On July 28, 2025, the Director of Civil Works for the United States Army Corps of Engineers (USACE), Eddie Belk, issued guidance for analyzing induced flooding. This guidance is intended to help USACE Districts and Divisions manage an analysis of induced flooding in planning studies (and those projects under design or construction) due to “recent court decisions on the takings implications of induced flooding.” While several cases affect how USACE thinks about this, the primary case that seems to be causing reevaluation is Ideker Farms v. United States, 71 F.4th 964 (Fed. Cir. 2023). This 2023 decision of the Federal Circuit appears to be causing USACE to think more conservatively about the possibility of USACE actions resulting in later “takings” of private property, and this is affecting the way in which USACE is thinking about real estate questions that come up in the context of planning studies.
The Fifth Amendment to the U.S. Constitution includes the Takings Clause, which instructs: “nor shall private property be taken for public use, without just compensation.” When performing planning studies, USACE has always examined the hydraulic impacts of a proposed project to determine whether third-party lands would be affected by a takings, and therefore whether those third parties needed to be compensated for the impacts. In fact, guidance on the books from 1996, 1998, and 2000 all confirm this approach. But in doing so, USACE had historically only looked at certain events to determine impacts. To create a simple example, if USACE’s modeling showed that in every 2-year event there was flooding of non-project lands, USACE would include a purchase of the affected land rights (either fee or easement) in the list of project costs to be considered for the benefit-cost ratio. In contrast, if USACE’s modeling showed that only in a 10,000-year event there was flooding of non-project lands, USACE would typically not include the purchase of those land rights as part of the project costs. USACE’s written guidance directed that a takings analysis should be prepared when “significant induced flooding” is anticipated, and USACE staff has reported that all events up to the 100-year event were generally used as the significance threshold.
As a result of the recent court decisions, including Ideker, USACE has now concluded that “it should not be assumed, in lieu of a takings analysis, that takings will not result merely because any induced flooding will be slight, or because the induced flooding is temporary or intermittent in nature.” As explained below, the guidance then proceeds to provide a framework for how to analyze these issues going forward. But to understand the guidance and to apply the judgment that is required by the new guidance, a proper understanding of takings law in the context of flooding is required.
The United States Supreme Court appears to have first considered whether flooding constitutes a taking in the case of Pumpelly v. Green Bay & Mississippi Canal Co. in 1871. In that case, the State of Wisconsin passed a law allowing for the raising of a dam, and the effect of the raise was to inundate the property of the plaintiff. Pumpelly stands for the proposition that where the government floods your land for the construction of a reservoir, lake, or navigable way, the land is “taken” even though you might still hold fee title to the property as demonstrated with a deed. Pumpelly was a clear case as the land was inundated 100% of the time, but it left open the question of what happens when land is only inundated part of the year. That question was answered by United States v. Cress. In Cress, a landowner owned property that was flooded for half of the year due to the construction of a lock, and the value of the property was cut in half due to the flooding. The property owner still held the deed to the property, still could use the property for part of the year, and had not lost the entire value of the property, and as a result the government argued that the seasonal flooding of the property did not constitute a taking. In response, the Court explained that a taking can occur where the flooding occurs during part of the year because that flooding was permanent and ongoing, year after year. Thus, for permanent inundations of lands, whether the land is flooded by a reservoir all of the time, or just some of the time, inundation from a dam or lock when caused by the government constitutes a taking.
In 2012, the Supreme Court had the chance to consider the case of Arkansas Game and Fish Commission v. United States. Whereas Pumpelly and Cress arose out of the construction of new infrastructure that created a permanent impact, Arkansas Fish and Game Commission arose out of regulatory changes in the operation of a dam. In Arkansas Fish and Game Commission, USACE reoperated dams to allow water to flow in the river at times that farmers needed the water, but those changed flows caused inundation and damages to trees owned by the State of Arkansas, causing USACE to revert the dam operations to the original operation that did not cause damage. Writing for a unanimous Court, Justice Ginsburg explained that the Court’s past decisions state that if “government action would qualify as a taking when permanently continued, temporary actions of the same character may also qualify as a taking.” The Court found “that recurrent floodings, even if of finite duration, are not categorically exempt from Takings Clause liability.” Thus, the Court said that if the government took action that resulted in flooding of land, and then took further actions to ensure that the flooding was only temporary (as opposed to permanently ongoing), the government could still be liable for a taking.
