Armstrong et al v. United States of America

Filing 13

FINDINGS OF FACT AND CONCLUSIONS OF LAW. Signed by Judge Stanwood R. Duval, Jr on 4/12/2013. (filed in master case doc 21172)(my, )

UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF LOUISIANA
IN RE KATRINA CANAL BREACHES
CONSOLIDATED LITIGATION

CIVIL ACTION
NO. 05-4182

PERTAINS TO:
Armstrong, C.A. No. 10-866

SECTION "K"(2)

FINDINGS OF FACT AND CONCLUSIONS OF LAW
The last of the Katrina floodwall breach cases came to trial in this Court on September
12, 2012, and continued through September 28, 2012. Plaintiffs Kenneth and Jeannine
Armstrong, Fred Holmes, the Succession of Ethel Coats, Alvin Livers, and Clifford Washington
(“Plaintiffs”) allege that the negligent remediation of the East Bank Industrial Area (“EBIA”)
performed by United States and the Washington Group International, Inc. (“WGI”) resulted in
the North and South Breaches of the EBIA floodwall. 1 That floodwall ran along the east bank
of the Inner Harbor Navigational Canal (“IHNC”) providing protection to the Lower Ninth Ward
and parts of Chalmette. On the morning of August 29, 2005, when Hurricane Katrina
(“Katrina”) came on shore, these breaches resulted in the inundation and destruction of the
Lower Ninth Ward and parts of Chalmette. Plaintiffs were all victims of this flooding.
Jurisdiction over this action is exercised pursuant to 28 U.S.C. § 1331 (federal question
jurisdiction), 28 U.S.C. § 1346(b) (Defendant United States); 28 U.S.C. § 2671, et seq. (Federal
Tort Claims Act); and 28 U.S.C. § 1332(d)(2) as amended by the Class Action Fairness Act of
2005, Pub. L. 109-2; 28 U.S.C. § 1367.

After considering all testimony and evidence presented

at trial and the deposition testimony that the Court reviewed prior to the trial, the Court is

1

See DX DM-0015 Silva Tula Demonstrative at 11.

prepared to rule as follows. To the extent a finding of fact constitutes a conclusion of law, the
Court adopts it as such. To the extent a conclusion of law constitutes a finding of fact, the Court
adopts it as such.

A.

FACTUAL BACKGROUND
1. THE IHNC AND THE EBIA

The IHNC, also referred to as the “Industrial Canal” was built in conjunction with a lock
in 1923 to provide a means of marine navigation between the Mississippi River and Lake
Pontchartrain. It runs generally in a north-south direction. The most southern point of the canal
feeds into the Mississippi River and its most northern point feeds into Lake Pontchartrain.
During World War II, the Gulf Intracoastal Waterway (“GIWW”) was rerouted through the
IHNC with the Federal Government leasing the lock and a 2.1-mile reach of the canal. At that
time, the United States assumed its operation and maintenance and eventually purchased the lock
in 1986.
The River and Harbor Act of 1956 authorized the Mississippi River-Gulf Outlet
(“MRGO”), a navigation channel which was completed in the mid-1960's and which runs from
the Gulf of Mexico to the GIWW (Reach 2) in a north-south direction. The MRGO then turns
westward (Reach 1) where it then is co-extensive with the GIWW. The MRGO/GIWW then
forms a “T” intersection with the IHNC just above the Florida Avenue Bridge. At the time of
Katrina, the MRGO provided a hydrological connection between the IHNC and Gulf of Mexico.
The portion of the IHNC that is relevant for purposes of this litigation is that which runs

2

between the Florida Avenue Bridge at the north and the Claiborne Avenue Bridge to the south.
This portion of the IHNC transects the Ninth Ward of New Orleans.
The IHNC lock is located immediately south of the North Claiborne Avenue Bridge.
Because of the lock’s presence, there is no hydrological link between the IHNC and the
Mississippi River. The East Bank of the IHNC acts as the western border for the Lower Ninth
Ward and is protected by a floodwall constructed by the United States Army Corps of Engineers
as part of the Lake Pontchartrain and Vicinity Hurricane Protection Plan (“LPV”). The LPV was
a program designed to construct a series of control structures, concrete floodwalls and levees
intended to protect the Greater New Orleans area from flooding. The LPV was passed by
Congress on October 27, 1965 as the Flood Control Act of 1965, 29 Stat. 1073, 1077 (42
U.S.C.A. §1962d-5).

2. THE EBIA FLOODWALL
When the IHNC was first constructed, earthen levees were built to protect the Upper and
Lower Ninth ward neighborhoods from storm surge. The focus of this trial was on the
improvements made as part of the LPV to the East Bank Industrial Area Floodwall (“EBIA
Floodwall”). That floodwall runs parallel to Jourdan Avenue extending from the Florida Avenue
Bridge (the northern boundary) to the North Claiborne Avenue Bridge (the southern boundary)
and the lock system to the south. As part of the LPV construction process, the Corps undertook
geotechnical evaluations and analyses with the aim of building an I-Wall which would remain
stable in the event a hypothetical 13-foot storm surge would be propagated into the IHNC. This

3

storm surge was projected as part of the "Standard Project Hurricane" that formed the basis for
all levees and floodwalls designed and constructed by the Corps pursuant to the LPV.
In 1965, the top of the earthen IHNC levee was increased to an elevation of +9 ft and
19.5 foot long sheet piles driven through the levee and foundation soils to an elevation of -8
NGVD, embedding the sheet pile in a layer of fat clays above soft organic clays. Using current
datum NAVD88, this resulted in a bottom elevation of -10.5. (Expert Report of Timothy Stark at
14) From 1969 to 1970, the Corps shaped and sculpted the levee to provide a wider crown.
Then, a concrete flood wall crown (8 ft high and 2 ft wide) was constructed to cover the existing
sheet piles that had been installed in 1965 to create the EBIA Floodwall that is at issue herein.
(JX-01672, March 12, 2012,Expert Report of Francisco Silva-Tulla ("Silva") at 0021). No
erosion protection on the landside of the floodwall was authorized, designed or constructed to
protect the landside of the levee in the event that storm surge waters overtopped the floodwall.
(Trial Transcript ("TT") of Silva at 3697).
At the most northern point along the floodwall, a small portion of the wall was replaced
in 1982.

