Ideker Farms, Inc. v. United States

In the United States Court of Federal Claims

No. 14-183L

(Filed: March 13, 2018)*

*Opinion originally filed under seal on February 23, 2018

)

IDEKER FARMS, INC., et al., )

) Fifth Amendment Taking; Missouri

Plaintiffs, ) River; Flooding; Liability; Causation;

) Foreseeability; and Severity.

v. )

)

THE UNITED STATES, )

)

Defendant. )

)

R. Dan Boulware, St. Joseph, MO, for plaintiffs. Edwin H. Smith, Seth C. Wright, and, R.

Todd Ehlert, St. Joseph, MO, and Benjamin D. Brown and Laura Alexander, Washington,

D.C., of counsel.

Terry M. Petrie, Environment and Natural Resources Division, U.S. Department of

Justice, Washington, D.C., with whom was Jeffrey H. Wood, Acting Assistant Attorney

General, for defendant. Jacqueline C. Brown, Laura W. Duncan, Carter F. Thurman,

and Daniela A. Arregui, Washington, D.C., of counsel.

TRIAL OPINION

FIRESTONE, Senior Judge

The pending action was brought by farmers, landowners, and business owners

from six states who claim a taking without just compensation in contravention of the

Fifth Amendment based on actions by the United States Army Corps of Engineers

(“Corps”) on the Missouri River. U.S. Const. amend. V. The plaintiffs claim that the

Corps has changed its management of the Missouri River and that these changes have

caused more flooding of their properties.

In order to manage the litigation, 44 plaintiffs were selected as representative or

“Bellwether” plaintiffs (“plaintiffs”). These plaintiffs own or farm properties that extend

from Bismarck, North Dakota to Leavenworth, Kansas. Various plaintiffs claim a taking

for flooding in 2007, 2008, 2010, 2011, 2013, and 2014. Together, these plaintiffs assert

takings claims based on over 100 flood events.

The litigation was also divided into two phases. Phase I was focused on the issue

of the United States’ liability. Each of the individual plaintiffs was called to testify or

present evidence to establish their property interest and the timing and approximate

duration of flooding on the relevant parcel of land. The court also heard testimony from

numerous expert witnesses and many federal government employees. These individuals

testified as to the changes the Corps has made to its management of the Missouri River,

whether the Corps’ changes have caused flooding or made flooding more severe for each

of the years at issue, and whether the flooding for the years at issue was a foreseeable

result of the Corps’ changes.

In Phase II of the litigation, the court will decide whether the United States has

any defenses to these plaintiffs’ claims and other legal and factual issues associated with

proving entitlement to just compensation. For those entitled to just compensation, the

court will also decide the appropriate amount of compensation.

Phase I of the trial began in Kansas City, Missouri on March 6, 2017 and was

moved to Washington, D.C. on April 24, 2017. The trial concluded on June 23, 2017 and

2

was resumed after post-trial briefing1 on November 13, 2017 for eight days of closing

arguments. During the 55 days of witness presentations, the court heard testimony from

over 95 witnesses and received over 3,250 exhibits into evidence.

Set forth below are the court’s findings of fact and conclusions of law for Phase I

of the trial. The opinion is organized into the following sections: I. Background Facts; II.

Legal Standards; III. Liability Findings (Expert Testimony); IV. Individual Plaintiffs; and

V. Conclusions.

I. Background Facts

1. The Missouri River Prior to Regulation

The Missouri River (“River”) travels 2,341 miles from its source in Three Forks,

Montana to its mouth near St. Louis, Missouri.2 PX16 at PLTF-00003114.3 The

Missouri River Basin includes most of the Great Plains and extends over 530,000 square

miles in ten states: Montana, Wyoming, North Dakota, South Dakota, Minnesota, Iowa,

Colorado, Nebraska, Kansas, and Missouri. PX99 at USACE0291936. Historically, the

River was largely “wide and shallow,” meandering across “a wide, unconstrained

1

The plaintiffs’ motion (ECF No. 396) to strike exhibit 1 to the United States’ response (ECF

No. 382-1) to plaintiffs’ post-trial brief is GRANTED. The court agrees with the plaintiffs that

the exhibit was in effect “briefing” and as such exceeded the page limit set by the court in its

July 21, 2017 post-trial order (ECF No. 370).

2

The Missouri River, which originally measured 2,546 miles in length, lost approximately 200

miles due to the intensive damming, straightening, and channelization that took place in the last

century. PX16 at PLTF-00003114.

3

Many of the background facts have been taken from the 2002 and 2011 reports of the National

Research Council whose members are taken from the National Academy of Science. PX16 and

PX17. In addition, many facts come from the Environmental Impact Statements that were

prepared in connection with various Corps activities on the River. See, e.g., PX99; PX110.

3

floodplain” resulting from constant bank erosion and deposition of sediment. PX390 at

USACE0465781; PX99 at USACE0291936-7. The River had “diverse wildlife habitats

within the meander belt and formed a natural Missouri River floodplain ecosystem that

included open shallow and deep waters, sandbars, wetlands, willow thickets, and riparian

woodlands.”4 PX99 at USACE0291936-7. This biodiversity was ensured by the River’s

transport and distribution of vast amounts of nutrient-rich sediment, which led to it being

known as the “Big Muddy.” PX16 at PLTF-00003159.

The River was known for its spring and summer rises due to snowmelt and rainfall

in the Plains (spring flooding) and in the Rocky Mountains (summer flooding). Id.

Historically, flooding was common and widespread on the Missouri River, drastically

impacting the appearance and functionality of the River, with “water spread[ing] across

its floodplain [thus] hydrologically connecting the channel[] to its floodplain and

backwaters[]” and creating new channels. PX16 at PLTF-00003158-9, PLTF-00003161-

5. Spring flooding tended to last “one to two weeks and was relatively localized,”

whereas summer flooding “lasted longer and inundated larger portions of the floodplain.”

Id. at PLTF-00003159. Although the River’s main channel was 1,000 to 10,000 feet

4

At the time of Lewis and Clark’s “Corps of Discovery” expedition, the Missouri River was

highly diverse, with a wide array of morphologies found in different parts of the River. In many

areas, the River “was a multichannel system, with a primary channel and often multiple

secondary channels . . . widespread bars, islands, and shallow sloughs[,]” while in others, it

comprised “natural levees, backwater lakes, large meander loops, oxbow lakes, and sandbars and

dunes[.]” PX17 at PLTF-00007916 (citing Hallberg et al. and Moody et al.). In addition,

various shallower channels and backwater habitats created “slower-moving waters [that were]

critical for the reproduction, shelter, and feeding of fish species[,]” while higher lands

encompassed “rich forests, prairie grasses, and thick underbrush that contained a myriad of plant

species.” PX16 at PLTF-00003165.

4

wide during normal flow periods, the width increased to 25,000 to 35,000 feet during

flooding, with the River “flow[ing] bluff-to-bluff and cover[ing] a width up to 17 miles”

in certain areas during heavy flooding. PX17 at PLTF-00007916.

2. Regulation of the Missouri River by the Federal Government

In the early twentieth century, the federal government determined that it was in the

national interest for the “wild, free-flowing” Missouri River “to be controlled for

purposes of human settlement and as a resource to support economic development.”

PX16 at PLTF-00003157, PLTF-00003098. To meet the demands of settlers and

minimize flooding, Congress adopted a series of laws starting with the Rivers and

Harbors Act in 1927, which regulated navigation, and the Flood Control Act (“FCA”) in

1917, “which placed flood control on equal footing with navigation within the Corps[.]”5

PX390 at USACE0465778; PX16 at PLTF-00003130.

As discussed in detail infra, flood control was to be achieved in part by controlling

flow into the River by constructing a series of interlocking dams and reservoirs with

controlled releases of water from Gavins Point Dam, the lowest of the six dams

constructed. PX99 at USACE0291944-5. In addition, the Corps would eventually

construct a series of federal levees to help contain flooding in sections of the River below

Gavins Point Dam. PX555 at USACE3590059, USACE3590183. The Corps also

constructed a series of structures within the River that were designed to stabilize the

5

Following the Great Mississippi River Flood of 1927, Congress passed the 1936 Flood Control

Act which made flooding a “federal responsibility” and created a national flood-control policy.

PX16 at PLTF-00003130.

5

River’s banks, limit erosion, and ensure that a deeper, self-scouring channel existed in the

center of the River to move flood waters more quickly through the River and allow for

navigation. PX16 at PLTF-00003129-30.

a. The Missouri River Mainstem Reservoir System

Construction of most of the dams and reservoirs by the Corps was authorized by

the 1944 FCA.6 The 1944 FCA identified six purposes that the construction and

operation of the Missouri River Mainstem Reservoir System (“System”) would serve:

flood risk reduction; enhanced navigation; generation of hydroelectric power; irrigation

and water quality; recreation; and fish and wildlife. Id. at PLTF-00003132-51. See also

PX390 at USACE0465778. The 1944 FCA also required the Corps to follow the broad

outlines of what is known as the “Pick–Sloan Plan.” PX17 at PLTF-00007903. Under

the Pick–Sloan Plan, the Corps was to operate the System to “reduce the river’s natural

hydrologic variability in order to provide a steady and reliable 9-foot deep navigation

channel[,]” making use of regulated storage of water in and releases from the reservoirs.

PX16 at PLTF-00003109. Congress strengthened the purpose of the Pick–Sloan Plan by

authorizing the Missouri River Bank Stabilization and Navigation Project (“BSNP”) in

the 1945 Rivers and Harbors Act so as to “facilitate navigation, control flooding, provide

water supplies, and meet other social and economic needs.” PX17 at PLTF-00007910.

The Corps constructed and operates six mainstem dams on the Missouri River:

6

The mainstem dams and reservoirs are those on the Missouri River itself as distinguished from

the dams and reservoirs that were also constructed on certain tributaries flowing into the River to

also assist in flood control and for other purposes. PX16 at PLTF-00003116; PX17 at PLTF-

00007891.

6

Fort Peck in northeastern Montana; Garrison in central North Dakota; Oahe, Big Bend,

and Fort Randall in South Dakota; and Gavins Point along the Nebraska and South

Dakota border. Tr. 6820:6-12, 6821:1-3. The System became fully operational in 1967

and is the largest reservoir storage system in the United States, with a total storage

capacity of 73.1 million acre-feet (“MAF”). PX390 at USACE0465783-4. Each of the

System’s reservoirs have four primary operating zones: a permanent pool at the bottom;

a carryover multiple use zone, which is the largest and is “designed to hold water that

can be used during periods of drought[;]” an annual flood control and multiple use zone;

and, at the top, an exclusive flood-control zone. Tr. 6832:15-6834:1. A secondary zone

is the surcharge zone, which is available in each reservoir in the space between the top

of the spillway gates in the closed position and the top of the spillway gates when all are

raised and releasing water underneath.7 Tr. 6836:20-24, 6843:2-24. The surcharge zone

has only been used in extreme situations when the reservoirs are filling beyond capacity.

Tr. 6844:23-6845:6. To date, the surcharge zone has only been used at Fort Peck and

Garrison Dams in 1975, 1997, and 2011. Tr. 6844:23-6845:14.

By design, the System reservoirs today are generally kept 75 percent full in order

to serve all of the FCA-authorized purposes. Tr. 6903:18-6905:12. Importantly, the

mainstem dams regulate only half of the Missouri River Basin. Tr. 6823:12-25. See

also DX3001-015. The dams cannot control the runoff from tributaries flowing into the

7

Each dam, depending on design, can release water from the hydropower facilities through

outlet tunnels or through a spillway when there are extreme releases. Tr. 6829:18-6831:1. On

dams that have spillways, the gates are raised and water passes underneath the gates, down the

spillway structure, and past the dam. Tr. 6841:15-6842:11.

7

River below the dams.8 Id. When heavy rainfall occurs in the downstream tributary

watersheds, the Corps can adjust releases from the mainstem dams to mitigate flooding.

Tr. 6825:10-24, 6826:22-6827:6.

The Corps’ operation of the six mainstem dams is formalized in a Master Manual

which contains “the Corps’ interpretation of its statutory responsibilities and operating

approaches developed in coordination with state agencies and other federal agencies.”

PX16 at PLTF-00003143-4. The Master Manual sets out the Corps’ basic objectives

and plans for operating the System for optimum fulfillment of the 1944 FCA uses. PX8

at FWS_00000095-7. To supplement the Master Manual, the Corps prepares a more

comprehensive Annual Operating Plan (“AOP”) every year. PX16 at PLTF-00003143-

4. The Master Manual’s storage allocations and release procedures are determined from

operational studies of historical River flows and historical flood and drought events.

PX3 at USACE0004001. With regard to flood control, the Great Flood of 1881 is used

for determining the flood control storage and release procedures. Id. at

USACE0003997-8.

To minimize the effects of flooding along the River, the Master Manual calls for

vacant intermediate storage space in each reservoir at the beginning of each annual flood

season, “with evacuation scheduled in such a manner that flood conditions will not be

significantly aggravated if at all possible[.]” PX3 at USACE0004040. See also PX16 at

PLTF-00003143 (quoting the 1979 Master Manual). This means that, historically, the

8

However, tributary dams help to control this flow. See PX17 at PLTF-00007894.

8

Corps has endeavored to have enough space in the reservoirs to store runoff during the

early and later spring snowpack melt and the spring rains to prevent upper Basin

flooding and to avoid releasing the resulting stored water into the lower Basin during

high uncontrolled tributary flows. PX5 at USACE0121591. The Corps’ System

operational objectives and requirements were set forth in Section IX-A of the 1979

Master Manual. PX3 at USACE0004040. Section IX-A also provided for allocation

among the zones. Id. The allocations for each of the zones were carried over to the

revised 2004/2006 Master Manual (later referred to as the new Master Manual), as

discussed infra. PX5 at USACE0121594-6.

In operating the System, the Corps takes into account the effects of the BSNP

structures in providing flood control due to the “interdependence between components”

and because “the overall performance of the infrastructure is critical for [R]iver corridor

and conveyance reliability.” PX555 at USACE3590139. In fact, the Corps’ operation

of the System and its operation and maintenance of the BSNP work hand-in-hand to

provide flood control. Tr. 3983:5-14, 6661:13-16. See also PX390 at USACE0465783.

In the 1979 Master Manual, the Corps expressly provided that flood control was

its first priority and that fish and wildlife were the last priority. PX3 at USCE0004040-

1. In the 1979 Master Manual, fish and wildlife were only to be given consideration

“insofar as possible, without serious interference with the [other] functions[.]” Id. at

USACE0004041. See also PX16 at PLTF-00003144. This order of priorities was

identified in a filing by the United States before the United States Supreme Court, in

which the United States represented that “[t]he 1979 Master Manual sets forth a ‘general

9

approach’ of priorities for the interests served by the Main Stem System. . . . It gives

flood control the highest priority[.]” PX34 at PLTF-00000550. As discussed infra, this

priority approach changed in 2004, when the new Master Manual was issued, and,

again, in 2006, when the Master Manual was revised.

b. The Missouri River Bank Stabilization and Navigation Project

The hydrology of the River was not only transformed by the mainstem reservoirs

and dams but also by the BSNP. Tr. 40:24-41:12; PX18 at PLTF-00005729-3; PX390

at USACE0465790-804. As noted above, the BSNP is a series of river-control

structures the Corps constructed to help with navigation and flood control. PX16 at

PLTF-00003129-30. With the BSNP, there is now a nine-foot deep by 300-foot wide

navigation channel between Sioux City, Iowa and the mouth of the River near St. Louis,

Missouri. PX390 at USACE0465785.

The BSNP achieved channelization and stabilization of the banks by using “an

intricate system of dikes and revetments . . . [which] were designed to provide a

continuous navigation channel without using locks and dams.” PX99 at

USACE0291939. The stabilization of the banks also proved to be indispensable in

providing flood control. Stable banks protected the River “from meander[ing][,] [thus]

promoting floodplain infrastructure.” PX390 at USACE0465802. As such, the BSNP is

an integral part of the Corps’ efforts to provide flood control, with “the BSNP and

related structures hav[ing] effectively become components of the flood risk management

system[.]” PX555 at USACE3590059. Channelization and bank stabilization, however,

have also resulted in disconnecting the Missouri River from its floodplain, altering the

10

River’s natural hydrograph. PX16 at PLTF-00003169; PX99 at USACE0291937.

Specifically, by narrowing and deepening the channel into a fixed location, the BSNP

structures have resulted in the loss of shallow water habitat (“SWH”). PX17 at PLTF-

00007921.

The BSNP structures (dikes and revetments) are commonly constructed of pilings

and rocks. PX390 at USACE0465785. Dikes extend from the bank into the River,

perpendicular or nearly perpendicular to the water flow. PX17 at PLTF-00007921. The

dikes constrict the River channel to a fixed width in order to maintain high-velocity

water flows and to protect the banks from erosion. Id. at PLTF-00007963-4.

