever toxins they contain. The majority holds that there is no need to test sediments from the channel sides for two reasons: (1) because they will not be affected by dredging, and (2) because any material vulnerable to side-slope adjustment is previously-dredged material which must be toxin free. The record refutes this.
In its Biological Opinion, National Oceanic and Atmospheric Administration (NOAA) Fisheries identified the Columbia River’s toxicity to be “among the highest levels measured in estuarine sites in Washington and Oregon.” Biological Opinion at 36. These high levels of toxicity, just short of lethal, pose a serious threat to juvenile salmonids dependent upon the estuary, not to mention the potential adverse effects on humans and other species. The Army Corps explained the problem of “side-slope adjustment” noting that
After the initial deepening the riverbed would begin to adjust to the new channel depth. Riverbeds adjacent to the deeper dredge cuts will degrade as bedload is deflected down the cut slope and into the navigation channel. This process will continue for 5-10 years before the side-slopes reach equilibrium with the channel hydraulics. Sand eroded from these sides will become part of the active bedload transport on the riverbed.
FSEIS, Exh. J at 9.
By this process, channel deepening may re-suspend contaminants that had been isolated in the side slopes of the river-bed. Despite this concern, the Army Corps failed to test the toxicity of a sufficient number of sediment samples taken from outside the navigation channel, analyzing primarily samples collected within the channel. The Corps, therefore, has almost no data from which to analyze the impacts of side-slope adjustment on the overall toxicity of the river during and after channel deepening; without that data, the Corps cannot effectively analyze the impact of channel deepening not only on juvenile salmonids dependent upon the estuary, but other users of the river.
The Corps claims a database of 4,000 sediment samples from the Columbia River system. However, there are several problems with this database from the standpoint of an adequate NEPA analysis. First, only 586 of the 4000 samples have been tested for toxicity. Second, the sample set is very old. Many samples predate the 40-foot channel depth. Only 40 samples, taken from just 4 sites, were *1161taken later than 2000, and none of these are from side-slope areas.9
Salmon runs are at perilously low levels. The last thing that salmon need is another ladder to climb, so to speak. The Corps’ failure to thoroughly analyze toxicity in the side-slope sediment renders its NEPA analysis incomplete — the agency has not considered all “relevant factors” associated with the direct and indirect impacts of channel deepening on river toxicity.
B. Salinity
Appellant also points to the Army Corps’ inadequate analysis of the impacts of channel deepening on the salinity of the Columbia River estuary. Salinity has a tremendous impact on the estuary’s health. Like levels of toxins in the river, the salinity of the estuary is a matter of vital importance to the juvenile salmon which must rest and readjust in the estuary before transitioning from fresh water to salt water in the ocean. The administrative record documents that deepening the Columbia River channel over the past century and other changes to the structure of the river have “likely caused the largest changes in salinity intrusion and density stratification” in the estuary. Salmon at River’s End: The Role of the Estuary in the Decline and Recovery of Columbia River Salmon, NMFS Technical Memorandum at 99 (2001).
At issue is the accuracy of the Army Corps’ models for measuring impacts of channel deepening on salinity. Contrary to the majority’s assertion, NWEA does present a “cogent challenge” to the old model, the WES model, a challenge supported by the record. NOAA Fisheries warned the Corps not to use its older salinity model because of the imprecision of its measurements of salinity in the shallow areas that are favored by salmonids:
The WES model used to evaluate change in salinity intrusion was not validated nor designed to assess impacts in the shallow, side channel habitat used by salmon. In other words, an analysis is needed of whether the model would predict current conditions to determine credibility of its ability to predict future conditions. WES was not asked to conduct such an assessment, and commented that — if such validation were needed — a new model would have to be developed.
Notes from NOAA Fisheries meeting to “Clarify Science Issues for Columbia River Channel Deepening Project,” (Oct. 6, 2000).
Heeding this admonition, the Corps developed a new model for measuring salinity, the Oregon Health and Sciences University/Oregon Graduate Institute (OHSH/OGI) model. NWEA argues that the Corps rushed development of this new salinity model and used it only to evaluate the accuracy of the older model. Furthermore, the Corps appears to have employed the new salinity model too hastily, before its accuracy had been verified. The developers of the model expressed concern about-reliance on its results given the uncertainty surrounding them.10 However, *1162the Corps relied on the findings based on this highly uncertain new model to conclude that salinity would increase in the shallows by less than .5 ppt (parts per thousand) and that this would have no impact.
The district court apparently relied, not on evidence within the record, but upon its own Google search about the Corps’ salinity model. The Corps relied upon a model admittedly fraught with uncertainty to reach its conclusion that the channel deepening would produce no adverse impact on the salmonids from salinity in the estuaries. I suggest that studies employing an older, inadequate model followed by complete reliance on an untested model does not constitute the “hard look” required by NEPA.
IV. CONCLUSION
Fundamentally, the majority' takes an ostrich’s head-in-the-sand approach to reviewing the agency’s analysis, settling for the Corps’ explanation without undertaking the required review of its decision making. It is true, we are not permitted to substitute our judgment for the reasoned decision of the agency. Neither, however, are we permitted to rubber-stamp the agency’s decision of what factors must be considered and what factors need not be considered without taking a detailed look at whether the agency’s reasoning is sound. Here, it is not.