Pumpelly and Cress were analyzed as per se takings, meaning that if the land was inundated, it was an automatic taking. In contrast, in Arkansas Fish and Game Commission, the Court found that because the flooding was only temporary, the courts needed to consider a balancing text to determine if a takings occurred. The Court noted the following factors to be considered by the lower courts in evaluating liability:
- “time is indeed a factor in determining the existence … of a compensable taking.”
- “the degree to which the invasion is intended or is the foreseeable result of authorized government action.”
- “the character of the land at issue and the owner’s ‘reasonable investment-backed expectations’ regarding the land’s use.”
- the “[s]everity of the interference figures in the calculus as well.”
The Supreme Court remanded the case back to the lower courts to consider those factors. When the lower court applied those factors on remand, the Federal Circuit concluded that the government had caused a taking in the form of a temporary flowage easement.
This then takes us to Ideker. The Court in Ideker was faced with the following facts: a group of farmers brought suit when their lands were flooded in six specific years over an eight-year period when USACE changed the operation of a dam on the Missouri Mainstem System for the benefit of downstream fish and birds. While the Mainstem System was beneficial for economically productive land use, it had a deleterious effect on fish and wildlife, requiring USACE to work with the Fish and Wildlife Service as mandated by the Endangered Species Act. The Fish and Wildlife Service issued several biological opinions containing proposed operational changes for the Mainstem System to reduce impacts on affected fish and wildlife, but USACE declined to make those changes “because it determined [Fish and Wildlife Services’] proposed actions would exacerbate flood risks to land adjacent to the [Missouri] River.”
While USACE remained resistant to changing the operation of the Mainstem System, it ultimately adopted a new Master Manual for the operation of the Mainstem System in 2004, which was intended to restore the Missouri River to a more natural state as requested by Fish and Wildlife. While the previous Master Manual had recognized the prioritization of flood control over wildlife, the 2004 revision eliminated that prioritization. Following a drought that ended in 2006, significant rain fell in 2007, 2008, 2010, 2011, 2013, and 2014, causing damage to plaintiffs’ lands and leading to the filing of the case in 2014. On appeal, the court found a taking as a result of flooding in 2007, 2008, 2010, 2013, and 2014 (not 2011). The court first noted that Arkansas Fish and Game Commission dealt with temporary intermittent flooding (a period of seven years) that came to an end when the United States changed its operations and for which the Supreme Court confirmed it was a temporary, not permanent, taking. In contrast, Ideker Farms dealt with intermittent flooding for which the United States had not proposed to return to its former operations plan, and thus property will likely continue to be flooded into the future. The court in Ideker noted that the multi-factor test from Arkansas Fish and Game Commission was used in that case to determine whether the flooding rose to the level of a temporary takings, but the plaintiffs in Ideker were instead alleging a permanent taking (even if for only six of eight years), making those factors irrelevant. Thus, the Court in Ideker found that this case was a per se taking, such as in Pumpelly or Cress. Relevant to USACE’s new planning guidance, USACE has reviewed language from Ideker and concluded that, because takings liability can take place even when flooding is intermittent, USACE cannot automatically consider “significance” as the simple test to determine whether induced flooding for infrequent events would result in a takings. Instead, the guidance calls on planning staff to work with Office of Counsel, possibly in an iterative process, in which USACE evaluates potential inundation as alternatives are developed up to and including the selection of the TSP, or tentatively selected plan. Once hydraulic impacts are identified, staff is to consider a range of options, including structural mitigation measures to offset the increased risk, or property acquisition through easement or fee. The guidance acknowledges that whether mitigation is adopted or not, Counsel must still evaluate the hydraulic impacts and determine if a taking has occurred. Where additional measures are adopted, the non-Federal sponsor is to be informed.
The guidance acknowledges that the takings determination must be based on engineering analysis and further directs that maps be generated that shows differences in lands to be inundated, flood depths, flood velocities, and periods of inundation both pre-project and post-project for the considered alternatives. This information is to be provided to Counsel along with information on property uses, projected impact of flooding, and impacts on property values. The guidance then directs Counsel to prepare the legal opinion on whether induced flooding rises to the level of a taking of real property. The conclusion of the legal opinion will be used in the actual Feasibility Report, but the opinion itself will not be included. However, what the new USACE guidance does not do is advise Counsel on how to make the determination of whether a hydraulic impact is a taking. But understanding more about Ideker is helpful in determining how to do that analysis.