Revetments are constructed parallel to the flow of the River, “on and along the channel

bank,” either to establish and protect a bank or to guide a flow consistent with a desired

alignment. Id. These structures have resulted in the narrowing of the Missouri River’s

channel “to as little as one-half to one-third of its original width.” PX17 at PLTF-

00007921.

The BSNP structures stabilized the banks of the River by providing erosion

control and directing the River’s flow to the center of the channel. PX16 at PLTF-

00003129-30. See also DX1089. This promoted scouring of the bottom of the channel,

causing it to grow narrower and deeper, thus increasing the River’s velocity and

resulting in a “self-scouring channel[.]” PX16 at PLTF-00003129-30. The BSNP also

led to the accretion of tillable farmland along the River because the sediment began to

accumulate on the River’s banks, thereby extending the land into what was previously

water. PX99 at USACE0291978. Many of the properties involved in this litigation

11

were created from the above-described accretion.

To further stabilize the River’s banks and channelize the River as part of the

BSNP, the Corps has also eliminated most of the chutes or side channels that existed

naturally in the lower Missouri River.9 PX17 at PLTF-0007905; PX390 at

USACE0465821. The chutes provided some of the channel’s natural lateral migration

that was vital to maintaining the Missouri River Basin ecosystem, but also contributed to

frequent flooding by allowing the River to meander into the floodplain. PX16 at PLTF-

00003159-60. The removal of chutes and side channels was necessary because they

promoted the River’s “braided pattern with no single, distinct river channel,” which

resulted in frequent flooding. PX17 at PLTF-00007903.

Finally, due to the changes to the River created by the BSNP, the River has lost its

natural flood pulses, i.e., “the predictable rising and falling of water in a natural river-

floodplain ecosystem as the principal agent controlling the adaptations of most of the

biota[,]” which is “essential to the health of river-floodplain ecosystems[.]” PX16 at

PLTF-00003162. The BSNP alterations to the fluctuations in the River stages not only

lessened the frequency of flooding, but also lessened the severity and shortened the

duration of the flooding. PX18 at PLTF-00005729. In this connection, the BSNP led to

an alteration in the normal drainage and seepage characteristics of the Missouri River

Basin.10 PX16 at PLTF-00003169-70; PX99 at USACE0291952. As a result, flooding

9

The lower Missouri River in this opinion refers to the portion of the River below the mainstem

dams.

10

“Degradation of the river channel disconnects the river channel from its floodplain. Channel

degradation not only makes it more difficult for the river to overflow its banks, but it also affects

12

from blocked drainage and seepage that had historically led to property damage was

significantly lessened due to “reduced fluctuations in the floodplain [ground]water table.”

PX16 at PLTF-00003169-70.

c. Fish and Wildlife Impacts from the Corps’ Actions Taken Prior

to 2004

While the Corps was very successful in transforming the River for flood control

and navigation purposes, according to the National Research Council (“NRC”) and the

United States Fish and Wildlife Service (“FWS”),11 the Corps’ actions resulted in

significant hydrological and geomorphological changes, primarily due to the loss of

“extreme high and extreme low flows[.]” PX16 at PLTF-00003165. See also PX8 at

FWS_00000136-7. Specifically, before 1900, the Missouri River used to transport

approximately “400 million metric tons per year of sediment from the interior United

States to coastal Louisiana.” PX17 at PLTF-00007909 (citing Meade and Moody, 2009).

The construction and operation of the System and BSNP structures altered the

hydrograph of the River, leading to sediment being scoured faster than it could be

replaced from upstream, thus causing the River bed to erode, a phenomenon known as

“degradation.” PX8 at FWS_00000136. This degradation greatly harmed Missouri River

the floodplain water table. Most importantly, the lack of flooding removes a source of periodic

recharge water for infiltration to the groundwater table. In addition, because the water table (an

alluvial aquifer) is hydrologically connected to the river channel itself, there is a consequent

lowering of this aquifer in association with the lowering (incision) of the river channel.” PX16

at PLTF-00003169.

11

The FWS is responsible for administering the Endangered Species Act (“ESA”). 16 U.S.C. §

1536. As discussed infra, under the ESA the FWS prepares biological opinions setting forth an

agency’s responsibilities to meet its obligations under the ESA. 16 U.S.C. §1538.

13

Basin wildlife, especially certain species of fish like the pallid sturgeon, because the

reduced amount of sediment, together with “the loss of channel chutes, oxbow lakes, and

wetlands[,]” destroyed the natural habitat of those species. PX8 at FWS_00000136-7

(citing Corps, 1981). Additionally, sediment was being trapped behind the dams,

reservoirs, and BSNP structures, “resulting in sediment imbalances and marked channel

incision” that further destroyed Missouri River Basin fish, wildlife, and habitats. PX16 at

PLTF-00003188. See also PX17 at PLTF-00007894, PLTF-00007905, PLTF-00007932;

PX8 at FWS_00000136.

The Corps’ actions are estimated to have led to the destruction of vast numbers of

acreage of Missouri River Basin fish and wildlife habitat, as well as the interruption of

breeding cues. PX16 at PLTF-00003105-6. River channelization, bank stabilization,

levee construction, and inundation have transformed roughly “3 million acres of natural

riverine and floodplain habitat” and have reduced the production of benthic

invertebrates―an important food source for the River’s native fishes―by 70 percent.

Id. Notably, of the 67 fish species native to the Missouri River, “51 are now listed as

rare, uncommon, and/or decreasing across all or part of their ranges.” Id. at PLTF-

00003106. The Corps’ actions have also interfered with habitat needed for various bird

species. Importantly, three species are currently on the federal Endangered Species

List―the Interior Least Tern (Sternula antillarum), the Piping Plover (Charadrius

melodus), and the Pallid Sturgeon (Scaphirhynchus albus)―due to the elimination of

14

what was once a rich, abundant ecosystem.12 Id.

When Congress became cognizant of the damage caused by the construction and

operation of a vast network of dams and related support structures across the United

States, it passed the Water Resources Development Act (“WRDA”) in 1986 to address

that harm. See, e.g., PX390 at USACE0465813; PX17 at PLTF-00007964; PX280 at

USACE1207214. Under the 1986 WRDA, Congress authorized the creation of the

Bank Stabilization and Navigation Fish and Wildlife Mitigation Project (“BSNFWMP”)

to address some of the damage that had been done to the Missouri River Basin

ecosystem. See, e.g., PX17 at PLTF-00007964-5; PX192 at USACE0266545; PX280 at

USACE1207214. The BSNFWMP is now part of the Missouri River Recovery Program

(“MRRP”) which, as discussed infra, is the Corps’ umbrella program for returning the

Missouri River to a more natural state to aid in the recovery of the Missouri River Basin

ecosystem. See, e.g., PX187 at FWS_00305189-90. See also PX110 at

USACE0005085. The BSNFWMP was aimed at mitigating the habitat losses caused

from construction of the BSNP structures by reconnecting the River to its floodplain

through the creation and restoration of habitat areas. The BSNFWMP provided

authority and direction to the Corps to complete projects to mitigate BSNP habitat losses

by returning the River to a more natural state.13 See, e.g., PX94 (report to Congress

12

The 2002 NRC report contains a lengthy description of how the Corps’ actions have led to the

destruction of habitat and threatened and endangered species. As discussed infra, the destruction

is also catalogued in the various Biological Opinions prepared by the FWS. See, e.g., PX9-B;

PX10.

13

The Corps constructed various projects under the BSNFWMP to create and restore habitat.

The Corps also commenced a review of the 1979 Master Manual in 1989 as part of its

15

presenting the results of a study undertaken by the Corps to analyze the cost of

mitigation projects under the WRDAs).

In 1999, the WRDA reauthorized the BSNFWMP and gave the Corps authority

to purchase 166,750 acres from willing sellers, including some of the plaintiffs in this

case, (“Willing Sellers Program”) along the Missouri River in order to replace lost

habitat. The area authorized represented approximately one-third of the habitat lost due

to the Corps’ regulation of the River. PX280 at USACE1207213-5; PX390 at

USACE0465813. Land acquired under the Willing Sellers Program was to be converted

to habitat for native Missouri River species to compensate for the habitat that had been

lost as a result of the BSNP. PX192 at CLMT0304-00000186.

The Willing Sellers Program was the preferred choice for helping to restore the

habitat lost, even though the Corps could have used eminent domain to acquire

easements over private land in the targeted acreage. See, e.g., PX47 at

USACE0000222-6; PX192 at USACE0266545; PX280 at USACE1207215. The

Willing Sellers Program has had some success. As of September 30, 2009, the Corps

has acquired 56,606 acres of the targeted acreage. See PX280 at USACE1207215;

PX610.

The BSNFWMP was further expanded by Congress in the 2003 WRDA, which

was aimed at developing shallow water habitat (“SWH”) for native Missouri River

Endangered Species Act obligations to address the changes that could be made to the System’s

operation to ameliorate some of the damage caused to the Missouri River Basin ecosystem, in

particular in terms of threatened and endangered species.

16

Basin aquatic species, in particular the pallid sturgeon, and emergent sandbar habitat

(“ESH”) for native Missouri River Basin bird species. PX277 at USACE0719805;

PX17 at PLTF-00007957. The SWH program currently extends from Ponca, Nebraska

downstream to the Missouri River’s mouth near St. Louis, Missouri and it aims at

having “in place 20-30 acres of SWH per river mile by 2024.” PX277 at

USACE0719805. The ESH program aimed at having ESH along the entire stretch of

the Missouri River. Specifically, the ESH program’s goal was to have 50 acres of ESH

per river mile below Garrison Dam, 20 acres of ESH per river mile below Fort Randall

Dam, 80 acres of ESH per river mile at Lewis and Clark Lake, and 80 acres of ESH per

river mile below Gavins Point Dam by 2015. PX174 at PLTF-00005901.

3. U.S. Fish and Wildlife Biological Opinions

In addition to responding to the mandates of the WRDA, the Corps was also

being pressured by the FWS to address the harm to the Missouri River Basin ecosystem

the Corps had caused in operating the Missouri River Mainstem Reservoir and Dam

System and in constructing the BSNP. Under Section 7 of the Endangered Species Act

(“ESA”), 16 U.S.C. § 1531 et seq., all federal agencies that permit, fund, or carry out

activities involving fish and wildlife must consult with the FWS to ensure that their

actions will not jeopardize the continued existence of any listed species.14 PX33 at

14

“Jeopardize the continued existence of a listed species” is defined as “to engage in an action

that reasonably would be expected, directly or indirectly, to reduce appreciably the likelihood of

both the survival and recovery of a listed species in the wild by reducing the reproduction,

numbers, or distribution of that species.” 50 C.F.R. § 402.02.

17

PLTF-00001365. Under Section 7, the Corps may be required to provide a biological

assessment (“BA”) to the FWS, assessing the impact of its actions on threatened and

endangered (“T&E”) species. Id. After filing a BA, the Corps is required to enter into

Section 7 formal consultation with the FWS to determine how best to avoid jeopardy to

the T&E species or prevent destruction or adverse modification of their critical habitat.

PX33 at PLTF-00001365. Thereafter, the FWS is required to issue a Biological Opinion

(“BiOp”) that recommends reasonable and prudent alternatives (“RPA”) to avoid further

jeopardy. Id. See also PX8; PX9-B; PX10. The Corps must follow the BiOp or offer

some alternative measures to comply with the ESA. PX33 at PLTF-00001365. As part

of a BiOp, the FWS will often include an incidental take statement, which allows an

agency, like the Corps, to harm (known as “take”) a limited number of protected species

without triggering the civil or criminal liability provisions of the ESA. 16 U.S.C. §

1539. See also PX33 at PLTF-00001365-6.

Through the issuance of a series of BiOps, starting in 1990, the Corps began

negotiations with the FWS over the changes it would need to make to its System

operations and the BSNP structures in order to prevent further damage to T&E species

and to comply with the ESA. See PX8; PX9-B; PX10. In the 1990 BiOp, the FWS

concluded that the Corps’ “operations of the System [were] likely to jeopardize the

continued existence of the endangered interior least tern . . . and the threatened piping

plover . . . because operations eliminate[d] essential nesting habitat and could result in

the loss of at least 12 percent of the . . . interior least tern population and 22 percent of

the . . . piping plover population[.]” PX8 at FWS_00000092. The FWS also made six

18

recommendations to the Corps, which the Corps never adopted. Id. at FWS_00000144-

8; PX230.

The Corps’ failure to accept the recommendations in the 1990 BiOp led to a new

ESA Section 7 consultation with the FWS about other possible RPA measures the Corps

could take to comply with the ESA. PX230. This consultation gave rise to the 2000

BiOp, which “repeatedly acknowledged that both flow changes and habitat creation

[were] necessary to prevent jeopardy to [T&E] species.” PX117 at PLTF-00000517.

See also PX9-B. The 2000 BiOp recommended an elaborate RPA program “involving a

combination of reservoir operational changes, structural modifications, and non-

structural actions[,]” such as hydrograph and temperature changes, chute restoration,

and floodplain acquisition or easements. PX117 at PLTF-00000517; PX9-B at

FWS_00029104-6, FWS_00029382.

The Corps was also resistant to adopting the FWS’ 2000 BiOp RPAs, believing

the RPAs would cause more flooding. PX231. Specifically, the Corps was concerned

that under the BiOp, there would be more and different types of flooding. PX645. The

Corps was particularly averse to implementing flow modifications recommended by the

FWS because it was concerned they would “pose significant effects to the natural and/or

human environment.” PX15 at USACE0001757. Rather, the Corps proposed a

combination of flow and non-flow measures to meet its ESA obligations, PX15 at

USACE0001762-74, stating:

The Corps is committed to reconnecting the river to its floodplain

wherever possible; however, several conditions must be met to ensure the

goals can be attained. These include: [a]cquisition of necessary real estate

19

interests (willing seller requirement and easements); [r]eceipt of

appropriate funds; [l]and acquired must allow floodplain reconnection

without induced damages to neighboring lands; and [e]xisting project

purposes such as navigation and flood control are not adversely impacted.

Id. at USACE0001854.

The Corp’s proposal was set forth in a 2003 BA. In response, the FWS issued its

2003 BiOp, providing for a “multi-faceted approach[]” to attaining the habitat

objectives, which included a combination of changes to the System and BSNP. PX118

at PLTF-00008890. See also PX10. This back and forth between the Corps and FWS

continued into 2004 with the two agencies still not able to reach an agreement as to the

System and BSNP changes necessary to meet the Corps’ obligations under the ESA.

4. Multi-District Court Litigation

Numerous lawsuits were brought against the Corps by various Missouri River

Basin states, environmental groups, and other Missouri River Basin stakeholders to

force the Corps to make changes to its management of the Missouri River. In 2002, the

Eighth Circuit addressed some of those claims in South Dakota v. Ubbelohde, 330 F.3d

1014 (8th Cir. 2003). In the meantime, American Rivers had filed a lawsuit seeking the

Corps’ compliance with the ESA. American Rivers v. USACE, 271 F. Supp.2d 230

(D.D.C. 2003). Ultimately, the pending cases against the Corps, including American

Rivers and Ubbelohde, were consolidated by the Federal Judicial Panel on Multi-District

Litigation before the U.S. District Court for the District of Minnesota. In re Operation

of the Mo. River Sys. Litig., 363 F. Supp.2d 1145 (D. Minn. 2004) (“MDL case”). The

purpose of the consolidation was to determine, inter alia, whether the Corps had to

comply with the ESA by following the 2003 FWS BiOp. Id.

20

The MDL Court recognized that the Corps’ simultaneous compliance with the

Flood Control Act (“FCA”) and the ESA was problematic on its face because

management actions that benefited one would likely and predictably harm the other. Id.

at 1175. Nevertheless, the Court ultimately denied the Corps an extension to try and

reconcile its obligations under both statutes. PX108 (In re Operation of the Mo. River

Sys. Litig., 305 F. Supp.2d 1096 (D. Minn. 2004)). As a result, on February 26, 2004,

the Court ordered the Corps to revise its 1979 Master Manual by March 19, 2004, to

address the 2003 BiOp directives to come into compliance with the ESA. Id.

As ordered by the MDL Court, on March 19, 2004, Brigadier General William T.

Grisoli, who was the Corps General responsible for the Corps’ management of the

Missouri River, signed the Record of Decision (“ROD”), which, inter alia, adopted a

new Master Manual. PX114. Before signing the ROD, General Grisoli reviewed the

March 2004 Final Environmental Impact Statement (“FEIS”) the Corps had prepared to

evaluate the changes to the Master Manual. Id. In the ROD, the Corps expressly stated

that both the 2004 FEIS and the 1944 FCA did not assume a priority of purposes in the

operation of the System, but it recognized that “there may be occasions where conflicts

[will] exist between the individual authorized purposes.” Id. at PLTF-00000537. To

ensure compliance with the ESA, the Corps committed to implementing all elements of

the 2003 BiOp and agreed to changes in the operation of the System and the operation

and maintenance of the BSNP. PX116. See also PX799.