This new I-wall was constructed with longer sheetpile which was approximately 13

feet longer–34.5 feet in length and its tip penetration was at least 15 feet deeper (approximately
-25 feet below NAVD88). It ran for about 30 feet in a north-south direction (that is as the EBIA
floodwall was originally constructed. Then, it made a right angle to the west or toward the
IHNC. These two attributes–longer length of sheet pile and right angle–acted as a restraint to
hold that part of the wall from moving landward when the storm surge occurred. (TT of William
Marr ("Marr") at 1847).

4

In addition, there was a structural anomaly at the North Breach. The only mechanisms by
which the old and new portions of the floodwall were connected were by a joint and a field weld;
the original 1969 I-Wall with 19.5 foot sheet piles was connected directly to the 1982 I-Wall
with 34.5 foot sheet piles with no transition. (TT of Silva 3759-60). The connection between
these two sheet piles of different lengths was an “undesirable design feature” that made
structural failure, caused by increased stress concentrations at the connection point, more likely
during a storm surge event. (Id.; TT of Marr 1847-48). The failure to use a “step connection”
resulted in a stress concentration when the Katrina storm surge occurred. In addition, the joint
and field weld at the connection was very crude and had a very, very low resistance to tearing
when stress was applied. (TT of Silva at 3767).
While the levee was built to +13.5 feet, because of the compressible soils underneath the
levee, settlement and ground subsidence occurred. The effect was that parts of the EBIA
Floodwall were one to two feet below the original construction top of the wall height, and indeed
the lowest points coincided with the South Breach and North Breach locations. (TT of Marr at
1862; see JX-1864, Marr Report, Fig. 3-1, at -00342). By 1999, the lowest top of floodwall
elevation was +11.3 NAVD88 and included the section of the floodwall that failed at the North
Breach. The second lowest top-of-floodwall elevation was +12.1 feet and included the section of
the floodwall that failed at the South Breach. ((JX-01672, Silva Expert Report 12 March 2012 at
-0022).

2

The Court has used the Bates Stamp numbering of all of the exhibits cited. It is for this reason that each
such number is preceded by a hyphen.

5

3. IHNC SUBSURFACE GEOLOGY
The subsurface geology of the land from the IHNC to the landside of the EBIA floodwall
requires explanation as well. As noted, in order for the Corps to determine what was necessary
to insure the floodwall's stability, it had to have an understanding of the soil in which the
sheetpile would be driven. The EBIA area is nestled between the remnants of two major
distributaries, the Bayou Sauvage system and the Mississippi River system which means that
during the 4000 years of this area's geological formation, the EBIA area would have been subject
to swamp conditions in the area as it developed. (TT of Marr at 1757).
In 1966, in preparation for the construction of the new floodwall, the Corps prepared a
Generalized Soil and Geologic Profile (See JX-04, Design Memorandum No. 3 for the LPV, at
-0193). Using the Unified Soil Classification System (USCS), the Corp produced a cross-section
representation of the soil underneath EBIA floodwall. Id. In doing so, various soil borings
which were used to create this cross section were identified. The South Breach occurred
between Borings 4 and 5; the North Breach occurred in between Borings 7 and 8.
From this profile, as well as a number of subsequent investigations, it is clear that the
soil underneath the EBIA floodwall is composed of various types of clays. The bottom tip of the
sheetpile is for the most part imbedded in what is referred to as "fat clay" (CH) with the next
strata below categorized as "clay with organic matter" (CHO) below. Beneath that there is a
small lens of "sandy silt" (ML) which stretches from Boring 1 slightly past Boring 4 (the South
Breach area). This lens is very narrow at the site of the South Breach and is located somewhere
below 22 feet NAVD88. It is even more narrow between Borings 4 and 5. Below that strata is
again more fat clay reaching down to approximately 42 feet NAVD88 and deeper. (TT Marr at

6

1763-64). Boring 8 (the North Breach area) is quite similar, without any sandy silt lens;
however, it was noted in cross-examination that there are some incidents of "SIS" which
indicates that there are some layers of shell present in those borings.
Thus, for geotechnical engineering purposes, the Corps properly treated this soil as
having the permeability, strength, and compressibility characteristics of clay. The Corps prior to
the construction of the 1966 floodwall performed two types of geotechnical analysis–stability
and seepage. (See JX-004, Design Memorandum 3 at -0114). As to seepage, the Corps stated
in its Design Memorandum 3, "Based on the soil conditions along this part of the project and the
short duration of hurricane floods, hazardous seepage or hydrostatic uplift on the protected side
is not anticipated." (Id.) Dr. Patrick Lucia, a registered civil and geotechnical engineer,
testified at trial that the engineers knowing the soils had significant clay content with low
permeability would have no expectation that there would be development of flow or significant
hydraulic pressures under hurricane conditions. (TT of Lucia at 2925) Furthermore, he
confirmed that flow and seepage are interchangeable terms in the geotechnical lexicon (TT of
Lucia at 2927).
As to the Corps's stability analysis, it utilized undrained shear strengths to calculate the
resisting force of the clay soils against the anticipated 13-foot storm surge that would rapidly
load against the face of the floodwall (TT of Lucia 2950-51). Undrained shear strengths are
universally used by geotechnical engineers for clay soils subjected to short-term loading
conditions. Dr. Lucia explained that these measurements (undrained shear strengths) are
appropriate to model undrained flooding conditions because hydraulic pressures are unknown
and not predictable in these types of soils. ( Id.; TT of Marr at 1899). Engineers in designing