The “hard look” here went awry. The Corps, as it must, acknowledged profound consequences from erosion if large quantities of sand are removed from the littoral system. Anyone familiar with the Washington coastline has seen the devastation from past erosion (consequences the Corps admits were caused by its own past bad practices). The Corps acknowledges that it has designated a deep-water disposal site to hold huge quantities of dredge spoils, but has no plan of mitigation if that site is used for its intended purpose. Nor does the Corps analyze when and how much erosion is likely to occur — only that it will be profound and devastating. Its analysis of increased toxicity that may result from dredging is completely inadequate, as is its analysis of possible changes in salinity. Last, but certainly not least, the economic analysis is highly suspect.
My bottom line is that the Corps has substantially more work to do. Hence my dissent.
. Cf. Felicity Barringer, Senate Backs New Controls for Projects by Engineers, N.Y. TIMES, July 20, 2006, at A18 (detailing a Senate bill requiring independent panels of scientific and economic experts to evaluate potential Corps water projects); Editorial, A Chance to Reform the Corps, N.Y. TIMES, July 19, 2006, at A20 (praising the Senate bill and noting several studies finding that the Corps regularly inflated economic benefits and understated potential environmental costs).
. The Port at Grays Harbor has recently expanded the rail loop at its grain terminal, allowing for more efficient unloading of grain. In addition, AGP, a midwestern soybean exporter, built a new terminal at the port. The economic analysis undertaken by the Corps did not consider this expansion which may cut traffic to both lower-Columbia and Puget Sound ports. Port of Grays Harbor Grain Terminal Loop Track at http:// www.wsdot.wa. gov/Projects/Rail/grays — harbor/ (last visited June 26, 2006); see also http://www.portofgraysharbor.com/ (last visited June 26, 2006). The Port of Grays Harbor has the advantage of cutting short the many miles' travel in Puget Sound to the Ports of Seattle and Tacoma and the even longer journey to the ports of the Columbia (Portland, Vancouver, and Kelama).
I wonder what to make of the fact that the proposal to dredge Portland’s port on the Willamette River, where its docks are located, has been postponed indefinitely because of the toxicity of the Willamette. For the foreseeable future, large ships cannot go into Portland, except lightly loaded, whether or not the Columbia is dredged. I suggest this should also be part of the economic analysis of container ship traffic.
. FEIS, Exh. H at H-6.
. "[T]he quantity of dredged material that will be placed in proposed Site E and the North Jetty Site is uncertain due to the dynamics of the sites. Some quantity of dredged material will likely have to be placed in the proposed Deep Water Site each year; this will result in some material being permanently lost from the littoral zone.” FEIS, Exh. H at H-68.
. The majority calls use of this term a "misnomer.” Maj. Op. at 1136. The majority's breezy dismissal of the Corps' use of "intended” epitomizes the problems with the majority's analysis: on the one hand the Corps tells us that extensive use of the deepwater disposal site is a "worst-case” scenario; on the other, it tells us that the "intended” use of the site will lead to “adverse” effects. Which is it? The Corps' inconsistency and the ambiguity of its plans are at the heart of what troubles me so deeply about this case, and the majority's ready dismissal of those inconsistencies is inconsistent with NEPA.
. Factor 4 is "types and quantities of waste proposed to be disposed and proposed methods of release, including methods of packaging the waste, if any.” Factor 7 is "existence and effects of present or previous discharges and dumping in the area (including cumulative effects).” FEIS, Exh. H at H-55.
. Criteria c states that "[i]f at any time during or after disposal site evaluation studies, it is determined that existing disposal sites presently approved on an interim basis for ocean dumping do not meet criteria for site selection set forth in Sections 228.5-228.ó, the use of such sites will be terminated as soon as suitable alternative disposal sites can be designated.” Criteria d states that "[t]he sizes of ocean disposal sites will be limited in order to localize, for identification and control, any immediate adverse impacts and to permit the implementation of effective monitoring and surveillance programs to prevent adverse, long-range impacts. The size, configuration, and location of any disposal site will be determined as part of the disposal site evaluation or designation study.” FEIS, Exh. H at HISS.
. The toxicity of the Columbia River is much in the news. The federal government recently reached a limited agreement with a Canadian mining company, Teck Comineo, that is believed to be responsible for a century's worth of pollution of the stretch of the Columbia between the Canadian border and the Grand Coulee Dam. Although the agreement requires Teck Comineo to fund a short-term study of human health and the environment of Lake Roosevelt, tribal stakeholders and the State of Washington are expressing reservations and concern that the agreement will allow Teck Comineo to walk away from what they believe is the company’s obligation to pay for its pollution of the Columbia. Pollution Study Ordered: Teck Comineo Pact Targets Fouling of Columbia, Karen Dorn Steele, The Spokesman Review (June 3, 2006). In fact, this court just held that a law-suit seeking to hold Teck Comineo liable under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) could proceed because, although Teck Comineo is a Canadian company, it released hazardous substances into the United States. Pakootas v. Teck Cominco Metals, Ltd., 452 F.3d 1066 (9th Cir.2006).
. Following its assertion that side-slope sediment need not be tested, the majority holds, in the alternative, that the Corp has in fact tested side-slope sediments. It bases this holding on a pathetic sample size- — two samples from outside the navigation channel. To hold that two samples from outside the navigation channel, when dredging is planned for more than one hundred miles of the Columbia, satisfies NEPA standards borders on the absurd.
. Uncertainty as to the accuracy and adequacy of the model is compounded by the impacts of climate change on the Pacific Ocean and Columbia River — how will now-certain rising of sea level impact salinity of the estuarine lower Columbia River? We don't know because the Corps has not considered this significant factor.