5. The Corps System and River Changes After 2004

This litigation arises from the changes the Corps has made to its operation of the

21

Mainstem Reservoir and Dam System, hereafter “System Changes,” and the changes

made to the BSNP and under the MRRP, hereafter “River Changes,” to meet its ESA

obligations under the 2003 BiOp. Together, these changes are referred to as “System and

River Changes.” Set forth below is a description of those System and River Changes.

The court’s findings regarding whether these System and River Changes caused flooding

or made flooding more severe than would have otherwise occurred without the System

and River Changes is discussed in the Liability Findings (Expert Testimony) section of

this opinion. The court’s findings regarding the impact of flooding on the individual

plaintiffs are in the section of the opinion on Individual Plaintiffs.

The System and River Changes that have been made to comply with the ESA are

numerous and are constantly evaluated and evolving. The new Master Manual issued on

March 19, 2004 provided that “[d]ecisions concerning implementation of additional

measures or modification of existing measures, including potential release changes out of

Gavins Point Dam, [would] be made through the adaptive management process.” PX114

at PLTF-00000540. Importantly, to comply with the ESA, the Corps must always

consider the impact of other FCA-authorized purposes on fish and wildlife because “[t]he

Endangered Species Act has a higher precedence than other authorized purposes.”

PX324 at USACE0328103. Thus, adaptive management does not give the Corps

unfettered flexibility. Rather, it is an approach that “promotes carefully-designed

management actions, assessment of these actions’ impacts, and subsequent policy

adjustments.” PX16 at PLTF-00003210. It aims at “maintain[ing] or restor[ing]

ecosystem resilience,” i.e., an ecosystem’s ability “to persist and adapt over time in the

22

face of natural and anthropogenic challenges.” Id. For example, the creation and

monitoring of each SWH and ESH project is a component of the adaptive management

approach adopted by the Corps.

The Corps’ commitment to complying with the ESA is also reflected in the

changes made to the Corps’ operating budget. PX701. See also PX860; PX860-A;

PX860-B. Prior to the ROD, the Corps was spending proportionally far more on the

other FCA-authorized purposes than on fish and wildlife. However, compliance with the

2003 BiOp under the ROD represented a significant change in the Corps’ management of

the River as reflected by its spending. PX701. See also PX860; PX860-A; PX860-B.

For example, the Corps’ fish and wildlife mitigation budget increased from $17.5 million

in 2003 to $69 million and $82.8 million in 2005 and 2006, respectively. PX860-A. This

represents a significant change in the focus of the work the Corps was doing in managing

the River—from flood control to River restoration work. PX701. See also PX860;

PX860-A; PX860-B.

a. System Changes after 2004

The Master Manual issued in 2004 to comply with the ESA had called for greater

releases in the Spring (“Spring Pulse”) than the Corps was prepared to make and so the

Corps instead, with the FWS’ approval, embarked on an “unprecedented” SWH

construction program in 2004 as an alternative means of promoting ESA compliance.

PX123 at USACE0510835. See also Tr. 217:5-24, 490:13-491:9, 2374:11-2379:16,

3737:7-25, 4283:3-16, 8272:6-8273:24, 10371:9-16; PX114; PX115. The Master Manual

was revised in 2006 to reflect the Corps’ approach. Both the 2004 and 2006 Master

23

Manuals, hereafter the “new Master Manual,” struck the language in the 1979 Master

Manual providing a sequential priority of the FCA-authorized purposes in operating the

System. Instead of giving flood control first priority and fish and wildlife last priority,

the new Section VII.7-01 of the Master Manual provides that, in operating the System,

the Corps will “balance [the FCA] functions in order to obtain the optimum development

and utilization of the water resources of the Missouri River basin to best serve the needs

of the people.” PX4 at USACE0002644; PX196.

The new Master Manual also contains two significant operational guidelines that

are at issue in this litigation. First, the new Master Manual authorizes the Corps to keep a

larger amount of water in the reservoirs for the benefit of other purposes, including fish

and wildlife. PX117-A at PLTF-00008836; PX756. In this connection, the Corps

acknowledges that during years of high early runoff from rain and snowpack melt above

the System dams, if the System does not have enough storage to impound all of the

runoff, the Corps may have to choose between making higher early releases, even if that

would likely wash away nesting birds and contribute to early flooding downstream, or

holding more water in the reservoirs and hope that spring rains are below normal. See,

e.g., PX10; Tr. 4620:5-4626:22.

Second, the new Master Manual addresses the need to return the River to having

more varied river stages for the benefit of T&E species. Under the new Master

Manual, water is released from the dams when needed to prevent the least tern and

piping plover from nesting on low-lying areas, which can be later washed away, and to

provide spawning cues for the pallid sturgeon, hereinafter referred to as “T&E

24

releases.” Tr. 6959:11-16, 6964:24-6965:5. Although the Corps is not required to

conduct the once-proposed Spring Pulse, the Corps consults with the FWS throughout

the nesting season to determine the extent of releases necessary to comply with the

BiOp.

As discussed infra, for each of the years of flooding, except for 2011, the Corps

made T&E releases during periods of high River flows with the knowledge that flooding

was taking place or likely to soon occur.15 In 2010, for example, the Corps made

releases for T&E species despite receiving direct communications from Corps personnel

regarding flooding already in progress in the lower River. PX967. The response from

Corps personnel responsible for the System was that the releases were needed for the

benefit of the least tern and piping plover. Id. This happened again in 2015. PX2308.

b. River Changes After 2004

To coordinate the various Corps projects needed to restore the River to a more

natural state, i.e., before it was so highly engineered, the Corps established the Missouri

River Recovery Program (“MRRP”). PX110 at USACE0005085. In carrying out the

objectives of the MRRP, the Corps, as it does with its T&E release decisions, has

partnered closely with the FWS. See, e.g., PX187; PX911 at PLTF-00018455; DX576

at USACE0089347. The MRRP coordinates the Corps’ authority under: (1) the 1986

15

The government presented expert testimony from Mr. Woodbury who calculated the

percentage of flow attributable to releases from Gavins Point Dam during the relevant periods of

flooding in 2007, 2008, 2010, 2013, and 2014 at key gages below Gavins Point Dam. In 2007

and 2008, contributions from Gavins Point Dam ranged along the River from five to fifteen

percent. In 2010, the range was from 15 to 20 percent. For 2013 and 2014, the contributions

ranged from 10 to 30 percent. See Def.’s Br. at 99-100; DX3015.

25

and 1999 WRDAs; (2) the 2003 BiOp; and (3) the 2007 WRDA. PX390 at

USACE0465813. As part of the 2007 WRDA, Congress authorized the Missouri River

Recovery Implementation Committee (“MRRIC”) to provide guidance to the Corps

from the affected community on its MRRP activities and projects. PX17 at PLTF-

00007956.

To accomplish the MRRP objective of restoring the Missouri River to a more

natural state, the Corps has modified BSNP structures and has reopened previously

closed chutes to create shallow water habitat (“SWH”). As explained infra, Corps

studies explain that as of 2014, the Corps had undertaken 1,697 dike notching actions,

354 major modification actions, 63 dike lowering actions, 36 dike extension actions, 39

side-channel chute actions, 20 revetment chute actions, 14 backwater actions, and 3

channel widening actions. PX277 at USACE0719815.

The Corps has also undertaken work to reopen the natural chutes it had removed

years ago. PX169. The Corps has acknowledged that in undertaking these BSNP

modifications, it is destabilizing the banks of the River, allowing it to meander into the

floodplain, as it had been allowed to do in its natural state.16 See, e.g., PX98 at

USACE0090827; PX390; PX911. The Corps has also recognized that in reopening

chutes, there are potential flooding impacts. See, e.g., PX911 at PLTF-00018460-1. See

also Tr. 9243:7-9245:20, 9499:7-9501:4, 9538:8-9540:23, 10872:16-10873:1. In the

16

“To achieve significant amounts of SWH, it will be necessary to erode real estate to increase

the river’s top-width.” PX98 at USACE0090827.

26

environmental reviews for reopening chutes, however, the Corps has clearly stated the

need for minimizing flooding impacts. See, e.g., DX0286 at USACE5031174,

USACE5031192.

As discussed in greater detail infra, the creation of SWH “involves reclaiming

areas that were historically part of the active [R]iver channel but were converted to land

by the sediments trapped by the BSNP.” PX390 at USACE0465814. Because these

endangered species “depend on bare sandbars for successful nesting and fledgling[,]”

which are limited along the Missouri River, the Corps has been increasing the amount of

ESH “by creating new sandbars largely from dredged material and by clearing

vegetation from existing sandbars.” PX17 at PLTF-00007965-6.

As also discussed in detail infra, the evidence has established that River Changes

by the Corps, such as the aggradation of sediment from notching and the degradation of

dikes and revetments, have had the effect of raising the Missouri River’s water surface

elevations (“WSEs”) in periods of high flows. See, e.g., PX2089-A; Tr. 4754:22-

4777:14, 4793:23-4794:23, 4838:9-4839:19, 10544:19-10545:18. In 2011, the NRC

reported that both SWH and ESH programs affect the amount of sediment transported

and thus have repercussions “for sediment loadings and transport and therefore for

channel morphology and habitat maintenance.” PX17 at PLTF-00007979. More

specifically, the NRC reported that although sediment transported across the Missouri

River Basin “has value for habitat formation[,] [it] has both positive and negative

influences on infrastructure.” Id. The NRC explained that the BSNP “structures make it

difficult to establish and maintain complex channel and floodplain habitats that depend

27

on the temporary storage and frequent alteration of sediment accumulations in the form

of channel bars and off-channel water bodies.” Id. at PLTF-00007981-2. This is

because these structures are aimed at facilitating “a narrower and more rapid flow that

scours a deeper and less complex channel[,]” resulting in greater flood control. Id. at

PLTF-00007982. As such, the NRC reported that changing the River to create SWH

and ESH will likely lead to increased flooding of homes, farms, and infrastructure

(transportation, telecommunications, power) within the floodplain. Id. at PLTF-

00007982. The NRC recognized that the impacts on homes and farms would likely be

greatest during high-water events. Id. See also PX911. The NRC concluded that “a

reversal of tradeoffs that would favor ecosystem restoration may . . . [result in] winners

and losers in a new operations scheme who will need to be carefully considered and

perhaps compensated.” PX16 at PLTF-00003209 (emphasis added).

6. Flooding on the Missouri River After 2004

As discussed above, in its pre-regulated state, the Missouri River regularly

flooded. The periods of flooding after the System and BSNP structures were

constructed have been less frequent and less severe. Between 1967, when the System

was completed, and 2004, there was some limited flooding in the 1980s and one

significant flood in 1993. The period from 2000 to 2006 was largely a period of drought

on the River. Flooding on the River returned, starting in 2007. In fact, 2007, 2008,

2010, 2011, 2013, and 2014 have been among the worst flooding years in the River’s

history. PX2008. The evidence established that 2009 and 2012 were drought years.

The plaintiffs’ takings claims are based on flooding in 2007, 2008, 2010, 2011, 2013,

28

and 2014. The year 2014 was selected as the cut-off year for purposes of proving

flooding by the Corps’ System and River Changes; however, some plaintiffs have

continued to experience flooding. The plaintiffs claim that there is now a pattern of

increased flooding caused by the Corps’ System and River Changes. The court allowed

testimony regarding flooding in 2015 and 2016 for purposes of establishing a pattern.

Set forth below is a brief description of the flooding on the Missouri River for the

years in question. There is no dispute that the flooding at issue in this litigation has

occurred in the context of high River flows that were naturally occurring during periods

of high precipitation or high tributary inflows, except for the flood in 2011. The 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. Tr. 14293:16-24. The government

contends that none of the Corps’ actions have caused higher WSEs on the River than

would have existed without these actions. The plaintiffs contend that the System and

River Changes have caused higher WSEs on the River than would have existed had the

System and River Changes not been made. The plaintiffs also contend that the higher

WSEs were a foreseeable consequence of the System and River Changes and led to the

flooding and takings claims at issue in this case.

a. Flooding in 2007

Out of the 113 year historical record the May 2007 regional precipitation in

South Dakota ranked 104 wettest and in North Dakota ranked 108 wettest. DX3015-28.

However, the System storage on March 1st was 22.7 million acre feet (“MAF”) below

29

the base of the annual flood control zone, and the upper three reservoirs were 31 to 34

feet below normal. Tr. 7157:3-8; DX0480. That year, the Corps conducted System

releases for the benefit of T&E species, which added flow to the River during a period

when the River was lower.17 Tr. 4557:24-4558:1, 7160:4-20; DX3001-243.

b. Flooding in 2008

In April through June 2008, Iowa experienced very heavy rainfall leading to the

wettest record in 114 years. Tr. 10796:20-10797:5; DX3015-29. Heavy and

unprecedented rains were also recorded in Montana, South Dakota, Nebraska, and

Missouri. Tr. 10797:6-13, 13500:12-14; DX3015-29; DX0094. System storage on

March 1st was up from 2007, but still below the normal start of the runoff season, and

the upper three reservoirs were 26 to 35 feet below normal levels. Tr. 7161:15-20;

DX0481. The Corps conducted System releases for the benefit of T&E species that

were higher than those in 2007. Tr. 4574:9-16, 7163:18-7164:5; DX3001-247. These

releases also added flow during a period of lower Basin flooding.18

c. Flooding in 2010

During the spring and summer of 2010, the Missouri River and its tributaries

experienced extensive flooding caused by heavy spring rainfall and heavy plains

snowpack. Tr. 1292:7-12, 1561: 21-23, 2845:22-2846:2, 2967:15-19, 3505:15-20,

3452:5-7, 7187:9-7189:16. There was very heavy rainfall in Iowa, Nebraska, and South

17

See fn. 15. In 2007, the releases from Gavins Point Dam contributed five to ten percent of the

flow.

18

See fn. 15. In 2008, the releases from Gavins Point Dam contributed five to fifteen percent of

the flow.

30

Dakota from May through July 2010. Tr. 10797:14-10798:2; DX3015-30. As a result,

runoff during 2010 was 156 percent above normal, generating the third highest runoff at

that time in the last 65 years in St. Joseph. Tr. 10806:14-15; DX0483 at

USACE0014563. Nevertheless, the Corps conducted T&E releases which contributed

to the flooding.19

d. Flooding in 2011

In 2011, runoff and rainfall in the upper Basin was unprecedented in magnitude

and duration. The untimely combination of events caused record WSEs and extensive

flooding in the upper and lower Basin from June through August. Tr. 13858:16-

13859:2. During the winter and spring of 2011, record snow fell across a large portion

of the northern Rocky Mountains and eastward into the Northern Plains. Id. See also

Tr. 10798:14-10799:9; DX3015-32. A cool spring held snowpack in place later than

usual, and a rapid snowmelt coincided with record-setting rains in May and early June

over Montana and western North Dakota. Tr. 13843:1-13848:2, 13853:1-6; DX3007-

07. Despite summer drought conditions in the lower Basin, it was extremely wet in

portions of Montana, Wyoming, Colorado, and the Dakotas. Tr. 13500:20-13501:2;

DX0094. In addition to record highs in mountain snow water equivalent, May

precipitation was 400 percent above average in the Basin, while runoff was 320 percent

above average above Sioux City. Tr. 13843:1-13844:4, 13847:13-13849:14; DX3007-

07; DX0192 at DX0192-0044. The high snowpack and delayed snowmelt contributed

19

See fn. 15. In 2010, the releases from Gavins Point Dam contributed 15 to 20 percent of the

flow.

31

to a May to July runoff in the upper Basin that was twenty percent higher than the

previously highest runoff ever recorded in 1997. Tr. 13853:1-6; DX0192 at DX0192-

0054, DX0192-0060, DX0192-0067. The runoff in 2011, totaling 61 MAF, was the

highest in the upper Basin since 1898, at 247 percent above normal levels. Tr. 7115:25-

7116:5; DX0484 at USACE0014651. Upper Basin runoff was 25 percent greater than

the next highest runoff year, i.e., 1997. Tr. 7117:16-7117:25.

Although the System storage on March 1st was 57.6 MAF, all 2010 runoff had

been released and the runoff season was at the base of the annual flood control zone by

January 27th. Tr. 7098:8-17, 7137:13-7138:17; DX0484. In 2011, there were no T&E

releases because the Corps had downstream flood control concerns. Tr. 7122:8-22. The

massive releases in 2011 starting in May were taken when Corps personnel determined

that the record runoff levels in the upper Basin were putting the entire mainstem System

in jeopardy. In response, the Corps released a record 160,000 cubic feet per second

(“cfs”) from Gavins Point Dam for 65 days with higher than normal releases continuing

until October. DX0192 at DX0192-0047-8. Previously, the record had been 70,000 cfs

released in 1997. Tr. 7392:13-15.

e. Flooding in 2013

A drought developed in 2012 that remained over much of the Missouri River

Basin in early 2013. Tr. 7202:1-4. System storage on March 1st was 8.3 MAF below

the normal start of the runoff season, and the upper three reservoirs were 10 to 12 feet

below normal levels. Tr. 7202:11-17; DX0484. Heavy localized rainfall later in the

year in Nebraska, Missouri, Kansas, and Iowa resulted in high flows downstream of

32

Nebraska City.20 Tr. 7203:8-20. Despite these higher downstream flows in late May

and early June, the Corps continued to make T&E releases which contributed to the flow

during the highest flow periods. DX3015-227; DX3015-308; DX3015-406; DX3015-

561.21

f. Flooding in 2014

In 2014, there was high runoff in the upper Basin and low runoff in the reach

between Sioux City and St. Louis. Tr. 7205:2-18; DX3001-276. June, however, saw

heavy rainfall in the lower Basin, with the Sioux City runoff being 140 percent above

normal levels. Tr. 7205:14-15, 7206:22-7207:9; DX0536 at USACE7739484; DX0094.