7

these levees and floodwalls use equations using various geotechnical measurements that produce
what is known as a Factor of Safety. In order for a floodwall to maintain global stability, that is
to be able to withstand short term loading conditions by storm surge, that Factor of Safety must
be equal to or greater than 1.3 for a flood condition. (JX-1777, Expert Report of Patrick Lucia
("Lucia") at-0066; DX-DM-1105-0038.).
The Corps also calculated a minimum control line which is a boundary line defining the
critical area where disturbing soils could potentially undermine the stability of the floodwall
during a storm surge loading condition. (TT of Lucia 2944-45; JX- 0004, Design Memorandum
No. 3, Plate 38 at -0193). That control line ran roughly 15-20 feet from the center line of the
floodwall. (TT of Lucia at 2946). Disturbing soils within this critical area could make it more
likely that the floodwall would lose the resisting force necessary to maintain stability against the
driving force of the projected storm surge–the force that the floodwall was designed to
withstand–thereby increasing the likelihood the floodwall could destabilize and be toppled into
the Lower Ninth Ward. (DX-DM-1005-0036-39; TT of Lucia 2944-45, 47; JX-1777-0066).
Moreover, Dr. Lucia made clear that in his opinion, any excavation outside of the minimum
control line would have no impact on the stability of the floodwall.
Lastly, the Corps considered the forces that could lead to "overturning moments" when
the design flood condition reached the top of the wall, evaluating various sheet pile depths for
stability against overturning forces using a factor of safety of 1.5. The final penetration depth
was chosen based on the Corps's determination that the floodwall would retain stability against
shear and overturning forces with floodwaters at the top of the wall. (JX -0004, Design
Memorandum No. 3 at -0114, Plates 37, 38, 39, 35).

8

4. THE EBIA REMEDIATION
The East Bank Industrial Area consists of the land area between the subject floodwall
and the IHNC upon which a number of marine related industrial sites were located. The Corps
purchased the site between 2000 and 2007 in preparation for the construction of a new lock
system to be located between the Mississippi River and Florida Avenue. (JX-1672, Expert
Report of Silva, 3/12/12 at -0011). As part of the overall plan, a by-pass canal was to be dredged
through this land. Thus, the Corps undertook the remediation of the area in 2004-2005. Two
concerns were presented in this undertaking–environmental and mechanical. The Corps had to
insure that any contaminated soil in the area caused by the previous industrial tenants would be
removed safely. Also, any obstruction to the dredging process had to be eliminated. Thus, all
remaining structures, both on land and subsurface, such as piers, pilings, slabs and the like, had
to be removed.
The EBIA is approximately 32 acres in size and is subdivided into six properties based
on historic land use. (JX-1780, Expert Report of Stark at -0027). These properties included
(from north to south along the IHNC) Boland Marine, McDonough Marine, Indian Towing,
Mayer Yacht, Saucer Marine and the International Tank Terminal. ( Id. at 29, Figure 5-2). The
North Breach occurred in the area of the Boland Marine property and the South Breach in the
area of Saucier Marine property. Each propery had its own structures and detritus, and
excavations were necessary to remove existing infrastructure and contaminated soils.
To that end the Corps utilized the services of WGI with whom it had an umbrella contract
known as a Total Environmental Restoration Contract ("TERC"). (JX-00048). Issuing its Task
Order 26, the Corps required WGI to draft a recommendations report and work plans to outline

9

what work was to be done in order for the site to be in compliance with Louisiana Department of
Environmental Quality ("LDEQ") standards. (JX-00049). The work would include site
characterization, demolition and environmental clean-up. Clearly from the evidenced adduced at
trial, the Corps oversaw and approved virtually all activity WGI undertook with respect to this
entire project.
The subsurface characterization consisted of the identification of buried structures and
foundations that could interfere with construction of the bypass channel. The removal process
involved a tremendous undertaking including building and slab removal, buried foundation
removal, pile removal, soil remediation, transite (or asbestos containing material) removal and
grid trenching, (TT of Morris at 129-30).3 The contamination of the soil at the EBIA included
that caused by fuels, oils, solvents and heavy meals deposited from dumping, spills and leaks. In
addition to buildings, foundations, and other buried strictures, other items had been buried at the
site, ranging from barges to garbage and debris. (JX-1672, Report of Silva, 3/12/12 at -0025).
In executing the contract, WGI was tasked with not only removing these obstructions but
then backfilling those areas disturbed and tamping the soil to fill these holes. Breaches occurred
opposite of the Boland Marine Site and the Saucier Site, thus a more specific review of the
activities at those areas is required.

A. Boland Marine Site
Focusing on the Boland Marine site in the vicinity of the North Breach, activities
included transite excavations, soil remediation excavations, excavations of structure, including
3

For a comprehensive pictorial overview of the area and its history, see Figure 11-2:2003 EBIA found in
JX-1672, Expert Report of Silva, 3/12/12, at -0018.

10

the "wedding cake" structure and piling, grid trenching and utility removals. ( See Expert
Report of Silva, 3/12/12, JX-1672- 0025; JX-01674, Silva Report, App. B, Figure 3, at -0004).
The closest transite excavation to the North Breach was 85 feet away form the floodwall and the
depth of such excavations was between 2 and 5 feet deep. These excavations were backfilled
with either sand or borrow pit material.4 Contaminated soil excavations were conducted in
accordance with Louisiana's Risk Evaluation and Corrective Action Program (RECAP) in a total
of sixteen areas on the Boland Marine site. The depths of these excavations ranged from 4 to 8
feet. These excavations were backfilled with native fill from the borrow pit and in some
instances imported river sand. (See JX-1672, Expert Report of Silva, 3/12/12, at -0026). Grid
trenching also occurred at this property on "25-foot spacings to a depth of 5 feet." (Id. at 0027). Furthermore, daily construction records indicate that spoil was used to backfill the
trenches.
As to the structures that were demolished on the Boland Marine cite, those structures had
pilings beneath their foundations. Also on the site were eight subsurface concrete and steel
foundations beneath the building slab. The deepest of these was dubbed "the wedding cake"
structure. According to Dr. Silva's report, that excavation was made to a total depth of 26.5 feet,
187 feet from the floodwall and 590 feet from the southern edge of the North Breach. Id. at 0028 and JX-1674, Silva Report, App. B, Figure 5 at -0006.