Interestingly, while the Court in Ideker did find a taking, it only found a taking occurred in five of the six years in which the Plaintiff’s property was flooded. For the five years in which liability was found, the sizes of the storms were large, but not especially uncommon. However, for 2011, the year in which liability was not found, the storm event was approximately a 500-year event. In other words, liability did not exist for the very infrequent 500-year event as the evidence in the case did not prove that even without the new Master Manual, flooding would not have occurred anyway. As noted by the trial court, “Flooding in 2011 occurred in both the upper and lower Basin because inflows exceeded the amount of storage available in the six reservoirs that make up the mainstem System, not because the Corps knew that would happen and then chose to protect one group of landowners over another.” Importantly as to the other five years where was liability was found, the trial court also noted that “[t]he United States concedes that none of these higher flows and the associated weather that caused them are an intervening cause that breaks the chain of causation presented by the plaintiffs regarding the System and River Changes.” As a result of this concession, USACE also seemed unable to argue that some of the other events might also have been caused significantly by the size of the storm, and not simply by the new Master Manual.
USACE’s new guidance is absolutely correct that one cannot simply look at events up to the 100-year event to determine if a taking has occurred. There is no connection between that threshold and the Constitutional prohibition against a taking. But the fact that there was no liability in Ideker for 2011 (a 500-year event) and the fact that USACE had made certain concessions in Ideker also does not mean that liability would automatically be found in the case of flooding for an intermittent event. Concluding that there would be liability just because some lands are hydraulically impacted would be jumping to a conclusion not justified by Ideker. Instead, USACE must look at a particular parcel of property, determine its use, and determine the hydraulic impacts of the proposed project on that property for a number of different storm frequencies, examining the change in depth of flooding, duration of flooding, and velocity. To the extent that a project impacts thousands of properties, the new guidance states that this analysis may be required for each and every one of them.
There is one more Supreme Court case that is relevant to analyzing these cases of potential takings. In Ideker, the Court considered the recent Supreme Court case of Cedar Point Nursery v. Hassid, 594 U.S. 139 (2021). Cedar Point was not a flood case but was instead a challenge to a California regulation that allowed a union to physically “take” a portion of the property of a company for a part of the year to allow union organizers to reach agricultural workers: up to three hours a day and for up to 120 days in a year. In it, the Supreme Court found that the regulation effected a physical taking because it allowed union organizers to come onto the property by permission of the State of California, which precluded the property owners from excluding the organizers from private space. It was a per se taking, even though it only took the property for a portion of the year, much like in Ideker. In responding to criticism about the decision, Chief Justice Roberts made clear that the Court’s holding did not eliminate “the distinction between trespass and takings.” Instead, Chief Justice Roberts explained that a mere occasional trespass would not constitute a taking. The Court in Ideker used this explanation to acknowledge that single events, an isolated event, or a mere occasional event may in fact be a trespass, and not a taking.
Thus, Ideker creates three clear approaches (i) temporary and not permanent flooding, under which the multi-factor test of Arkansas Fish and Game Commission is applied to determine if there is a taking, (ii) permanently recurring flooding that foreseeably or intentionally results from government action (including Pumpelly, Cress, and Ideker), which Ideker tells us to be a categorical or per se physical taking not requiring application of the multi-factor test from Arkansas Fish and Game Commission, and (iii) isolated physical invasions not undertaken pursuant to a granted right of access, which the Cedar Point Court explains can be a mere trespass.
This third category of analysis, that of a single event, an isolated event, or a mere occasional event, must be considered by USACE in its updated induced flooding analysis. For example, imagine a situation where a USACE alternative raises a levee that currently provides accredited protection from the 100-year event but which will now be accredited for the 200-year event. Only in the case of a 100-year to 200-year event will the studied alternative create a hydraulic impact of some level of induced flooding. If the unleveed lands across the channel from the existing levee are currently inundated in all events up to the 100-year event, then the engineering analysis presented to Counsel would show that the analyzed alternative would increase the depth of flooding, would potentially increase the period of inundation, and might increase the velocity of water over the already inundated property. But assuming a 50-year life span for the proposed project, this means that there is a very small chance that the lands would be impacted from an event larger than the 100-year event. This low likelihood appears to fit squarely in the bucket of an “isolated event” or a “more occasional event.” At the other end of the spectrum, one can imagine a circumstance where an alternative would generate significant new flooding of lands not historically subject to flood for a 125-year event. In such a case, the “mere occasional flooding” bucket would not apply and Ideker would be squarely on point.
Going forward, USACE should work with non-Federal sponsors to consider these fact-specific inquiries. Non-Federal sponsors can add both resources and perspectives to these processes and should have a role more than simply being “informed” at the end. The laws of the 50 states are different in terms of the impacts on real property and whether diminution of value leads to a taking, and if so, what property interest might be taken. For this reason, USACE must lean on its partners to work through a project-by-project analysis in true partnership.
[View source.]