The System storage on March 1st was 5.4 MAF below the normal start of the runoff

season, and the upper three reservoirs were five to 12 feet below normal levels. Tr.

7205:20-24; DX0536. Although the Corps reduced Gavins Point Dam releases to a

minimum 10,000 cfs in response to the June rainfall, it maintained its T&E releases,

which coincided with the highest downstream flows. Tr. 4649:22-4651:1.22

II. Legal Standards

1. Causation

20

For instance, the Big Nehama River had the second highest flows of record. Tr. 7203:8-14;

DX3001-275. May rainfall in the lower Basin was also above normal. Tr. 10799:12-14;

DX3015-33; DX0094.

21

See fn. 15. In 2013, the releases from Gavins Point Dam contributed 20 to 30 percent of the

flow.

22

See fn. 15. In 2014, the releases from Gavins Point Dam contributed 10 to 20 percent of the

flow.

33

The threshold issue to be decided in this case is whether plaintiffs can establish

that the Corps caused the flooding upon which plaintiffs base their takings claims.

Plaintiffs assert that they may prove causation based on the cumulative and combined

effects of the Corps’ System and River Changes. Specifically, they argue that if they can

prove that “but for” the Corps’ System and River Changes, the flooding in the years

identified would not have occurred or been as severe, they will have met their causation

burden. Plaintiffs argue that they are entitled to prove causation by demonstrating that

the numerous actions the Corps has taken to “retransform” the Missouri River from a

highly engineered river to a more natural state for the benefit of T&E species has led to

increased and more severe flooding. In support of their approach to causation, the

plaintiffs point to the numerous System and River Changes discussed above in the

Background Facts section which were taken to meet the Corps’ ESA obligations under

the 2003 BiOp. The plaintiffs cite to evidence, largely taken from government

documents and testimony, which catalogues the Corps’ River Changes after 2004,

including the reopening of chutes, the creation of ESH, and the construction of SWH.

They also rely on evidence, again primarily from government documents and testimony,

which shows how the Corps has made System Changes to meet its ESA obligations by

releasing water from the dams during periods of high River flows to protect and promote

T&E species.

The plaintiffs contend that their reliance on the cumulative and combined effects

of the Corps’ System and River Changes to establish causation is consistent with the

“single-purpose analysis” that the Federal Circuit approved in Arkansas Game & Fish.

34

Pls.’ Br. at 96 (citing Ark. Game & Fish Comm’n v. United States (“Ark. Game & Fish

III”), 736 F.3d 1364, 1370 (Fed. Cir. 2013) (determining that because the Corps’ yearly

deviations from its water-management plan “were directed to a single purpose”

(benefiting agricultural interests) and “had a consistent overall impact on the

Management Area[,]” it is appropriate to view the flooding as lasting for seven years).

The plaintiffs assert that in Ark. Game & Fish III, the Federal Circuit “recognized that

flooding that is alleged to have occurred because of [the] Corps[’] actions done for a

single purpose are to be analyzed as one continuous flood or on a macro basis, not as

individual floods.” Pls.’ Resp. at 3 (citing Ark. Game & Fish III, 736 F.3d at 1370). The

plaintiffs argue that in cases where “there is a ‘single purpose’ for the Corps’ actions that

are alleged to have caused the flooding, in determining whether there has been

appropriation of property, the Court must assess the collective effects of all the MRRP

flooding resulting from that single purpose[.]” Pls.’ Br. at 95-6 (citing Ark. Game & Fish

III, 736 F.3d at 1370).

The plaintiffs contend that the government’s assertion that to prove causation the

plaintiffs had to isolate each individual Corps action and connect that action to each

separate flooding event on an individual property is not supported. The plaintiffs explain

that this court has long recognized that a taking can be established based on a series of

different government actions taken for a single purpose. The plaintiffs point to Cotton

Land Co. v. United States, where the plaintiffs maintain that the court held that “a

sequence of events acting in concert” and which result in the physical occupation of

property can establish a taking. Pls.’ Resp. at 22 n.10 (relying on Cotton Land, 75 F.

35

Supp. 232, 233 (Ct. Cl. 1948) (finding that a taking had occurred, although the injury

occurred years after the initial act of constructing the dam, when a dam caused sediment

to deposit in the riverbed upstream of the dam, and over time, the sediment raised the

riverbed and water level until the river overflowed its banks and flooded the plaintiff’s

property. The court rejected the government’s defense that the erection of the dam was

too remote a cause on which to base liability because the flooding did not occur directly

from the erection of the dam but through a chain of events). The plaintiffs also rely on

Avery v. United States, 330 F.2d 640, 644-45 (Ct. Cl. 1964) and Cary v. United States,

552 F.3d 1373, 1379 (Fed. Cir. 2009). Finally, plaintiffs argue that if the government’s

theory of causation is adopted and plaintiffs are required to prove how each individual

action of the Corps caused or increased flooding on each individual property for each

separate flooding event, no plaintiff could muster the evidence to prove a case because it

would be impossible to separate out each action and pinpoint the precise cause of

flooding.

The government argues that the plaintiffs’ reliance on the cumulative and

combined effects of the Corps’ System and River Changes to prove causation should be

rejected on the grounds that “[p]roving causation in a takings claim requires a detailed

analysis of the property in question including quantifying the alleged harm directly

attributable to a precise government action.” Def.’s Resp. at 4. The government asserts

that “‘several distinct actions viewed in concert’ are ‘too broad’ to properly allege a

takings claim.” Id. (quoting Acceptance Ins. Cos. Inc. v. United States, 584 F.3d 849,

855 (Fed. Cir. 2009)). According to the government, plaintiffs bear the burden to

36

pinpoint “the precise step in a sequence of events that constitutes conduct the government

cannot engage in without paying compensation.” Id. The government also argues that

the plaintiffs’ reliance on Cotton Land is misplaced, on the grounds that the case involved

only one governmental action, i.e., the impoundment of water behind the dam, whereas

here the Corps has taken numerous actions.

The government further argues that a single-purpose flood theory is inapplicable

here, on the grounds that, unlike Arkansas Game & Fish, “this case does not involve a

single set of discrete water control manual deviations proactively made to serve a single

purpose and affect[ing] a single parcel for a multi-year period, and even [p]laintiffs do

not contend that the Corps’ actions are the sole cause of the various flood events” at

issue. Def.’s Resp. at 7. The government argues that here “the various actions the Corps

has undertaken pursuant to the [new] Master Manual . . . and MRRP have varied each

year and served multiple purposes.” Id. (citing Def.’s Post Trial Brief (“Def.’s Br.”). at

8, 14-6, 35-44). The government also asserts that the plaintiffs allege different amounts

of flooding in different years. Id. at n.9 (citing Def.’s Br. at Tables 3-7, Columns A, I

(flooding at different properties); Pls.’ Br. at Table A (varying amounts of flooding on the

properties)). For all of these reasons, the government argues that plaintiffs are required to

prove how individual actions of the Corps caused or increased flooding at each plaintiffs’

property for each separate flooding event.

The court has considered the parties’ arguments regarding the proper law to apply

and concludes for the reasons that follow that the plaintiffs’ causation theory is supported

and is the appropriate theory to apply in this case. Specifically, the court agrees with

37

plaintiffs that an individual plaintiff can meet the causation burden if that plaintiff can

prove that: (1) the Corps’ System and River Changes were made for a single purpose; (2)

the cumulative and combined effects of the System and River Changes made for that

single purpose led to higher WSEs than would have existed without the System and River

Changes; and (3) the higher WSEs led to flooding, or more severe flooding on the

property owned or farmed by that individual plaintiff than the flooding the plaintiff

would have experienced without the Corps’ System and River Changes.

First, the court finds, contrary to the government’s contentions, that this case is

similar to Arkansas Game and Fish. The Federal Circuit found that “[t]he government

cannot obtain an exemption from takings liability on the ground that the series of interim

deviations were adopted on a year-by-year basis, rather than as part of a single multi-year

plan, when the deviations were designed to serve a single purpose and collectively caused

repeated flooding and timber loss on the Commission’s property.” Ark. Game & Fish III,

736 F.3d at 1370 (emphasis added). Based on the reasoning in Ark. Game & Fish III, the

court finds that it is proper to consider the series of changes made by the Corps for a

single purpose—here, meeting the Corps’ ESA obligations—to determine whether those

System and River Changes have caused flooding that would not have occurred without

those System and River Changes.

Second, the court’s conclusion is consistent with other cases in this court

involving flooding. For example, in Turner v. United States, the United States Claims

Court held that the “plaintiffs have demonstrated that authorized actions of the

Government resulted in the taking of an easement on their property for periodic

38

inundation through flooding.” Turner v. United States, 23 Cl. Ct. 447, 460 (1991). The

court found that “there is a causal connection between the Corps’ activities and the

damage to plaintiffs’ land.” Id. at 455 (emphasis added) (footnote omitted). In that case,

the court concluded that channelization had resulted in higher flow velocity and flood

stages, which had subsequently led to the transfer of greater amounts of sediment that

“raised the river stage along the channel adjacent to plaintiffs’ land.” Id. The court

explained that “[o]ver time, the process had occurred to such an extent that the

tremendous flows captured in the upper reaches of the creek could no longer be contained

in the now shallower lower reaches. This was exacerbated by the damming effect of the

sediment deposits at the East Fork.” Id. (footnote omitted). The court analyzed multiple

Corps actions, and the plaintiffs did not have to pinpoint the specific Corps’ action(s) that

caused the flooding on their property. It was, as these plaintiffs would say, the

“combined and cumulative” impacts of the Corps’ actions over time that constituted a

taking.

Third, the court also finds, as plaintiffs contend, that the government’s reliance on

Acceptance is misplaced. Acceptance is a regulatory takings case and the United States

Supreme Court has made clear that the “longstanding distinction between acquisitions of

property for public use, on the one hand, and regulations prohibiting private uses, on the

other, makes it inappropriate to treat cases involving physical takings as controlling

precedents for the evaluation of a claim that there has been a ‘regulatory taking,’ and vice

versa.” Tahoe-Sierra Pres. Council, Inc. v. Tahoe Reg’l Planning Agency, 535 U.S. 302,

323 (2002) (footnote omitted).

39

Finally, the court also agrees with plaintiffs that if the court were to require each

plaintiff to prove causation with evidence of how each of thousands of Corps’ System

and River Changes individually led to flooding or increased flooding on their property for

each flooding event, no plaintiff would be able to prove causation.

2. Foreseeability

The standards for proving foreseeability in a takings case are well settled. In this

case, to prove foreseeability, the plaintiffs must show either (1) that the Corps intended to

take plaintiffs’ property interests by its actions in making the System and River Changes

it instituted to comply with the ESA or (2) that the invasion of the plaintiffs’ property

interests was the ‘“foreseeable or predictable result”’ of the Corps’ System and River

Changes. Ark. Game & Fish III, 736 F.3d at 1372 (quoting Moden v. United States, 404

F.3d 1335, 1343 (Fed. Cir. 2005) (determining that an injury is “the direct, natural, or

probable result” when the injury was “the foreseeable and predictable result of the

authorized [government act].”)). See also Ridge Line, Inc. v. United States, 346 F.3d

1346, 1355 (Fed. Cir. 2003); Cary v. United States, 552 F.3d 1373, 1377 (Fed. Cir.

2009).

Here, the plaintiffs acknowledge that the Corps did not intend to invade each

plaintiff’s property interest. Rather, the plaintiffs argue that they can establish their

takings claims by proving that the flooding which invaded their properties was the

“direct, natural and probable result” of the Corps’ System and River Changes. Pls.’ Br. at

85 (citing Moden, 404 F.3d at 1343; Cary, 552 F.3d at 1377). The plaintiffs also assert

that in meeting their burden, it is not necessary for plaintiffs to establish that flooding was

40

the direct, natural, and probable result of each individual System and River Change taken

by the Corps. Pls.’ Br. at 87-9. Rather, plaintiffs argue, as they did with regard to

causation, that so long as the Corps’ actions have a ‘“single purpose[,]”’ foreseeability

should be judged taking all of the Corps’ System and River Changes into account. Id. at

87 (quoting Ark. Game & Fish III, 736 F.3d at 1370). In this regard, the plaintiffs

maintain that by endeavoring to restore the River and by making System releases to

benefit T&E species, the Corps knew or should have known that the River’s flooding

pattern would change and that more flooding would occur. Id. at 4, 52-3, 90-1.

The plaintiffs assert that foreseeability, or determining whether flooding was the

direct, natural, and probable result of the Corps’ System and River Changes, is

determined based on an objective, not subjective, test. Id. at 86-7 (citing Ark. Game &

Fish III, 736 F.3d at 1372-3). Plaintiffs point to “extensive evidence” in the form of

government reports, records, and studies, as well as the testimony of government

witnesses before Congress that, at the time the Corps was considering the System and

River Changes it needed to make to comply with the ESA, the Corps knew or should

have known that additional flooding would likely result. Pls.’ Br. at 91; PX16 at PLTF-

00003207-9 (“There will be potential for flood damage on properties that are near the

channel. … There may be drainage problems on some floodplains that have been

converted to agricultural, industrial, or domestic uses.”). See also PX96 at

USACE0070654; PX99 at USACE0291869 (“Based on input from public scoping, the

following issues will be addressed either generally or through a specific evaluation in the

SEIS. . . . Increased flooding on adjacent private lands[;]. … Impacts to levees and flood

41

control structures[.]”); PX324 at USACE0328101 (“Any changes in the management of

the [R]iver will result in a wide variety of impacts to authorized purposes. Meeting the

requirements of the [ESA] will require some sacrifice by other purposes that have

benefited in the past to the detriment of listed species.”).23

The plaintiffs also contend the fact that the Corps did not undertake a

comprehensive study to determine the cumulative and combined impacts of its proposed

System and River Changes on raising WSEs in the River argues in favor of finding

foreseeability under Arkansas Game & Fish III, because foreseeability, according to the

plaintiffs, “encompass[es] not only what the Corps knew at the time it authorized the

MRRP about the flooding impacts of the MRRP, but what it could have known had it

conducted a reasonable investigation concerning those impacts.”24 Pls.’ Br. at 86 (citing

Ark. Game & Fish III, 736 F.3d at 1372-3 (upholding the Federal Court of Claim’s

finding that the flooding was foreseeable because “a reasonable investigation by the

Corps of Engineers prior to implementing the deviations during the 1993–2000 period

would have revealed that the deviations would result in a significant increase in the

number of days of flooding in the Management Area during the growing season.”); Ridge

Line, 346 F.3d at 1357 (citing Cotton Land Co., 75 F. Supp. at 233-4 (mentioning that

23

See PX566; PX911; Tr. 217:25-219:8, 223:5-20, 227:24-228:9, 1114:6-9, 1117:14-1118:6,

1145:4-1149:2, 13140:1-8, 13169:25-1370:11, 13261:2-22, 3042:22-3044:20, 3191:11-3192:10,

5741:19-5742:17.

24

The government acknowledges that “[a]n agency’s reasonable investigation into the results of

a complained-of action before it takes that action can bear on whether a result was foreseeable.”

Def.’s Br. at 79 (citing Ark. Game & Fish III, 736 F.3d at 1373).

42

“[i]f engineers had studied the question in advance they would, we suppose, have

predicted what occurred.”)).

The government agrees that to prove foreseeability plaintiffs do not have to prove

that the government intended to flood their properties but only that that flooding was the

“direct, natural, or probable result of authorized [g]overnment activity.”25 Def.’s Br. at 5.

The government contends, however, that “[t]he United States is only liable for a taking

for damages ‘directly attributable to government action,’ not secondary or contributory

factors that caused damage.” Id. at 77 (quoting Bartz v. United States, 633 F.2d 571, 593

(Ct. Cl. 1980)). According to the government, “[t]his ‘direct attribution’ occurs if the

injury is the ‘direct, natural, or probable result of an authorized activity and not the

incidental or consequential injury inflicted by the action,’ as incidental and consequential

injuries of an action lie in tort.” Id. (quoting Ridge Line, 346 F.3d at 1355). The

government also argued that foreseeability must be determined as of the time of the

governmental action alleged to have taken the property and thus the government argues

that plaintiffs’ failure to pinpoint individual Corps’ actions means that the court does not

have a concrete point in time to determine foreseeability.