A cofferdam was used in its

removal. The excavation was backfilled with spoil from the excavation and borrow pit material
in compacted 2 to 3 foot lifts. Id. (Silva Report).

4

The "borrow pit" was an area created at the McDonough Marine site and was the subject of the only
geotechnical evaluation that was made. This pit provided most of the material used to backfill excavations
undertaken at the EBIA and thus provided for native clays to be returned to the excavations. In fact the borrow pit
remained unfilled at the time of Katrina and no breach occurred anywhere near it.

11

The pilings penetrated to the deepest point in the EBIA area ranging in depths from 30 to
65 feet. All of those pilings were at least 275 feet away from the North Breach. There were
other deep pilings at the Mayer Yacht, Indian Towing, McDonough Marine and Saucer Marine
sites that were removed. Some were backfilled; however, there is no proof presented that all
were. The Court accepts Dr. Silva's opinion, however, that "any cavities left behind from pile
extractions would close in a short time (days), due to the lateral stresses in the soft subsurface
clays found at the EBIA.". Id. at -0028.
Finally, a gas line and water line was excavated at the Boland Marine area. As Dr. Silva
described:
WGI removed a two-or three-inch diameter gas line from the canal side of the
floodwall. The line was excavated, the concrete encasement around the pipe was
broken, and the pipe was cut with a pipe cutter. (QAR 691) The remaining pipe
was then excavated back to Surekote Road and the trench was backfilled and
compacted. (Id.). Excavations of the utilities were not expected to exceed two
feet in depth. (QAR 688) The closest utility was more than 400 feet away from
the North Breach.
Id.

B. Saucer Marine
At the Saucer Marine site, similar action was taken by WGI. Soil remediation
excavations, structural excavations, including the sewer lift station and pilings, grid trenching
and utility removal all were conducted. There were 15 areas of contaminated soil excavations.
Their depths ranged from 3 to 13 feet. The excavations were backfilled primarily with borrow
pit material. (Id. at -0029). Grid trenching was also conducted on 25-foot spacings to a depth of
5 feet. "Spoil" from the excavations was "apparently" used to backfill these trenches. Id.

12

A significant number of structures were removed from the Saucer marine site as well
with two structures having pilings underneath their foundations which were also removed.
Those piling depths were recorded as 39 ft and 28.6feet (See JX-1672, Report of Silva, 3/12/12,
- 0030-31, JX-1674, Silva Report, App. B, Figure 10 at -0012). All of these pilings were at least
150 feet away from the South Breach. While some were backfilled with soil, the daily field
records do not confirm that all were. Nonetheless, it appears to the Court that it is more likely
than not that because of the nature of the soils in the EBIA, the holes or cavities left behind from
the pile extractions would close in a matter of days or months due to the lateral stresses in the
soft subsurface clays in the EBIA. (See JX-1672, Report of Silva, 3/12/12,- 0030-31; TT of
Rogers at 723-24 citing Rogers Dep. II at 190:6-191:1).5
Another significant removal was that of a sewer lift station which was excavated to a
total depth of 22 feet. This structure was located 87 feet from the floodwall, and 193 feet from
the northern edge of the South Breach. (JX-1672, Report of Silva, 3/12/12,- 0030-31). A coffer
dam was installed for the process which included the station being cut into sections and removed
with a crane. Its slab foundation was broken up and removed as well along with exposed timber
pilings. This excavation was filled with borrow pit material, "spoil", and sand in bucketcompacted 24-inch lifts. (Id.)
A gas line and water line were also removed. Dr. Silva describes the process as follows:
WGI removed the lines from the canal side of the floodwall. The 4
September 2003 preparatory meeting called for the lines to be excavated and the
concrete encasement around the pipes to be broken (QAR) The lines were cut and
removed back to the Surekote Road curb. The shallow two-foot excavations were
5

The Court was not overly impressed with Dr. Rogers' new found belief which he "researched" after his
deposition that led him to equivocate with respect to his initial deposition testimony. (See TT-Jonathan Rogers at
723-24).

13

backfilled in compacted one-foot lifts. (QAR 690) In October 2004, additional
utilities were removed from under Surekote Road on the Saucer Marine site. The
backfill material was semi-compacted with multiple passes of a dozer. (QAR
931) Figures II-7 9WGI016083) and 11-8 (WGI016083) are photographs taken
during the waterline removal at Saucer Marine.
(Id. at -0031-32). It must be noted that similar activities were undertaken at all of the other sites
along the canal and no breaching occurred anywhere else along the length of the EBIA
floodwall–only the North and South Breaches.

c. McDonough Marine and the Borrow Pit
Perhaps one of the most telling excavations with respect to the non-failure of the
floodwall concerns the Borrow Pit that was excavated at the McDonough Marine site. This
location was chosen because of its relative lack of contamination and obstructions; furthermore,
it is clear that no "clay layer" was installed there as the pit itself was made of silts and clays.
Indeed, this excavation is one of only three excavations for which a contemporaneous
geotechnical evaluation was made prior to its execution.

(TT of Jonathan Rogers at 747-48).

The depth of this pit varied with the greatest being 18 ft., decreasing to 14.8 ft, 10 ft, 6, ft,
5 ft and 4 ft at various points in the pit6 and the eastern edge was about 53 feet away from the
floodwall at McDonough Marine (JX-1672, Report of Silva, 3/12/12, -0033-34, JX-1674, Silva
Report, App. B, Figure 12 and 13, 0015-16). However, it is clear that the borrow pit pierced the
lower organic clay lens and reached the lower organic clay level. (TT of Stark at 3428-29) . At
the conclusion of its use, the berm on the canal side of the pit was breached to allow water to

6

These depths are not corrected to NAVD88; for instance, a depth of 14.8 ft would translate to -11.4 ft.
NAVD88. Id.