The government asserts that if the Corps’ System and River changes have only

increased the “risk” of flooding, the plaintiffs cannot show foreseeability. Def.’s Br. at

78-9. The government argues that plaintiffs must show that the Corps should have

25

The government, as discussed infra, contends, for example, that the 2011 flood was not

foreseeable because it was caused by excessive snowmelt and rainfall in the upper basin above

Gavins Point Dam and that the 2011 flood was not a direct, natural, or probable result of keeping

water in the reservoirs during March and April 2011.

43

known that the System and River Changes would in fact cause additional flooding and

not merely the risk of additional flooding. Relying on Cary, the government argues that

taking a risk or an action that “increases the risk of a detrimental result ‘does not equate

to making the detrimental result direct, natural, or probable.’” Def.’s Br. at 78 (quoting

Cary, 552 F.3d at 1378). The critical question, the government contends, is what set the

chain of events in motion and then whether each event inevitably followed to cause the

result. Id. (citing Cary, 552 F.3d at 1379).

The government contends “[t]he Corps did not intend, nor did it expect, that any

increased flooding would likely occur as the result of either changes to the [Master]

Manual or the construction of habitat projects[.]” Id. at 6. The government describes

how multiple engineers discussed the Corps’ efforts to analyze possible changes from the

Master Manual revision and to model, design, and construct habitat projects to minimize

and avoid adverse effects on flood control.26 Id. at 5-6, 143-5. Relying on its experts and

Corps witnesses, the government contends that the “effects [of the Corps’ actions] on

water levels from project construction are minor and localized.” Id. at 146.

Having considered the parties’ arguments, the court concludes with regard to

plaintiffs’ burden to establish foreseeability as follows. First, plaintiffs must prove that

the flooding at issue was the “direct, natural, or probable result” of the Corps’ authorized

26

The government’s reliance on its efforts to minimize flooding is misplaced. As this court in

Bettini explained, “[t]he inquiry is not whether the result was intended nor whether it occurred

despite the efforts of the Government to prevent the damage, but whether it was the probable or

foreseeable consequence of a deliberate governmental act.” Bettini v. United States, 4 Cl. Ct.

755, 760 (1984) (citing Berenholz v. United States, 1 Cl. Ct. 620, 628 (1982)).

44

actions. Ridge Line, 346 F.3d at 1355 (internal quotations and citations omitted). The

Federal Circuit has explained that to prove a direct, natural, or probable result, a

“property owner must prove that the asserted government invasion of property interests

allegedly effecting a taking ‘was the predictable result of the government action,’. . .

because it was ‘the direct or necessary result’ of the act.” Vaizburd v. United States, 384

F.3d 1278, 1282-3 (Fed. Cir. 2004) (quoting Ridge Line, 346 F.3d at 1356) (other

citations omitted). This court in Baird v. United States explained that “it is the

‘likelihood of the outcome’ of the government’s action that distinguishes its takings from

its torts.” Baird v. United States, 5 Cl. Ct. 324, 330 (1984) (quoting Bettini v. United

States, 4 Cl. Ct. 755, 760 (1984)); Berenholz v. United States, 1 Cl. Ct. 620, 628 (1982)

(“The likelihood of the outcome serves to distinguish conduct which is taking from that

which is tortious.”), aff’d, 723 F.2d 68 (Fed. Cir. 1983)). In other words, “the probability

and foreseeability of the damage is a primary determinative element in whether a taking

or tort occurred.” Baird v. United States, 5 Cl. Ct. 324, 330 (1984).

In this regard, the court finds that Cary is plainly distinguishable from the present

case. Cary involved a forest fire started by a hunter where the government was sued for a

taking because it had left dead wood in the forest which made the forest fire worse than it

would have been without the dead timber. The government has focused on the portions

of Cary finding that there was only a long sequence of decisions, some of which

increased risk, and others which decreased risk, which was insufficient to demonstrate

that the loss of the plaintiffs’ property was the direct, natural, or probable result of the

Forest Service’s timber removal policies. The plaintiffs, on the other hand, rely upon

45

Cary’s recognition that a series of events can give rise to a taking. Cary, 552 F.3d at

1379. In Cary, the Federal Circuit discussed and distinguished flooding cases, including

Cotton Land, holding that “[t]he key difference between the flood cases and the instant

controversy is that the policy of suppressing fires did not set the [fire] in motion as the

dams did the floods.” Id. Thus, in flooding cases, foreseeability has been found where

the Corps removal and subsequent repair of a section of a dike protecting the plaintiffs’

property from Lake Erie caused the dike to weaken and erode, because “the washout of

the dike and flooding of the farm were the natural and probable consequences of

defendant’s conduct in creating the breach.” Berenholz, 1 Cl. Ct. at 627. The court in

Cotton Land held that a taking had occurred because “[t]he events which occurred,

although they took some time, were only the natural consequences of the collision of

sediment-bearing flowing water with still water, and the progress upstream, of the deposit

begun by that collision.” Cotton Land, 75 F. Supp. at 233. The court found that “a

succession of events was initiated which, when the events had all occurred in their natural

order, deprived the company of the beneficial use of its land.” Id.

Similarly, in Barnes v. United States, the Court of Claims determined that the

government’s control of the Missouri River through its construction and operation of the

Fort Randall and Gavins Point Dams on the Missouri River had the “natural

consequence” of producing a sediment delta on the Niobrara River, a tributary, resulting

in the intermittent and inevitably recurring flooding of the plaintiffs’ lands. Barnes v.

United States, 538 F.2d 865, 872 (Ct. Cl. 1976). The court found that “the delta’s

unprecedented size and tenacity were natural consequences of the closing of the dams in

46

the area, and that the flooding [of the plaintiffs’ properties] was a natural consequence of

delta growth.” Id. The court also held that “[t]he flooding has been occasioned by

authorized Government action for public use–construction and operation of the Fort

Randall and Gavins Point Dams.” Id. The court mentioned that “[t]he stipulation shows

that defendant anticipated the creation of a delta and a rise in the groundwater elevations

in the area.” Id. at 873.

The Federal Circuit in Ark. Game & Fish III affirmed the Court of Federal Claim’s

finding that the flooding was foreseeable because “the Corps of Engineers could have

foreseen that the series of deviations approved during the 1990s would lead to

substantially increased flooding of the Management Area and, ultimately, to the loss of

large numbers of trees there.” Ark. Game & Fish III, 736 F.3d at 1372-3. The Federal

Circuit additionally noted that the Court of Federal Claims also “found that during the

deviation period[,] the Commission put the Corps of Engineers on notice of the impact

the deviations were having.” Id. at 1373. Although Commission representatives had

complained that the long flooding periods caused by the Corps’ deviation policies were

damaging timber resources, the Corps had continued to approve said deviations during

the growing season of 2000. Id. In view of the foregoing precedents, the court finds that

the plaintiffs will be able to satisfy their burden if they can show that higher WSEs are a

direct and natural consequence of the cumulative and combined effects of the System and

River Changes taken by the Corps to meet its ESA obligations.

Second, the court also finds that foreseeability is judged on an objective basis. See

Moden, 404 F.3d at 1344 n.3. Thus, the Corps’ subjective foresight of injury is not

47

required. Cotton Land, 75 F. Supp. at 235 (“it is not necessary, in order to take

jurisdiction of a suit for compensation for property taken, to find that the Government’s

agents were aware that their acts would result in its taking, so that their performance of

the acts can be regarded as a somewhat tenuous promise to pay.”).

Finally, as will be discussed in the Liability Findings (Expert Testimony) section,

the issue of causation and foreseeability is different for the flooding in 2011 than the

other years. Plaintiffs are relying on a single purpose theory to prove that the Corps’

System and River Changes together, caused foreseeable flooding in each year. A

distinction must be drawn from the flooding in 2011, which was tied directly to System

releases aimed at protecting the integrity of the dams and reservoirs, and the flooding in

2007, 2008, 2010, 2013, and 2014. The Corps made System releases in 2007, 2008,

2010, 2013, and 2014 for T&E species which, plaintiffs argue, together with the River

Changes had the direct, natural, and probable effect of causing flooding or more severe

flooding in 2007, 2008, 2010, 2013, and 2014. The Corps’ System releases in 2011 were

not part of the single purpose of meeting the Corps’ obligations to protect T&E species.27

As such, the plaintiffs will need to separately prove causation and foreseeability for

flooding in 2011.

3. Severity

27

Plaintiffs argue that the 2011 flood was a consequence of the Corps changing its priorities and

deciding that flood control would no longer be considered the Corps’ first priority. The

plaintiffs, through their expert, Dr. Christensen, conceded that the System releases in 2011 were

not related to the Corps’ ESA obligations under the 2003 BiOp.

48

To prove liability for a taking, plaintiffs must not only establish that the flooding

was caused by authorized government action and was the intended or natural, direct, and

probable consequence of the government’s actions, but also that “the injury constituted a

sufficiently severe invasion that interfered with the landowner’s reasonable expectations

as to the use of the land.” Ark. Game & Fish III, 736 F.3d at 1370 (citation omitted). In

this phase of the litigation the court did not require plaintiffs to establish the full extent of

the injury to their property interests. Issues regarding the full extent of the injury and of

valuing the interest taken have been reserved for the second phase of the litigation.

The plaintiffs argue that for the purpose of proving severity, they need only show

that the injuries to their properties have been more than de minimis and have “interfered

with the landowner’s reasonable expectations as to the use of the land.” Pls.’ Br. at 99

(citing Ark. Game & Fish III, 736 F.3d at 1370). The plaintiffs also argue that in

evaluating the severity of the flooding, this court should again consider “the collective

effects of all the MRRP flooding resulting from [the] single purpose[.]” Id. at 96 (citing

Ark. Game & Fish III, 736 F.3d at 1370).

The plaintiffs maintain that because flooding involves a physical occupation of

land, “[t]he damage can be substantial even if the flooding does not take the whole of the

property.” Id. at 97 (citing United States v. Cress, 243 U.S. 316, 328 (1917) (“[I]t is the

character of the invasion, not the amount of damage resulting from it, so long as the

damage is substantial, that determines the question whether it is a taking.”)). In this

connection, plaintiffs explain that only a few inches of surface flooding or an increase of

several inches in the groundwater table can cause drainage and/or seepage flooding that

49

interferes with planting for an entire year. Id. at 97-8. Plaintiffs assert that, for many

plaintiffs, where there have been repeated years of flooding, the Corps’ System and River

Changes “threaten[] to wipe out an entire way of life.” Id. at 98.

The government argues that plaintiffs can only establish a taking if the flooding

they prove “exceeded ‘a range that the property owner could have reasonably expected to

experience in the natural course of things,’ or the government action ‘impaired the use of

the lands for agricultural purposes.’” Def.’s Resp. at 45 (quoting Ark. Game & Fish III,

736 F.3d at 1374-5 (citation omitted)). The government insists, contrary to plaintiffs’

contentions, that “the question is not simply whether [p]laintiffs experience—for

example—longer, or more severe flooding now compared to before the agency action

that changes the character of the use of the land, but rather, how much—if any—of the

increase is caused by the agency action, and whether that amount, relative to the other

causes, results in a taking.” Def.’s Br. at 80 (citing Alost v. United States, 73 Fed. Cl.

480, 494 (Fed. Cl. 2006)). The government explains that “when natural flooding will last

a large number of days and at depths that would already cause damage—such as

destroying crops—the addition of an incremental length of days of flooding or depth due

to government action is unlikely to ‘materially enhance’ the damage caused by the flood

and result in a taking.” Id. at 81 (quoting Leeth v. United States, 22 Cl. Ct. 467, 487

(1991)). The government contends that in establishing the severity necessary to prove a

taking, courts consider the character of the land, such as the flood-prone nature of the

property, historical flooding events, and historical weather patterns, when assessing

causation and severity. In addition, the government argues based on Keystone

50

Bituminous Coal Ass’n v. DeBenedictis, that the court should also consider the “parcel as

a whole” because plaintiffs “have not alleged, nor can they show, that there has been a

permanent physical occupation of their property or a complete seizure of a portion of

their properties.” Def.’s Br. at 81 (citing 480 U.S. 470, 500 (1987)).

The court finds that at this stage of the litigation, for purposes of establishing

severity, it is sufficient for plaintiffs to show that government-induced flooding has

interfered with plaintiffs’ ability to use their land for its intended purposes. The court

heard uncontroverted testimony from each plaintiff as to how the flooding interfered with

their use of their property and the nature of the damages suffered. In Ark. Game & Fish

III, the Federal Circuit rejected the government’s argument that “the marginal increase in

flooding did not constitute a sufficiently severe invasion of the Commission’s property

rights to support a takings claim.” Ark. Game & Fish III, 736 F.3d at 1374. The court

further explained that “[t]he point is not that there was flooding before the deviations; the

point is that after the deviations began the flooding lasted for significantly longer periods

of time and had much more serious consequences than the flooding of the pre-deviation

period.” Id. The Federal Circuit explained that it is not “unreasonable to measure the

severity of the interference with a property owner’s rights by looking to the effects of the

interference[;] [the] interference with the Commission’s property rights [is considered to

be] as depriving the Commission ‘of the customary use of the Management Area as a

forest and wildlife preserve.’” Id. at 1375 (citing Ark. Game & Fish, 568 U.S. at 37-8).

The Federal Circuit mentioned that “[i]ndeed, it may often be difficult to say, in the

abstract, whether a particular intrusion is severe or only incremental in nature;

51

consideration of the effects of the intrusion on the property owner will often make that

distinction easier to draw.” Id.

In this connection, the court finds that the government’s reliance on Keystone

Bituminous Coal and its related arguments regarding the “parcel as a whole test” are

misplaced and unpersuasive. Keystone Bituminous Coal is a regulatory takings case and

thus is not applicable here. See Tahoe-Sierra Pres. Council, Inc., 535 U.S. at 326-7. Nor

is the “parcel as a whole” test applicable or appropriate for determining the severity of

government-induced flooding in a physical takings case. See id. In Tahoe-Sierra, the

Supreme Court made clear that “[w]hen the government physically takes possession of an

interest in property for some public purpose, it has a categorical duty to compensate the

former owner, regardless of whether the interest that is taken constitutes an entire parcel

or merely a part thereof.” Id. at 322 (citing United States v. Pewee Coal Co., 341 U.S.

114, 115 (1951)).

At this stage of the litigation, plaintiffs who are able to prove an invasion by

government flooding that interfered with that plaintiff’s use and enjoyment of the

property for some period of time (which can be established by combining the impacts of

flood events over multiple years) will be allowed to proceed to the next phase of the

litigation.

III. Liability Findings (Expert Testimony)

The plaintiffs retained three experts to support their takings claims based on

flooding in 2007, 2008, 2010, 2011, 2013, and 2014. In keeping with their theory that

that the flooding was caused by the cumulative and combined effects of the Corps’

52

System and River Changes,28 the plaintiffs focused their experts’ testimony on how the

Corps’ System and River Changes together caused higher water surface elevations

(“WSEs”) in the River, which led to atypical flooding, namely flooding that would not

have occurred or was more severe.29 The plaintiffs retained Dr. Ronald Kurt Christensen,

Ph.D., P.E., J.D., Dr. Ted Hromadka II, Ph.D., Ph.D., Ph.D., D.WRE, P.E., P.G., P.H.,

and Mr. Glenn Tofani, P.E., G.E.

Dr. Christensen’s opinion testimony focused on a “but for” analysis to show

WSEs in a world without the Corps’ System and River Changes, as compared to WSEs in

the “actual” world with the Corps’ System and River Changes for each year of flooding.

Dr. Christensen opined that the difference in WSEs between the “but for” and “actual”

world was explained by the Corps’ System and River Changes, and that together the

Corps’ System and River Changes caused flooding that would not have occurred or

would not have been as severe.

Dr. Hromadka’s opinion testimony focused on the hydrological reasons for how

and why the Corps’ River Changes caused a rise in WSEs, and thus why the Corps’ River

Changes, together with the System Changes, caused plaintiffs to experience more

flooding or more severe flooding than before the System and River Changes were

undertaken by the Corps. He also opined as to why the flooding caused by the Corps’

28

Throughout the trial, the plaintiffs argued that the Corps’ System and River Changes were

made to meet the Corps’ ESA obligations. They also argued that by virtue of the Corps’ ESA

obligations, flood control was no longer the Corps’ first priority and that this shift in priorities

also led to increased flooding. See, e.g., Tr. 4517:23-4525:11 (Christensen).

29

The plaintiffs relied on the testimony of each individual plaintiff to establish the severity of

flooding for the years in question.

53

System and River Changes was foreseeable as a natural, direct, and probable

consequence of the System and River Changes. Finally, he testified regarding the

flooding claimed by each plaintiff and offered his opinion as to whether the Corps’

System and River Changes were a foreseeable cause of the flooding on each plaintiff’s

property. In this connection, Dr. Hromadka explained that there were different reasons

for flooding on various plaintiffs’ properties. For the properties adjacent to the River,

there was overbank flooding, blocked drainage, and seepage. For properties further away

from the River, including properties behind levee systems, there was flooding associated

with levee overtopping and with blocked drainage and seepage. Dr. Hromadka did not

testify as a levee expert, but was qualified to testify regarding blocked drainage and

seepage, which are effects from higher WSEs. In formulating his opinions regarding

flooding from levee failures, Dr. Hromadka relied on the testimony of Mr. Tofani, an

expert in levee construction who was retained by plaintiffs to opine on the effect of

higher WSEs on levees in the vicinity of plaintiffs’ properties, including the effects of

levee overtopping, blocked drainage, and seepage.

Mr. Tofani offered his analyses and opinions regarding the effect of WSEs on

levees and on the failure of various levees during the years in question. There are many

levees protecting properties from flooding along the lower River. As discussed infra,

many of the levees at issue are private levees or levees that are part of the federal

government’s Public Law 84-99 (“PL 84-99”) program that failed and caused flooding in

some of the years at issue. In addition, several properties were flooded when a few

federally owned and operated levees failed in 2011. Mr. Tofani opined as to each

54

plaintiff claiming flooding based on levee overtopping and/or levee failure. He relied, as

did Dr. Hromadka, on the WSE analysis of Dr. Christensen and testified as to how those

higher WSEs translated to higher water levels on affected levees, which caused them to

overtop with some then failing. Mr. Tofani offered his opinions as to whether any of the

levees would have overtopped or failed in the “but for” world modeled by Dr.

Christensen. Mr. Tofani also prepared a separate analysis in connection with one of the

federal levees that failed in 2011 because it was adjacent to a re-opened chute. Finally,

Mr. Tofani offered his opinions regarding the effects of blocked drainage and seepage

that he determined were caused by the Corps’ System and River Changes.

The government, in response to the plaintiffs’ experts, presented the testimony of

experts of its own, as well as the testimony of many professional government employees

from several agencies, including the Corps, who challenged the analyses and opinions of

the plaintiffs’ experts with their own opinions. The government retained Dr. Robert

Mussetter, Ph.D., P.E., Mr. Mark Woodbury, M.S., Dr. Jeffrey Schaefer, Ph.D., P.E.,

P.G., and Dr. Andrew Kopania, D.Env., P.G.

The government presented Dr. Mussetter, who offered his analyses and opinions

to show, based on modeling he had conducted, that the Corps’ changes to the River he

modeled did not cause flooding, but in fact had a positive effect on lowering WSEs in

many situations by allowing more water to spread across the floodplain.

The government also relied on Mr. Woodbury, a civil engineer, who, based on his

modeling and Dr. Mussetter’s calculations, opined that the changes to the System and

River that he modeled, which were not the same as those modeled by plaintiffs’ experts,

55

did not have a significant impact on WSEs in the River and thus did not cause flooding or

make flooding significantly more severe in the years in question. Importantly, the

LIDAR imagery30 presented by Mr. Woodbury, as part of his analysis, confirmed that

virtually all of the flooding complained of by plaintiffs in fact occurred and was almost

always attributable, at least in part, to elevated WSEs in the Missouri River. Based on his

modeling, Mr. Woodbury presented his opinion as to each plaintiff’s flooding claim and

opined as to whether, based on the changes he modeled, the Corps’ actions caused more

or fewer days of flooding than would have occurred without the changes he modeled.

The government also presented the expert testimony of two witnesses to respond

to Mr. Tofani’s opinions regarding levee failure, blocked drainage, and seepage. Dr.

Schaefer was called as the government’s levee expert. He focused his testimony on the

2011 levee failures. Dr. Schaefer opined that the 2011 levee failures were caused by the

extraordinary high flows in the River and that Mr. Tofani’s contention that a re-opened

chute contributed to one of the federal levee failures was not supported. Dr. Schaefer

also examined whether the Corps’ changes, as modeled by Mr. Woodbury, caused

increased seepage and concluded they did not. Relying upon the same modeling, Dr.

Schaefer concluded that none of the Corps’ actions could have led to any erosion on

plaintiffs’ properties either. The government also offered the testimony of Dr. Kopania,

who relying on Mr. Woodbury’s modeling, offered his opinion that he did not find

30

LIDAR is a remote-sensing surveying method produced by the government that creates three-

dimensional pictures of particular points on the Earth’s surface on specific dates.

56

evidence to show that groundwater levels rose due to increases in WSEs, and thus the

Corps was not responsible for causing additional seepage.

The government further relied on the testimony of several professional Corps

employees who work on the Missouri River, who offered their opinions as to whether the

increased WSEs measured on the River could be instead explained by the plaintiffs’

actions. Specifically, these Corps employees testified that by building private levees, the

plaintiffs had constricted river flows and caused higher WSEs. These government

witnesses did not, however, present any analysis or data to support their assertions. The

government also presented the testimony of other government witnesses, most

particularly Jody Farhart, who was responsible for System operations during the relevant

time period and who explained that the Corps did not ignore flood control when it made

releases, most especially in 2011.

Finally, the government presented witnesses from other government agencies,

including FWS, Federal Emergency Management Agency (“FEMA”), and National

Oceanic and Atmospheric Administration (“NOAA”), who testified to various facts

associated with the Corps’ ESA compliance history, weather patterns during the flood

years in question, and where plaintiffs’ properties were located on FEMA maps within

the Missouri River floodplain.

In the sections that follow, the court will review the plaintiffs’ experts’ testimony,

the government’s experts’ testimony, and the government’s witness testimony in detail,

with the understanding, as discussed in the Legal Standards section, that the court agrees

with the plaintiffs that where, as here, the Corps made systemic changes to its

57

management of the System and to the River in order to meet its ESA obligations, it is not

necessary for the plaintiffs to pinpoint the precise Corps actions at each property to

establish “but for” causation and foreseeability. The plaintiffs’ “but for” causation

burden rises and falls on their ability to show for each year of flooding that the

cumulative and combined effects of the Corps’ System and River Changes resulted in

foreseeable flooding that was either greater than what would have occurred without these

Changes or would have been prevented altogether. In this connection, as also discussed

above, the plaintiffs can only meet their foreseeability burden by establishing that the

flooding was either the intended result of the Corps’ actions or the direct, natural, and

probable consequence of the Corps’ actions.

1. Causation and Foreseeability

a. Dr. Christensen

Dr. Ronald Kurt Christensen, Ph.D., P.E., J.D., is a professional engineer with 36

years of experience in civil, reservoir, and river environmental engineering. Tr. 4461:22-

4462:1; PX2050. He holds a B.S. in Watershed Science from Utah State University,

College of Natural Resources (1978). Tr. 4464:5-18; PX2050. He received his M.S. in

Civil Engineering in 1980 from Utah State University, where he studied Water

Hydrology and River and Reservoir Management. Tr. 4463:20-4464:3; PX2050. He

received a Ph.D. in Civil and Environmental Engineering from Utah State University in

1996. Tr. 4463:7-19; PX2050. He also holds a J.D. from the University of Utah College

of Law (1999). Tr. 4462:4-6; PX2050. He is a licensed Professional Engineer in

Alabama, New Mexico, and Utah and is a member of the American Society of Civil

58

Engineers. Tr. 4464:19-24; PX2050. Dr. Christensen was an adjunct professor of Civil

and Environmental Engineering from 1998 to 2002 at Utah State University. Tr.

4465:11-14; PX2050. In 2007, he was an adjunct Hydrology instructor at Utah Valley

State College. Tr. 4465:9-11; PX2050.

Dr. Christensen has had extensive experience studying the hydrology and

operations of rivers, dams, and reservoirs. Tr. 4465:15-22; PX2050. As the owner of

Water and Environmental Services, LLC, a consulting firm, he has worked specifically

on issues of reservoir and river operations, including work studying the Flathead Lake

and Reservoir in Montana between 2015 and 2016, where he did an analysis of flooding,

hydropower, and erosion. Tr. 4464:22-4467:4, 4472:12-16; PX2050. He also studied

flood operations and hydrology for the R.L. Harris Dam on the Tallapoosa River in

Alabama. Tr. 4468:15-23; PX2050. In conducting his work, Dr. Christensen explained

that he has applied various modeling techniques, including the Corps’ HEC-RAS

computer model on several occasions.31 Tr. 4470:4-4472:20, 4482:12-4485:20; PX2050.

He explained the circumstances in which he determined that the Corps’ HEC-RAS model

is helpful and, as discussed infra, why he did not use the HEC-RAS model in his work

31

HEC-RAS is a modeling system that is used by the Corps and others to determine the effects

of individual changes to a river, such as the construction of a bridge across a river which will

likely lead to constricting the river. See, e.g., Tr. 4470:4-4471:2 (Christensen). Dr. Christensen

testified that most recently he had used HEC-RAS to analyze the levee breach on the Missouri

River at the Iatan Power Plant in Missouri. Tr. 4471:15-4472:4. He has also used the computer

program for single-event analysis and floodplain mapping. Tr. 4471:1-2. Dr. Christensen

explained that it is not used to model an entire complex river. Tr. 5026:7-11. See also Tr.

11605:12-15, 11611:13-16 (Woodbury; admitting that one-dimensional models such as HEC-

RAS cannot effectively capture “three-dimensional effects” or “the hydrodynamic complexities”

of a flooding river.); PX2229 at 3 (“there is concern over the ability of 1D [hydraulic] models

[such as HEC-RAS] to effectively capture the hydrodynamic complexities of a river in flood[.]”).

59

for this case. Tr. 4702:10-4712:18. The court finds Dr. Christensen to be a highly

reliable and credible witness with both impressive credentials and extensive relevant

experience on the issues of causation and foreseeability in connection with reservoir

operations and river hydrology.

Dr. Christensen testified that he was engaged by plaintiffs to evaluate, analyze,

and assess the changes the Corps has made to its policies and procedures regarding the

operation and management of the mainstem reservoir System (“System Changes”).32 Tr.

4487:5-25. In this context he explained the new Master Manual “made changes to

navigation releases for conservation” and “made changes to releases for T&E species

protection.” Tr. 4537:10-27, 4539:3-4. He further explained “[t]he BiOp requires two

distinct types of releases to benefit T&E species. First, variable flows to benefit the pallid

sturgeon . . . [a]nd second, releases designed to aid the nesting of and avoid the taking of

least terns and piping plovers.” Tr. 4539:5-7, 21-23. He testified that he was also

engaged by plaintiffs to evaluate, analyze, and assess the changes the Corps has made and

is continuing to make to the Missouri River channel (“River Changes”).33 Tr. 4487:5-25.

32

Dr. Christensen identified six significant System Changes: (1) the Corps “deprioritized flood

control” by balancing all System authorized purposes equally (Tr. 4517:23-4524:11); (2) the

Corps reduced the amount of System flood control storage (Tr. 4525:4-4529:7); (3) the Corps

modified its navigation releases for conservation, resulting in less reservoir storage capacity (Tr.

4529:25-4530:4, 4535:15-4539:2); (4) the Corps replaced Plate 44 in the 1979 Master Manual

with Plate VI-1 in the new Master Manual, which meant that release minimums were replaced by

advisory guidance (Tr. 4529:20-24, 4530:5-4535:7); (5) the Corps increased the frequency of

T&E releases (Tr. 4539:3-4547:16); and (6) the Corps adopted an “adaptive management”

approach to meet its ESA obligations, which gave it more operational flexibility to protect T&E

species (Tr. 4513:13-16, 4514:17-24, 4524:15-4525:3).

33

Dr. Christensen identified four significant River Changes: (1) the Corps modified, removed,

and failed to maintain dikes and revetments (Tr. 4662:16-23, 4667:13-23); (2) the Corps created

chevrons, chutes, and backwater areas (Tr. 4662:16-23, 4667:13-23, 4674:6-4675:3); (3) the

60

He explained that the “2003 amended BiOp . . . ordered that and directed that the Corps

needed to restore the connectivity of the floodplain to the river.” Tr. 4661:7-10.

Specifically, he testified that the 2003 amended BiOp “ordered that shallow-water habitat

be constructed or be created some way or another for [the] endangered pallid sturgeon.”

Tr. 4661:20-22. Additionally, he explained that “[i]n 2004 as required by the ROD, the

Corps deprioritized flood control and accelerated . . . its modification of river control

structures[,] . . . the reintroduction of chutes and, . . . the construction of various habitat

projects[.]” Tr. 4662:18-23. He testified that he was asked by plaintiffs to offer opinions

based on his assessments and analyses as to whether and to what extent the Corps’

System and River Changes have contributed to the increased incidents of flooding and to

the severity, frequency, and duration of flooding on the Missouri River since those

System and River Changes have been implemented by the Corps. Tr. 4487:15-18.

Although, as discussed below, Dr. Christensen analyzed each of the flood years

separately, Dr. Christensen offered the following generalized opinions. First, he opined

that “[t]he Corps’ changes to its operation and management of the [S]ystem post-2004

contributed to cause higher water surface elevations downstream of the [S]ystem and

flooding of greater frequency, severity and/or duration than would have otherwise

occurred.” Tr. 4488:6-11. Second, he opined that “[t]he Corps’ changes to its operation

and management of the [R]iver channel had the cumulative effect of reducing the flood

Corps constructed SWH and ESH projects for T&E species (Tr. 4662:16-23, 4663:23-4665:14,

4676:12-4677:15); and (4) the Corps added sediment to the River (Tr. 4673:22-4674:5, 4680:11-

16, 4691:18-19).

61

carrying capacity of the [R]iver, allowing the lateral migration of water to the floodplain

during high-water events, sloughing and eroding the riverbanks, and contributing to cause

high water surface elevations and flooding . . . of greater frequency, severity, and/or

duration than would have otherwise occurred.” Tr. 4488:12-21. Third, Dr. Christensen

opined that “[b]ut for the Corps’ changes to its policies and procedures for the operation

and management of the [S]ystem and . . . the [R]iver, the flooding of [p]laintiffs’

properties either would not have occurred or would have been of lesser frequency,

severity and duration.” Tr. 4488:22-4489:2. Finally, he opined that “[u]nless the Corps

reverts to its prior policies and procedures for the operation and management of the

[S]ystem and the [R]iver, the increased levels of flooding will continue.” Tr. 4489:3-6.

Dr. Christensen distinguished the flooding in 2011 from the other years of

flooding at issue in the case. Specifically, he acknowledged that not all of the flooding

experienced by all plaintiffs “could have or would have been avoided[]” in 2011. Tr.

4489:18-4490:1. In contrast to the other years of flooding, Dr. Christensen also opined

that the System releases in 2011 were made solely because of inadequate System storage

in the reservoirs due to high mountain snowpack and heavy plain snowpack. Tr.

4590:11-23, 4591:21-4592:3. He explained that the “[s]ummary of results are that the

lack of system storage forced the Corps to make releases of unprecedented duration and

volume throughout the summer of 2011, contributing to and exacerbating catastrophic

flooding throughout the basin.” 4591:24-4592:3. According to Dr. Christensen, the

extensive flooding in 2011 was caused by the Corps’ failure to release water from the

62

reservoirs early enough in the year and in sufficient quantities to minimize flooding. Tr.

4591:24-4594:8.

In contrast to the System releases in 2011, Dr. Christensen explained that the

System releases in 2007, 2008, 2010, 2013, and 2014 were made to meet the Corps’

obligations under the ESA. Tr. 4539:3-4540:2, 4546:6-4547:16. Dr. Christensen

testified that releases from the System for years other than 2011 were sometimes needed

to prevent protected bird species from nesting in low-lying areas.34 Tr. 4539:21-4541:14.

Other times, releases were needed to cue the pallid sturgeon to spawn. Tr. 4539:8-20.

Dr. Christensen acknowledged that some flooding would still have occurred at points

downstream of Gavins Point during 2010, but unlike in 2011, he testified that the Corps

in 2010 made T&E releases that made flooding more severe in extent and duration. Tr.

4581:22-4582:18; PX2094-A at 1-3.

The court finds that the releases and the reasons for the releases in 2011 and in

2007, 2008, 2010, 2013, and 2014 are sufficiently different from each other that the 2011

flood needs to be analyzed separately for both causation and foreseeability. Thus, the

court will first review Dr. Christensen’s analysis and opinions regarding the System

Changes for the 2007, 2008, 2010, 2013, and 2014 flooding, and then the court will turn

to Dr. Christensen’s analysis and opinions regarding the System Changes in 2011. The

court will thereafter examine Dr. Christensen’s opinions regarding causation and

34

Under the terms of the 2003 BiOp, the Corps now operates under an incidental take statement

which allows the Corps to kill a limited number of birds or eggs without triggering the liability

provisions of the ESA.