14

flood the excavation. (See JX-1672, Report of Silva, 3/12/12, at- 0033-34). Nonetheless, with
an open, un-backfilled, un-tamped large pit, no breach occurred at the EBIA in its vicinity.
Finally, two 4-foot deep contaminated areas under Surekote Road and adjacent to the
levee toe at McDonough were excavated with the Corps' geotechnical branch having provided
instruction for the backfill and compaction of these area.
Thus, between February 2001 and May 2005, WGI performed numerous definable
features of work, including but not limited to excavating the Borrow Pit, demolition and piling
removal; grid trenching; the Sewer Lift Station removal at Saucer Marine; the "Wedding Cake
removal at Boland Marine; soil remediation and utility line removals. WGI left the EBIA site
on May 26, 2005. After the Technical Completion Report was completed and approved by the
Corps on August 23, 2005, WGI's work was completed. (JX-1364 Technical Completion
Report).

5. HURRICANE KATRINA
Hurricane Katrina hit Miami, Florida on August 25, 2005, as a Category 1 hurricane on
the Safir Simpson scale. (Rec. Doc. 20920, Joint Pretrial Order, Uncontested Fact No. 65 at 36)
("PTO"). After crossing the southern tip of Florida, it entered the Gulf of Mexico on August
26th, where, under favorable conditions, it exploded into a Category 5 hurricane. (PTO,
Uncontested Fact No. 66 at 36). It then peaked in intensity the afternoon of August 28, 2005,
with maximum sustained surface winds of 175 miles per hour; hurricane force winds extending
100 miles from its center; and tropical force winds extending 22 miles from its center. (PTO,
Uncontested Fact No. 67 at 36).

15

On August 29, 2005, Hurricane Katrina made landfall to the east of New Orleans at
Buras, Louisiana at 6:10 a.m. It was a massive Category 3 hurricane with winds about 125 miles
per hour and generated the largest storm surge elevations in the history of the United States.
(PTO, Uncontested Fact No. 69 and 70 at 36). Storm surge is a wind driven process: the wind
puts energy into the sea surface through the waves and currents. The greater the wind, the
greater the storm surge will be. (PTO, Uncontested Fact 72, at 36).
With the existence of the MRGO and the massive destruction of the wetlands
surrounding it caused by the Army Corps of Engineers' failure to armor the banks of that
channel, the storm surge drove up the MRGO, into the GIWW and finally into the IHNC. By
5:00 a.m. on August 29th, the water level at the IHNC Lock was 10.3 ft NAVD88 (2004.65),
increasing to 11.3 ft by 6:00 a.m., 12.3 ft by 7:00 a.m., 13.8 ft by 8:00 a.m., peaking at 14.3 ft.
by 9:00 a.m. By 10:00 a.m. the height at the lock had reduced to 12.3 ft. (PTO, Uncontested
Fact No. 1 at 29).
Because the storm landed to the east of New Orleans, New Orleans was in the northwest
quadrant of Katrina's wind field with winds blowing from a north, north easterly direction prior
to the passage of the eye of the storm (PTO, Uncontested Fact No. 4 at 30). Thus the wind
direction did not shift to a westerly direction until after 10:00 a.m. on the morning of August 29,
20005.
The two aforementioned breaches occurred at the EBIA floodwall early on the morning
of the 29th. The North Breach occurred at approximately 6:00 a.m. on August 29, 2005 (JX1713-, Expert Report of Robert A. Dalrymple ("Dalrymple Report) at -0007, -00065) and
resulted in a 180-foot gap in the floodwall. The South Breach occurred along the Saucer Marine

16

parcel of the EBIA at approximately 7:00 a.m resulting in a 793-foot gap in the floodwall. (Id.
at -0008,

-00065; PTO, Uncontested Fact No. 75 at 37). These failures caused the

catastrophic flooding and immense damage to the Plaintiffs which resulted in their filing the
instant suit.

B.

ISSUES PRESENTED
1. Plaintiffs Allegations

Plaintiffs contend that by virtue of the defendants' failure to conduct a full and competent
geotechnical site assessment, failure to evaluate fully the impact of their activities on the
floodwall and failing to employ prudent engineering practices, WGI and the Corps breached
their respective duties to maintain and protect the integrity of the levee and floodwall system
along the EBIA. (Rec. Doc. 20945, Plaintiffs' Trial Brief, at 1). Plaintiffs posit:
WGI's allegedly negligent work created and/or exacerbated subsurface pathways
for Hurricane Katrina's surge water pressures, which were transmitted through
the improperly backfilled and compacted excavations directly into the underlying
swamp/marsh layer. This pressure transfer from the flood side to the protected
side of the floodwall resulted in significant uplift pressures beneath the
floodwall and played a substantial role in the ultimate failures at the North and
South Breaches adjacent to the Lower 9th Ward. These pressure transfers, uplift
pressures and floodwall failures could have been prevented had WGI and the
Corps complied with industry custom and practice, as well as the myriad of
policies and regulations mandating WGI and/or the Corps perform a geotechnical
analysis on the potential impact of this excavation activity on the adjacent EBIA
floodwall. Moreover, before Defendants completed their work, they knew (or
reasonably should have known) that the IHNC and the soils beneath the EBIA
were hydraulically connected; thus, an increased tidal or storm event in the IHNC
would result in enhanced underseepage and the transmission of hydraulic
pressures laterally across the EBIA site and beneath the floodwall.
Id. (emphasis added). For support of these theories, Plaintiffs primarily rely on the expert
testimony of Dr. Robert Bea which will be discussed in detail, infra.
17

2. Defendants' Responses
In response to Plaintiffs' allegations, WGI maintains that it owed no duty to the Plaintiffs
with respect to the harm they suffered as a result of Katrina because the risks were unknown and
attenuated. As to geotechnical evaluations, WGI argues that there was no duty because (1) the
Plaintiffs are not third-party beneficiaries of the WGI-Corps contract; (2) the Task Order 26 of
the TERC did not obligate or otherwise provide for WGI to conduct any geotechnical evaluation
of the levee and the floodwall; and (3) there is no other source of duty identified to create that
duty on the part of WGI. Furthermore, WGI maintains that it performed its contractual duties
with due care in all aspects.
Moreover, WGI contends that it is entitled to certain legal immunities. First it re-urges
its immunity from suit based on the Government Contractor Defense. It also argues that it is
immune from liability based on state law immunity pursuant to La. Rev. Stat. 9:2771 which
provides that a contractor is not liable for damage or injury caused by alleged defects in work
undertaken by him due to any "fault or insufficiency of the plans or specifications," assuming
that he performs the work according thereto.
Finally, WGI maintains that its environmental remediation was not a contributing cause
of the two EBIA floodwall breaches. In particular, WGI contends that Dr. Bea's analysis is
fatally flawed and that he has created a "third type of pressure" to explain the breaches which is
illegitimate and not based on proven scientific methodology. This argument is underscored by
what WGI maintains is the erroneous application of certain computer models– SEEP/W and
SLOPE/W.

18

The United States argues that Plaintiffs' claims are barred by the Flood Control Act of
1928, 33 U.S.C. § 702(c) because Plaintiffs' flood damages were caused by alleged negligence
primarily or substantially related to a flood control activity with respect to the activities
undertaken with respect to the lock replacement and the EBIA remediation. In addition, the
United States maintains immunity under §702(c) because the damages were caused by
deficiencies in the original design and construction of the EBIA floodwall. Moreover, the
United States maintains that Plaintiffs failed to prove the WGI's activities were a cause in fact of
the EBIA floodwall failures or that the Corps was negligent in approving WGI's work.
Likewise, the United States maintains that Plaintiffs' methodology and analysis used to
prove that the breaches were caused by uplift pressures is unsupported by accepted geotechnical
engineering principles. The United States contends that applying scientifically accepted soil and
fluid mechanics to the EBIA clay soil demonstrates that Plaintiffs' allegation that underseepage
and landside hydraulic uplift pressures were not contributing factors to either the North or South
Breach.

C. Burden of Proof
Plaintiffs' claims rest on the United States' and WGI's alleged negligence. Under the
Federal Tort Claim Act, the United States may only be held liable if, under the law of the place
where the alleged act or omission occurred, a private party would be held liable under like
circumstances. 28 U.S.C. §§ 1346(b)(1), 2674.

Thus the same legal standard applies for a

cause of action for negligence under the law of Louisiana as to both the United States and WGI.

19

A duty-risk analysis is the standard negligence analysis employed in Louisiana to
determine whether to impose liability under La. Civ. C. art. 2315. As stated by the Louisiana
Supreme Court:
The approach provides an analytical framework for the evaluation of
liability. One analysis requires proof by the plaintiff of five separate elements:
(1) the defendant had a duty to conform his conduct to a specific standard (the
duty element); (2) the defendant's conduct failed to conform to the appropriate
standard (the breach element); (3) the defendant's substandard conduct was a
cause in fact of the plaintiff's injuries (the cause-in-fact element); (4) the
defendants' substandard conduct was a legal cause of the plaintiff's injuries (the
scope of liability or scope of protection element); and (5) the actual damages (the
damages element). Fowler v. Roberts, 556 So.2d 1, 4 (La. 1989), reh'g granted
on other grounds and original opinion reinstated as supplemented, 556 So.2d at
13 (La. 1990). A negative answer to any of the inquiries of the duty-risk analysis
results in a determination of no liability. See Mathieu, 94-0952 at 11, 646 So.2d
at 326. See also, Daye v. General Motors Corporation, 97-1653, p. 9
(La.9/9/98), 720 So.2d 654, 660; Perkins v. Entergy Corporation, 98-2081 p.21
(La. App. 1 Cir. 12/28/99), 756 So.2d 388, 403, aff'd, 2000-1372, 2000-1387, 001440 (La.3/23/01), 782 So.2d 606.
Lemann v. Essen Lane Daiquiries, Inc. 923 So.2d 627, 633 (La. 2006).
The Supreme Court of the State of Louisiana has further noted that the initial inquiry to
determine if a party may be liable under the duty-risk analysis is cause-in-fact. Lasyone .v
Kansas City Southern R.R., 786 So.2d 682, 690-91 (La. 2001). Plaintiffs have the burden of
proving negligence and causation by a preponderance of the evidence, and proof is sufficient to
constitute a preponderance when the entirety of the evidence, both direct and circumstantial,
establishes that the fact or causation sought to be proved is more probable than not. Cay v. State
of Louisiana, Dept. of Transp. and Development, 631 So.2d 393, 395 (La. 1994). Stated
another way:
The plaintiff must prove that the alleged defect in the thing was a causein-fact of the plaintiff's harm. Brown, 707 So.2d at1242; Campbell, 648 So.2d at
902. A party's conduct is a cause-in-fact of the harm if it was a substantial factor
20