63

foreseeability with respect to the River Changes and his analysis of the cumulative and

combined effects of the Corps’ System and River Changes.

i) System-Related Flooding in 2007, 2008, 2010, 2013, and

2014

Dr. Christensen began his opinion testimony by briefly reviewing the history of

the Missouri River and how it has changed over time. Dr. Christensen explained that

historically the River was “a braided, multi-channel river . . . [which] flooded

regularly[.]” Tr. 4491:14-16. He explained that the River’s original connection to its

floodplain had allowed native species to flourish. Tr. 4491:16-19. He then explained, as

discussed above, that in the 1940s in order to encourage economic development in the

Missouri River Basin, “the Government began, through the Corps, to actively regulate

and manage all aspects of the [R]iver to provide, amongst other things, flood control and

downstream navigation.” Tr. 4491:23-4492:4. He testified that under the 1944 Flood

Control Act (“FCA”), the six large dams and reservoirs, Fort Peck, Garrison, Oahe, Big

Bend, Fort Randall, and Gavins Point, were built by the Corps to provide flood control.

Tr. 4494:13-19. He explained that each reservoir in the System has five storage zones:

(1) the inactive zone, which is the lowest zone where water is stored continuously; (2) the

carryover storage zone, where water can be stored over multiple years and which is

emptied and filled for System water uses, including, in order of priority under the 1944

FCA vc(flood control; navigation; hydropower; irrigation, municipal and industrial water

supply, and water quality; recreation; and fish and wildlife habitat); (3) the flood control

and multiple use zone, which needs to be emptied by March each year, so there is enough

64

room in the System to capture the snowmelt and hold it to protect against downstream

flooding; (4) the exclusive flood-control zone, which is for flood control only and is kept

empty except for when there are large floods; and (5) the surcharge zone, which is the

uppermost zone and is empty to accommodate large floods that exceed the exclusive

flood control zone. Tr. 4495:24-4499:20.

Dr. Christensen testified that flood control is allocated among the reservoirs. Tr.

4499:22-4500:14. See also PX2200. He explained that the upper Basin runoff captured

by the System includes: (1) plains snowmelt, which usually occurs during March and

April; (2) mountain snowmelt, which usually begins in late April and can go through

July; and (3) rainfall, which is random but occurs during these melting periods. Tr.

4501:3-4502:17.

Dr. Christensen described his understanding of the Corps’ System operations

under the 1979 Master Manual and later contrasted that understanding with his

understanding of the changes made by the Corps to the Master Manual after 2004, which

included changes to meet the Corps’ obligations under the ESA to comply with the 2003

BiOp. Tr. 4502:23-4547:16. Under the 1979 Master Manual, Dr. Christensen testified,

flood control was the highest priority and fish and wildlife protection had the lowest

priority. Tr. 4503:12-23. See also PX3 at USACE0004040-1. Dr. Christensen explained

that under the 1979 Master Manual, releases from Gavins Point Dam were significantly

reduced during periods of high downstream inflows from the tributaries into the Missouri

River. Tr. 4503:24-4505:7.

65

Dr. Christensen testified that in order to implement the BiOp, the Corps changed

the Master Manual in 2004 and further revised it in 2006, hereafter the “new Master

Manual.”35 Tr. 4513:4-12. According to Dr. Christensen, under the new Master Manual,

flood control is no longer the Corps’ first priority except when risks to health and safety

are imminent. Tr. 4517:23-4524:11. See also PX4 at USACE0002644; PX5 at

USACE0121577-8, USACE0121590; PX2201. Whereas the 1979 Master Manual clearly

identified flood control as the first priority, the new Master Manual states that “Congress

did not assign a priority” to any of the 1944 FCA-authorized purposes and “it was

contemplated that the Corps . . . would balance these functions in order to obtain the

optimum development and utilization of the water resources of the Missouri River

[B]asin to best serve the needs of the people.” PX5 at USACE0121590. The new Master

Manual further states that the Corps would now abide by four objectives: “first, to serve

the contemporary needs of the [B]asin and the Nation; second, to serve the

Congressionally authorized project purposes; third to comply with other applicable

statutory and regulatory requirements including environmental laws such as the

Endangered Species Act (ESA); and fourth, to fulfill the Corps’ responsibilities to

Federally recognized Tribes.” Id. See also PX2201.

35

Dr. Christensen explained that in addition to operating reservoirs and dams, the Corps’ pre-

2004 actions over the years had served to disconnect the Missouri River from its floodplain by

channelizing the previously shallow and braided river. Tr. 4492:5-14. In so doing, he testified,

“[t]housands of acres of habitat of native fish and wildlife were destroyed, including the habitat

of what are now considered threatened and endangered species . . . under the ESA[.]” Tr.

4492:15-19. As discussed previously, concerns over these environmental consequences led to

litigation forcing the Corps to make changes to its policies and practices to better protect and

promote T&E species. Ultimately, this litigation resulted in a court order requiring the Corps to

implement the 2003 BiOp of the FWS.

66

Dr. Christensen explained that to comply with the 2003 BiOp, the Corps began to

make very specific releases downstream to benefit T&E species (“T&E releases”).36 Tr.

4539:3-4540:2. First, to benefit the interior least tern and piping plover, he testified that

the Corps began making releases using cycled or heavy steady spring flows to prevent the

least terns and plovers from building low-lying nests for fear these nests would wash

away, and then the Corps would lower the releases downstream to prevent flooding once

the nests were made. Tr. 4539:21-4541:4. Second, to benefit the pallid sturgeon, Dr.

Christensen explained that the Corps began making additional releases downstream. Tr.

4539:8-20. These releases involved increasing flows in the spring in order to provide

spawning cues to the sturgeon. Id. The court understands that the Corps also lowered

releases in the summer in order to protect the SWH the Corps had created for the

sturgeon.37

Dr. Christensen opined that the T&E releases needed to meet the Corps’ ESA

obligations compromised flood control because cycled releases were made by the Corps

at the same time that the lower River was facing large tributary inflows or rain. Tr.

4541:8-14. According to Dr. Christensen, the Corps’ releases made to comply with the

ESA added to downstream flooding in 2007, 2008, 2010, 2013, and 2014.38 Tr. 4546:22-

36

Dr. Christensen acknowledged that the Corps had periodically made releases for species

protection starting in 1986, but explained that under the new Master Manual, these T&E releases

are now mandatory to comply with the ESA. Tr. 4541:5-14.

37

Dr. Christensen explained that reductions in releases for T&E protection can interfere with

flood protection whenever higher releases are needed to empty the reservoirs to provide space for

upper Basin runoff the next year. Tr. 4541:8-14, 4546:6-21.

38

The benefits to the least terns and plovers, as well as to the pallid sturgeon, from flooding were

acknowledged by both Corps and FWS witnesses at trial. Mr. Casey Kruse, the Missouri River

coordinator for the FWS, testified that big floods benefit the bird species by improving habitat.

67

4547:16. He reached that conclusion based upon a model he created to compare River

water levels in a “but for” world where the Corps did not have to make releases for ESA

compliance with those in the “actual world” with the T&E releases needed for ESA

compliance. Tr. 4547:24-4549:11.

Dr. Christensen characterized his model as “simple, because it [was] based on

measured data, actual Corps past practices, and the applicable criteria, rather than

elaborate computer simulation.” Tr. 4549:12-15. Dr. Christensen treated the six

reservoirs of the System as a single reservoir for modeling purposes, and thus he only

modeled releases from Gavins Point, the last reservoir in the System. Tr. 4552:22-

4553:2. He claimed that he accounted for differences in available storage in individual

reservoirs in his calculations of a single release from Gavins Point Dam, stating that at

“the end of the month [S]ystem storage [for all reservoirs was] equal to the previous

month storage, plus net inflow minus net outflow[.]” Tr. 4553:3-6. Regarding

assumptions, Dr. Christensen explained that in his “but for” world, flood control

remained the number one priority and that the Corps would have continued to make

releases that benefited flood control. Tr. 4551:7-20. He further explained that he

modeled the years 1993 through 1999, as well as the period from 2000 to 2015, in order

to capture two high flow years before the Corps implemented the new Master Manual.

Tr. 4553:8-15. He used 1993 through 1999 to model actual System releases and the

Tr. 1155:7-12. Mr. Bitner, from the Corps, testified that overbank flooding also helped the pallid

sturgeon by washing biota and nutrients into the River. Tr. 1740:14-23.

68

period from 2000 to 2015 to model old policy operations under the 1979 Master Manual.

Id. He then explained his modeling results for each year of flooding.

For 2007, Dr. Christensen testified that “the Corps’ actual releases in 2007 were

higher than they would have been under the [1979 Master Manual] and policies. The

primary reason for the difference was the changes the Corps made to its release schedule

for T&E low nesting prevention and conservation.” Tr. 4557:13-23. Specifically, he

explained that the Corps in 2007 made higher T&E releases in the spring despite high

downstream inflows and that because these higher releases coincided with the highest

downstream flows, there was increased flooding. Tr. 4556:22-4558:1. Under Dr.

Christensen’s “but for” model, the Corps would have limited its releases to 6,000 cubic

feet per second (“cfs”) during critical periods, instead of making T&E releases. Tr.

4559:5-9, 4567:5-9; PX2060-A at 2; PX2091-A at 1-2. Dr. Christensen acknowledged

that the additional releases for T&E species in the actual world in 2007, as compared to

overall flows downstream, were not necessarily large. Tr. 4558:12-4559:23. He opined,

however, that because they added flow, these T&E releases contributed to higher WSEs,

thus causing greater flooding in 2007 than would have occurred without these T&E

releases. Tr. 4557:16-4567:25. For example, he explained that on May 2nd, the Corps

began higher T&E releases, which, due to travel time, coincided with the May 7th and

8th critical high flows. Tr. 4573:9-13. Evidence presented by the government showed

that, during periods of peak flow, releases from Gavins Point made up approximately five

to ten percent of downstream flows in 2007. DX3015-207; DX3015-220; DX3015-221;

DX3015-300; DX3015-301; DX3015-394; DX3015-395; DX3015-552; DX3015-553.

69

Dr. Christensen presented his model results for 2008 and offered the same opinion

as he had for 2007, to the effect that releases in 2008 were higher to prevent low T&E

bird nesting, despite higher downstream inflows, and that because they coincided with

periods of higher downstream flows, these T&E releases caused more severe flooding

than would have occurred without these releases. Tr. 4573:23-4575:9. Dr. Christensen

explained that in the “but for” world, the Corps would have reduced releases in 2008

when the Corps had some reason to anticipate higher tributary flows. Tr. 4574:3-8,

4576:3-7; PX2060-A at 3; PX2091-A at 3-4. According to Dr. Christensen, the

differences between his models under the “but for” world and the actual world for May

and June were greater for 2008 than for 2007. Tr. 4576:21-24. Evidence presented by

the government showed that during periods of peak flow, releases from Gavins Point

made up approximately five to fifteen percent of the downstream flow in 2008. DX3015-

207; DX3015-220; DX3015-222; DX3015-300; DX3015-302; DX3015-394; DX3015-

397; DX3015-552; DX3015-554.

Dr. Christensen explained that System storage was a factor in the flooding that

occurred in 2010 because “reduced storage capacity . . . led to higher releases” from the

System. Tr. 4578:13-19. According to Dr. Christenson, “[c]arryover storage [in 2010]

was more full than it would have been pre-2004 due to the Corps’ new policies and

practices of reduced releases for conservation and reduced capacity due to

sedimentation.” Tr. 4578:23-4579:2. Specifically, Dr. Christensen testified that, due to

high plains snowpack melt and precipitation, the three largest downstream reservoirs

were filled into the exclusive flood control zone early: Oahe by the end of May; Fort

70

Randall by the end of June; and Garrison by the end of July. Tr. 4579: 3-6, 4580:5-13.

During June, July, and most of August, considerable downstream flooding occurred,

which, Dr. Christensen opined, was exacerbated by the high releases from the System,

including T&E releases that “coincided with the highest downstream peak flows[.]” Tr.

4582:4-25. Dr. Christensen conceded that this was one of the years of flooding in which

all flooding could not have been avoided in the “but for” world. PX2094-A at 1-3. Dr.

Christensen explained that under the 1979 Master Manual, the Corps would have had

more capacity in the reservoirs because it would not have retained as much water in the

reservoirs from the prior drought year, and when there was downstream flooding, the

Corps would not have conducted T&E releases. Tr. 4578:24-4579:2, 4583:2-5. Dr.

Christensen testified that in the “but for” world, the Corps would have kept releases

lower during June and July when the highest downstream flows occurred, and the lower

releases would therefore have reduced the flooding in 2010. Tr. 4588:21-4589:5;

PX2060-A at 4; PX2094-A at 1-3. Evidence presented by the government showed that

during periods of peak flows in 2010, releases from Gavins Point made up fifteen to

twenty percent of the flow. DX3015-207; DX3015-220; DX3015-223; DX3015-300;

DX3015-303; DX3015-394; DX3015-398; DX3015-552; DX3015-555.

Dr. Christensen testified that T&E releases in 2013 caused increased flooding. Tr.

4647:10-17. He explained that despite higher downstream inflows during late May and

early June, the Corps maintained higher T&E releases, which “coincided with the highest

downstream flows and increased the flooding.” Tr. 4647:4-17. Dr. Christensen testified

that in the “but for” world, the Corps would have reduced releases when there were high

71

downstream flows. Tr. 4647:6-12, 4648:18-23; PX2091-A at 5-6. Evidence from the

government showed that during periods of peak flows, releases from Gavins Point made

up twenty to thirty percent of the flow in 2013. DX3015-207; DX3015-220; DX3015-

227; DX3015-300; DX3015-308; DX3015-394; DX3015-406; DX3015-552; DX3015-

561.

Dr. Christensen testified that T&E releases increased flooding in 2014. Tr.

4650:4-4651:1. He testified that while the Corps reduced releases in June 2014 to

accommodate downstream flooding, the Corps maintained higher flows for T&E species

purposes, which “coincided with the highest downstream flows and increased the

flooding.” Tr. 4650:4-19. He explained that the Corps’ actions thus “caused higher and

longer duration flood flows.” Tr. 4650:4-7. Dr. Christensen testified that in the “but for”

world, the Corps would not have made T&E releases and would have reduced its

releases, reducing the flooding in 2014. Tr. 4650:4-12; PX2091-A at 7-8. Evidence from

the government showed that during periods of peak flows, releases from Gavins Point

made up ten to twenty percent of the flow in 2014. DX3015-207; DX3015-220;

DX3015-229; DX3015-300; DX3015-310; DX3015-394; DX3015-408; DX3015-552;

DX3015-563.

The government took issue with Dr. Christensen’s System model and his opinions

regarding the Corps’ System Changes for the years in question. The court has considered

those criticisms and has determined that the government’s criticisms do not undermine

the basic logic and factual support behind Dr. Christensen’s opinions. Set forth below are

some of the most repeated criticisms for each flood year and the reasons the court finds

72

that these criticisms do not undercut Dr. Christensen’s conclusions. First, the

government criticized Dr. Christensen by suggesting that his model required omniscience

because it assumed extremely low releases to avoid the highest downstream flows, which

the government contends could not be known.39 See Tr. 4572:20-4573:22, 4577:15-

4578:11, 4589:6-4590:10, 4648:24-4649:21, 4651:2-22 (Christensen). The court

disagrees. The 1979 Master Manual did not require releases for the benefit of fish and

wildlife and, based on the priorities articulated in that Master Manual, no releases for fish

and wildlife would have been undertaken unless the Corps was assured that there would

be no flooding in the lower Basin. PX3 at USACE0004040-1. See also Tr. 4503:12-23,

4530:19-22, 4586:12-21 (Christensen), 6611:2-23 (Ponganis). For example, in 2007, just

before it increased T&E releases, the Corps had reduced its releases for a brief period

because of flooding concerns but then raised T&E releases right at the time of greater

downstream inflows. Tr. 4573:1-22. See also DX3001-243. In the “but for” world, Dr.

Christensen correctly assumes that if flooding was already occurring in the lower River,

the Corps would not have resumed releases. The same was true for T&E releases in

2008. See, e.g., Tr. 4577:19-4578:11; DX3001-247. Similarly, in 2010, the Corps

continued T&E releases after flooding downstream was brought to the Corps’ attention.

Tr. 4581:22-4584:8. See also DX3001-263. After receiving notice that 14 or 18

Missouri River gages were above the flood stage downstream of Gavins Point, the Corps

continued to make T&E releases in 2010 because, as Ms. Farhart explained at the time in

39

The criticism was made by Mr. Woodbury. Mr. Woodbury’s background is described at

length infra in connection with Dr. Hromadka’s testimony.

73

an e-mail to a colleague, “[t]his is the peak of the tern and plover nesting season and if

we cut now it may be difficult to go back up if the birds come in and nest low.” PX967 at

USACE5569989. On June 11, 2010, Ms. Farhart explained to her fellow Corps

colleague, Colonel Roger Wilson, that she would continue to cycle releases for the least

terns and plovers, stating,

[a]s you may know, we have been cycling releases from 26,500 cfs to

28,000 cfs for 8 hours every other day to prevent terns and plovers from

nesting on low elevation sandbar habitat. As a result of the flooding

downstream, Gavins Point releases on the low part of the cycle will be

reduced from 26,500 to 22,000 cfs while continuing the 8 hours of 28,000

cfs every other day. This is the peak of the tern and plover nesting season,

so maintain[ing] the cycle is critical to prevent large numbers of nest[s]

from being established on newly exposed habitat. Once the bulk of the

nests are established we will have more flexibility to reduce or eliminate

the cycling.” Id. at USACE5569988.