in bringing about the harm. Graves, 703 So.2d at 570 (citing Frank L. Maraist &
Thomas G. Galligan, Louisiana Tort Law, § 4-3, at 86-88 (1996); Edwards v.
Horstman, 96-1403 (La. 2/25/97), 687 So.2d 1007. For example, the act is a
cause-in-fact in bringing about the injury when the harm would not have occurred
without it. While a party's conduct does not have to be the sole cause of the harm,
it is a necessary antecedent essential to an assessment of liability. Theriot v.
Lasseigne, 93-2661 (La. 7/5/94), 640 So.2d 1305, 1310; Dixie Drive It Yourself
Sys. v. American Bev. Co., 242 La. 471, 137 So.2d 298, 302 (1962). Whether an
action is the cause-in-fact of the harm is essentially a factual determination that is
usually left for the factfinder. Boykin v. Louisiana Transit Co., 96-1932 (La.
3/4/98), 707 So2d 1225, 1231; Theriot, 640 So.2d at 1310.
Laysone, 786 So.2d at 690-98. In this judge-tried case then, the Court is the ultimate fact finder.
As previously noted, trial of this matter took place over the course of three weeks. Prior
to that time, the Court reviewed the deposition testimony of John Greishaber and more than 7
other witnesses from the Corps. As to WGI witnesses, the Court reviewed, inter alia, the
deposition testimony of Steven Roe, Anne Veigel and Philip Staggs. During trial, the Court
heard the testimony of the Plaintiffs, except for Ethel Coats who is deceased; her testimony was
by deposition. Plaintiffs presented testimony of Chad Morris who evaluated the spatial
characteristics of the EBIA and Dr. David Rogers who provided expert testimony concerning
the engineering properties of the soils found at the EBIA and the standard of care for performing
earthwork at the EBIA. However, the seminal witness for Plaintiffs' was Dr. Robert Bea whose
credentials have been accepted by this Court and whose experience concerning the LPV
floodwalls and failures has been a centerpiece of his work since Katrina and the devastation
caused by the LPV's failures. Dr. Bea testified for a total of four days. The testimony of Dr.
Bea's assistant, Diego Cobos-Roa, was presented by deposition. Thus, the overarching focus of
the trial was the expert testimony with respect to causation and the geotechnical methods used to

21

support Plaintiffs' theory of causation.

The Court will now focus on whether Plaintiffs have

carried their burden of proof as to causation.

D. CAUSATION AS POSITED BY PLAINTIFFS
Plaintiffs maintain that the vertical and horizontal uplift pressures induced global and
lateral stability failure resulting in the North and South Breaches. Plaintiffs contend that the
excavations allowed certain pressure to move instantaneously through the soil skeleton. They
posit that these uplift pressures compromised the entire system by essentially levitating the wall,
which allowed the lateral forces from the canal to laterally displace the floodwall. Because the
pressure from the storm surge loading did not dissipate, the excavations were a substantial cause
of the North and South Breaches. This theory was developed by Dr. Bea based on his field
observations at the 17th Street Canal piezometers and the EBIA sites and his use of the SEEP/W
model and the SLOPE/W models to replicate the pressures which he observed at the 17th Street
Canal and which he believes explain the failures at the EBIA floodwall.
As previously mentioned, global stability is a geotechnical term which refers generally to
the ratio of the resisting landside forces to the driving canal-side forces trying to push a
floodwall landside. Soil mechanics provide the tools to determine stability by quantifying the
physical properties of soils and using mathematical equations to demonstrate how those physical
properties affect the ability of soil to bear weight and resist water. This realm is that of
geotechnical engineers who make the necessary tests and calculations to try to insure stability in
the design and construction of floodwalls and levees, and they too are uniquely qualified to
determine the cause of a stability failure.

22

Shear strength analysis is one benchmark used by scientists to determine whether a
floodwall placed to a certain depth in a certain type of soil will remain stable during a defined
flooding event. (TT of Marr at 1890-91). This type of analysis was done when the EBIA
floodwall was designed and constructed in 1966 to provide protection for a 13-foot hurricane
surge event. ( JX-1777, Report of Lucia, at -0006). "The condition for inboard stability
analysis during the design flood condition was evaluated. (USACE 1966 Design Memorandum
No. 3). A short term critical factor of safety of 1.3 was targeted for the design criteria." (Id.)
Furthermore, as previously noted, the minimum control line–a boundary line defining the critical
area where disturbing soils could potentially undermine the stability of the floodwall during a
storm surge loading condition–was determined. (TT, Lucia 2944-45; JX 0004, Plate 38). That
control line ran roughly 15-20 feet from the center line of the floodwall. (TT, Lucian at 2946).
The EBIA floodwall breaches demonstrate that some force or forces were at play which
destroyed that stability of that floodwall. Thus, Plaintiffs' burden of proof requires them to prove
by a preponderance of the evidence what caused the global stability of the EBIA floodwall to fail
with translational instability (the wall sliding landside while upright) and the lateral
displacement in two places along the EBIA.
Prior to this trial, the Court had before it in 2008 a Motion for Summary Judgment that
was filed by WG in this matterI. At that time, the focus of Plaintiffs's causation theory
concerned instability that was caused by the effects of underseepage of water. Plaintiffs then
maintained that the deep excavations of the sewer lift station at Saucer Marine site and the
wedding cake structure at the Boland site, "'altered the flow of sub-surface ground water, opened
portals to sub-surface permeable layers , and allowed the unabated underseepage through

23

permeable soils below the flood wall during Hurricane Katrina when the EBIA was covered
with high water" which allegedly cause the North and South Breaches. In re Katrina Canal
Breaches Consolidated Litigation, (MRGO), 2008 WL 5234369, *1 (E.D.La. December 15,
2008) (emphasis added), rev'd 620 F.3d 455 (5th Cir. 2010). The previous espousal of this
underseepage theory as cause for the failures of the EBIA floodwall by Dr. Bea was underscored
at trial during Bea's cross- examination not only concerning of his testimony in this regard in this
suit, but with specific discussion of his testimony as to the underseepage theory of causation
presented in the Robinson and the Barge trials. (See TT, Bea at 893-399).
However, as a result of the jointly developed soils exploration program that took place
over several months in the summer of 2011 for this trial, it became clear that the soil in question
was substantially less permeable than previously thought.. Geotechnical engineers were
retained by all the parties in this case, a protocol was established and locations were determined
where soils would be tested. The program confirmed that the EBIA is predominantly clay.