The Corps continued T&E releases in 2013 and 2014 in the face of increased

downstream flows and higher WSEs. Tr. 4647:10-17, 4649:14-18 (2013),

4650:10-19, 4651:22-4652:1 (2014). In fact, in 2015 the Corps continued to make

T&E releases after learning that the lower 500 miles of the Missouri River from

Nebraska City to St. Louis were three to six feet above flood stage. PX2308 at

USACE0356497. At that time, the Corps acknowledged that reduction in releases

74

would “reduce stages downstream about 1 to 1.5 feet[.]”40 Id. In view of the

foregoing facts, the court finds that the government’s criticism of Dr.

Christensen’s testimony based on Mr. Woodbury’s claim that Dr. Christensen’s

modeling requires “omniscience” is not persuasive.

Second, the government criticized Dr. Christensen’s model because he did not

present modeling or other evidence to show with specificity how he accounted for

releases between the individual reservoirs in the System, but instead presented modeled

releases from only Gavins Point. Tr. 11355:25-11356:9, 11367:25-11371:20

(Woodbury). The court acknowledges that Dr. Christensen presented modeling only for

releases from Gavins Point. However, for flooding in 2007, 2008, 2010, 2013, and 2014,

the court finds that the criticism is not well founded. For purposes of modeling

downstream flood control, the court is persuaded that releases from Gavins Point were

the only relevant releases to consider. The government did not present any evidence to

show how releases among the upper reservoirs would have been relevant in modeling a

“but for” world, where the flooding for the years in question was all below Gavins

Point.41

40

The court recognizes that the 2015 flooding is not at issue in this phase of the litigation, but

finds that this evidence confirms the plaintiffs’ position that in the actual world ESA compliance

has taken precedence over flood reduction and that in the “but for” world the additional releases

mandated by the ESA would not been made and thus would not have contributed to River flows

and flooding.

41

Dr. Christensen noted that the flooding in 2010 involved some System storage concerns, but

explained that the main contributions to additional flooding were the T&E releases made in the

face of downstream flooding. Tr. 4579:25-4582:21.

75

Third, the government also took issue with Dr. Christensen’s use of 6,000 cfs as

the minimum daily release in the “but for” world on the grounds that in more recent

years, the Corps has had to make higher releases to meet water supply needs below

Gavins Point. Tr. 11366:5-11367:23, 11698:1-11700:16 (Woodbury); DX3015-648. The

court again finds that the criticism is not well founded. To begin, the 1979 Master

Manual explicitly uses 6,000 cfs as the minimum release in order to meet water supply

requirements and Dr. Christensen was modeling the “but for” world using the 1979

Master Manual. Tr. 4570:20-4572:19; PX3 at USACE0004043, USACE0004057-8.

Moreover, to the extent that the minimum release number should have been 8,000 cfs, as

suggested by the government’s evidence, the court accepts the government’s concession

from Mr. Woodbury that an increase of 2,000 cfs would not “drastically change flooding

levels[,]” meaning that the minimum amount under either “but for” scenario would likely

be the same. Tr. 11702:12-19 (Woodbury).

Fourth, the government criticized Dr. Christensen for failing to take into account

the Corps’ actions in reducing its releases at various times when it became aware of flood

risks downstream. The court does not find this criticism persuasive. The fact that the

Corps knew its actions increased flooding during certain periods and changed its releases

for a limited period of time only confirms that the Corps made T&E releases during other

downstream high flow periods when it was on notice that additional releases would result

in higher WSEs and more flooding.42

42

As discussed later in the context of Dr. Hromadka’s testimony, the government’s evidence of

significant rainfall during periods of flooding does not undermine the credibility or reliability of

76

Finally, the court did not find any of the government’s other criticisms of Dr.

Christensen’s opinions regarding 2007, 2008, 2010, 2013, and 2014 persuasive. As

discussed below in detail in connection with Dr. Hromadka’s expert testimony, the court

does not find the opinions offered by Dr. Mussetter or Mr. Woodbury sufficiently reliable

to cause the court to question Dr. Christensen’s opinions that the T&E releases mandated

by the 2003 BiOp and reflected in the Corps’ new Master Manual would not have been

made in the “but for” world during periods where downstream flooding was evident or

highly likely.

The court finds that Dr. Christensen’s testimony is sufficient to show that the

Corps’ System Changes, which for the years in questions include T&E releases, caused

increased WSEs. The court also finds that Dr. Christensen’s testimony is sufficient to

show that it was foreseeable that by adding flow to the River, the T&E releases would

increase flooding in periods of already high River flows below Gavins Point Dam. The

increased flooding was a direct, natural, and probable consequence of making T&E

releases at times when there was known or expected increased downstream inflows or

flooding.

ii) System Changes and the 2011 Flood

There is no dispute that the 2011 flood was a record-setting flood that involved a

record 60.8 million acre-feet (“MAF”) runoff, the highest since record-keeping began in

Dr. Christensen’s testimony. Flooding usually occurs because of increased precipitation. The

plaintiffs contend that it was the Corps’ T&E releases in the face of foreseeable flooding,

combined with the River Changes made by the Corps to comply with its ESA obligations, that

satisfy plaintiffs’ causation and foreseeability burden in this case. The court agrees.

77

1898. DX0192 at DX0192-0007, DX0192-0016. It is also not disputed that starting in

June 2011, the Corps released a record 160,000 cfs from Gavins Point for many months

to address the record runoff. Id. at DX0192-0026. Prior to that time, 70,100 cfs had been

the record released from Gavins Point in 1997.43 Id.

To understand Dr. Christensen’s “but for” analysis for System Changes with

regard to 2011, the court will first review his testimony as to the differences between the

1979 Master Manual and the new Master Manual. According to Dr. Christensen, the

entire mainstem “[S]ystem is designed to control inflows up to 40 million acre-feet,

March to July, without outflows ever exceeding 100,000 cfs[.]” Tr. 4600:15-19. Dr.

Christensen opined that had the Corps followed the release levels set in the 1979 Master

Manual, and specifically the levels set in Plate 44 in that Manual, flooding in 2011 would

have been less severe. Tr. 4598:25-4599:15. Dr. Christensen opined that had the Corps

followed Plate 44 and the 1979 Master Manual, it would have released more water from

the reservoirs earlier, in late March or early April, and that had the Corps done so, the

Corps would have limited releases from Gavins Point downstream to 100,000 cfs and

thus limited the severity of flooding. Tr. 4592:8-4594:8; PX2060-A at 5, PX2094-A at 4-

16. He recognized in his opinion testimony that there still would have been significant

flooding at 100,000 cfs, but he explained that flood damage would have been greatly

reduced. Tr. 4600:15-4601:12. He explained that “the [2011] flood water levels had

43

The court understands that the 2011 flood was devastating for the plaintiffs. However, as

discussed infra, in order to establish liability for a taking, the plaintiffs were required to establish

causation and foreseeability with regard to the 2011 flood, which the plaintiffs failed to do.

78

increased by as much as five to six feet in the worst areas and at the very least more than

1.5 feet in all [p]laintiff areas[.]” Tr. 4794:11-15.

Dr. Christensen opined that the increased flooding in 2011 was also foreseeable or

a natural, direct, and probable consequence of the Corps having to use Plate VI-1 in the

Corps’ new Master Manual, rather than Plate 44 from the 1979 Master Manual. Tr.

4594:20-4600:11. Dr. Christensen based his opinion on a comparison of the Corps’

response to upper Basin runoff in 1997 and 2011. Tr. 4597:9-4598:4. Dr. Christensen

testified that upper Basin runoff in 1997 and 2011 were similar, but that the record runoff

in the upper Basin in 1997 did not lead to downstream flooding because of the proactive

measures the Corps took to minimize the flooding risk mandated by Plate 44. Tr. 4597:9-

4599:15, 4607:3-4612:3. Dr. Christensen produced a chart which contrasts 1997 releases

to 2011 releases to show that before the new Master Manual revisions, the Corps made

larger releases in 1997 than even the minimum amounts required by the 1979 Master

Manual to avoid flooding. Tr. 4607:3-4612:3; PX2203. He testified that if the Master

Manual had not been changed, Corps personnel would have been focused on flood

control and would have acted in 2011 as they had in 1997. Tr. 4598:1-16, 4624:8-12. He

explained that in the actual world, under the new policy of “balanced” priorities, the

Corps was not allowed to make earlier releases for flood control because flood control

only becomes the highest priority under the new Master Manual when flooding is deemed

imminent. Tr. 4595:5-8, 4598:12-4599:3, 4610:2-4612:3, 4625:17-25; PX5 at

USACE0121577 (“Flood Control carries the highest priority during significant runoff

events that pose a threat to human health and safety[.]”), USACE0121625 (“The flood

79

control purpose of the System will be given the highest System priority during periods of

significant runoff when loss of life and property could occur.”). Dr. Christensen noted

that e-mails show that it was not until April 25 that the Corps decided that upper Basin

runoff could no longer be contained by the reservoirs and only then did the Corps begin

making releases from Gavins Point. Tr. 4618:11-23; PX2206. He also noted that the

Corps was still consulting with FWS regarding T&E releases in April and did not begin

to evacuate the reservoirs until May 7. Tr. 4618:11-4619:5; PX787; PX2206.

Dr. Christensen testified that while Corps personnel consistently explained that

among the reasons for not releasing water from the reservoirs earlier was downstream

flooding concerns, the Corps’ explanation for not making earlier releases in late March or

early April based on concerns over downstream flooding was not consistent with the

Corps’ approach in 1997. Tr. 4622:16-4624:12. In 1997, he testified, the Corps released

water from the reservoirs “despite minor downstream flooding.” Id.

The court has considered Dr. Christensen’s testimony and concludes that

plaintiffs’ causation theory based on a “but for” world in 2011 fails for two primary

reasons. First, and foremost, the plaintiffs have not established how the Corps’ System

releases in 2011 can be considered part of the “single purpose” it has relied upon to

establish causation for the other flood years. The Corps’ System releases in 2011 had

nothing to do with ESA compliance. The court’s conclusion is confirmed by the

independent study commissioned by the Corps to evaluate the 2011 flood. In 2011, the

Corps commissioned four independent non-Corps experts to evaluate the Corps’

80

operation prior, during, and after the 2011 flood.44 The panel provided its findings and

recommendations in its report entitled Review of the Regulation of the Missouri River

Mainstem Reservoir System During the Flood of 2011. See DX0192 at DX0192-0094-9.

In the report, the experts concluded that the Corps’ “decisions related to storage and

evacuation of spring runoff were not influenced or affected by consideration of

threatened or endangered species.” DX0192 at DX0192-0074.45 In addition, in his

separate report entitled 2011 Missouri River Flood Report: Opinions, Bases, Reasons,

Facts and Data, Dr. Grigg explained that in 2011 “operations for purposes other than

flood control (including environmental purposes) were suspended or assigned secondary

priority once significant flooding started and that during the flood the Corps did not

operate for environmental or other purposes in a way to influence flood risk.”46 PX847 at

5. The plaintiffs have based their takings case on the theory that meeting the Corps’ ESA

obligations is the “single purpose” that allows the court to consider all of the Corps’

System and River Changes for all of the years together to determine whether the plaintiffs

have met their causation and foreseeability burden for establishing a taking. As discussed

44

The experts are Dr. Neil Grigg, professor of Civil and Environmental Engineering at Colorado

State University; Mr. Bill Lawrence, Hydrologist in Charge of the Arkansas Red Basin at the

River Forecast Center of the National Weather Service; Mr. Darwin Oekerman, Hydrologist at

the USGS; and Ms. Cara McCarthy, Senior Forecast Hydrologist at the National Resources

Conservation Service at the National Water and Climate Center. DX0192 at DX0192-0094.

45

The court has considered plaintiffs’ argument that the study was biased because Ms. Farhat

was involved in the study and because the Corps performed the study. The court has heard the

testimony and reviewed the study. The court does not find the study was biased.

46

The plaintiffs argue that portions of Dr. Grigg’s analysis support their position that earlier

releases would have led to less flooding. For the reasons stated in the next section of the opinion

the court finds that the argument regarding earlier releases is not supported.

81

in the Legal Standards section of the opinion, the court has accepted plaintiffs’ theory

that the cumulative and combined effects of the Corps’ System and River Changes can

give rise to a taking, to the extent the System and River Changes “serve a single

purpose.” Because the releases in 2011 were not part of the “single purpose” to comply

with the ESA, the flooding caused by the System releases in 2011 cannot be considered

with the other flood years to establish the plaintiffs’ takings claims.

Second, plaintiffs have not established that the 2011 flood by itself meets the

causation and foreseeability tests to serve as an independent basis for a takings claim.

Plaintiffs’ 2011 takings claims turn on their contention that when the Corps began

balancing priorities under the new Master Manual, the Corps abandoned flood control as

its first priority, and this change in priorities directly, naturally, and probably caused the

flooding in 2011. To prove their case, plaintiffs rely on a comparison of conditions in the

upper Basin in 1997 with those in 2011. Tr. 4590:11-4625:25; PX2203; PX2208. Dr.

Christensen testified that his “but for” model for 2011 was based on the Corps’ response

to upper Basin conditions in 1997. See, e.g., Tr. 4590:24-25, 4591:17-20. The court

finds, however, that the evidence established that there were significant differences

between conditions in the upper Basin in 1997 and 2011 and that the premise of

plaintiffs’ takings claims for flooding in 2011 is not supported.

The two major differences that led to different responses by the Corps to upper

Basin runoff in 1997 and 2011 are as follows: first, Basin runoff started earlier in 1997

than in 2011 because the weather was colder in 2011 and, as a result, inflows to the

reservoirs in 2011 were lower in March and April than in 1997. DX0192 at DX0192-

82

0036, DX0192-0050, DX0192-0053, DX0192-0081-2; DX3001-202; DX3015-656;

DX3015-660. Second, runoff in 2011 was twenty percent higher than in 1997, meaning

the runoff in 1997 was significantly less than in 2011 (49 MAF in 1997 compared to 60.8

MAF in 2011). DX0192 at DX0192-0010, DX0192-0016, DX0192-0061. These

differences mean that the Corps’ response in 2011 cannot be fairly equated to the Corps’

response to upper Basin runoff in 1997. Dr. Christensen’s modeling of the “but for”

world depended upon finding that upper Basin runoff in 2011 was virtually the same as

1997. The court finds that Dr. Christensen’s opinions regarding the “but for” world in

2011 did not take into account the significant differences in the timing and volume of the

inflows between 1997 and 2011. As such, the court cannot find that “but for” the new

Master Manual, the Corps would have begun to make releases from the dams earlier and

could have avoided the 160,000 cfs releases that resulted in devastating flooding in 2011.

For similar reasons, the court finds that the increased flooding in 2011 was not the

natural, direct, and probable consequence of the new Master Manual. The plaintiffs

argue that the 2011 flood was the foreseeable result of the Corps’ policy change, which

no longer placed flood control as the Corps’ first priority except when there were threats

to human health and safety. The new Master Manual requirement to balance priorities

did not mean that flood control was ignored by the Corps during March and April 2011,

thus making flooding inevitable. To the contrary, the Corps claims that it did not make

releases in March and April 2011 because it was in fact concerned with downstream

flooding. Tr. 7121:8-7122:6, 7123:11-7132:5 (Farhat); PX691 at USACE1012607 (map

showing the potential flooding hotspots). The Corps’ concerns in March and April 2011

83

demonstrate that flood control, particularly downstream of Gavins Point Dam, was

considered by the Corps in making its System release decisions in 2011. The court

appreciates that when the Corps was faced with downstream flooding in 1997, Corps

personnel made a different call than that of Corps personnel in 2011. The evidence

established that in 1997, Corps decision-makers elected to increase flooding downstream

in March and April because of the large upper Basin inflows the reservoirs were

receiving at that time. In 2011, in contrast, upper Basin inflows did not dictate that the

Corps should begin to make releases in March and April, which is when large releases

could typically cause or increase downstream flooding, because such inflows came later

in April and May. In both 1997 and 2011, Corps personnel made System release

decisions based on their best judgment in light of the information they had at the time.

The plaintiffs’ contention that the government took a flowage easement from

plaintiffs during the 2011 flood because the new Master Manual requirement to balance

priorities meant at some point the Corps would not have enough room in the reservoirs

and would thus need to flood properties in both the upper and lower Basin is not

supported. There are situations where the government has obtained flowage easements

when it knows it will have to flood an area in order to protect one group of landowners

during a flood event at the expense of another group. That is not the case here. 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

84

over another. The Corps was forced to flood both upper and lower Basin properties to

save the mainstem System.

The court also recognizes that the Corps could have exercised its judgment

differently in 2011 and perhaps minimized some of the terrible floodi

This text is long and has been trimmed here. Open the source document for the complete record.