1. PERMEABILITY
As the Court found previously, the soils of the EBIA are predominantly fine-grained,
low permeability clays. (TT of Dunbar at 1754-64; TT of Rogers at 606). Dr. Joseph Dunbar
testified convincingly concerning the characteristics of these soils particularly in light of his
participation in an effort to understand the depositional environments and how they correlate to
the engineering properties of the soils that define the work of geotechnical engineers. (TT of
Dunbar at 1752). As such, notwithstanding the fact that there are "lenses" of other materials
present in these soils, impermeable clay soils dominate both the North and South Breach sites.

24

This fact is bolstered by the borings and photographs taken during the aforementioned soil
exploration program. (TT of Dunbar at 1779).
Permeability is synonymous with the term "hydraulic conductivity" In other words,
permeability "is the ability of a porous material, such as soil or rock, to allow the passage of
liquid, such as pore water." (TT of Silva at 3715). Water moves quickly through gravel and
sand; however, clay does not allow water to flow freely. It is for this reason that clay is
considered "impermeable." Hydraulic conductivity is a more precise term in that it refers
directly to groundwater as the liquid source. Id. at 3716.
In Court Dr. Silva demonstrated credibly using a foot of material in a tube and applying
half a foot of water to that material results in the following:
Material

Permeability

Time for material to flow through
container

sand

10G2 cm/sec

10 minutes

silts

10G4 cm/sec

21 hours

clays

10G6 cm/sec

90 days

(TT, Silva at 3716-17). The parties stipulated that the permeability of the clay at the EBIA is
10G5 cm/sec. Dr. Stark presented compelling evidence which the Court accepts that the sitespecific permeability of the organic clay soil at the EBIA would result in it taking one minute to
generate four teaspoons of water to flow through that clay (TT of Stark at 3414). With this
inexorable fact confronting Plaintiffs' experts, a shift in causation from water seepage to
pressure as the primary cause of the uplift pressure and de-stabilization of the floodwall
occurred.

25

2. HYDRAULIC PRESSURE CONDUCTIVITY AND UPLIFT PRESSURE
As Plaintiffs state in their Post Trial Brief (Rec. Doc. 21110 at 64 of 121 through 68 of
121), there are three types of pressure that could effect the global stability of a floodwall: (1)
total stress pressures, (2) pore pressures due to flow, and (3) uplift pressures. Total stress
pressure is exerted by the IHNC waters and is ever present; however, Plaintiffs do not posit that
this played a causative role in the EBIA disaster. Pore pressures associated with water flow is
another type of pressure. The time required for this kind of pressure is a function of the
permeability and the compressibility of the soil.

(TT of Silva at 3737: 11-16) . Plaintiffs do not

contend that flow-related pressure was causative either.
Plaintiffs base their theory of causation on uplift pressure. Dr. Bea's testimony focused
on this source of pressure which he maintains is “the near-instantaneous transmission of pore
pressures that occurs in the saturated high water-content soils of the EBIA." (Plaintiffs' PostTrial Brief, Rec. Doc. 21110 at 65 of 121). Plaintiffs opine that "[t]his third type of pressure is
communicated through the pores in the soil, just as "flow" of water and its associated pressures
communicate through the pores in the soil, but the terms of "flow" and "uplift" pressures are
used to avoid confusion of these types of pore pressure." (Id.)
Dr. Bea used the analogy of a car brake to demonstrate his theory of uplift pressure.
"There need not be a significant 'flow ' of brake fluid for the brake pedal to activate the brake pad
against the shoe. The pressure from the pedal transmits immediately to the brake shoe and
"flow" (i.e., a leak) would actually undermine the system." (Id. citing TT of Bea at 1299-1301).
Dr. Bea testified, "The high-water content soils are the soils that are able to transmit hydraulic
pressures. If the water contents are . . of the order of 30 percent, it will be the soil structure that

26

is transmitting pressure, and that's much less efficient at transmitting pressure. So water content
is critical to pressure transmission." (TT of Bea at 1432, 1434). Thus, Bea concluded that high
water content is synonymous with an effective pressure transmission. (Plaintiffs' Post Trial
Brief, Doc. 21110 at 67 of 121, n. 205). Indeed, Plaintiffs contend that "[t]he flow of water
through this soil skeleton may take time, but the pressure transmission is nearly instantaneous
due to the high water content." (Id.). Indeed, Dr. Bea testified that "hydraulic pressures are
relatively insensitive to the hydraulic conductivity we have called permeability." (TT, Bea, at
903). In essence, Dr. Bea maintains that the uplift pressures caused by the instantaneous
pressure transmission compromised the entire system essentially levitating the wall, which
allowed the lateral forces from the canal to laterally displace the floodwall. (TT of Bea at 3946).

3. PIEZOMETER PRESSURE
Bea testified that he arrived at this theory of uplift as a mode of levee failure beginning
with testing done at the 17th Street Canal floodwall. Using piezometer readings collected from
piezometers7 installed on the Jefferson Parish side of that canal after Katrina, he stated:
. . . So I pulled the data out and plotted it and something just becomes so crystal
clear. As the water left rising and falling in the 17th Street Canal, the pressures in
the water of the swamp-marsh layer is following it. It's almost like a dog on a
leash, only the leash is damn short because the pressure responds is virtually
instantaneously. It's minutes. The data acquisition is not rapid enough to carry it
to seconds, but it sure as hell in hours. So you can track it up, you could track it
down.
And we said–or actually turned to a student and I said, "Do you have
those pressures in your lateral stability analysis?" And the student came back and
said, "No, that's not normal. We don't do that normally." And I said, "Well in
this case, God doesn't believe you're (sic) normally. She refuses to separate the

7

Piezometer is a method by which a scientist can measure the static liquid pressure.

27

effect. Let's find out what those pressures are. Let's put them in the stability and
let's see what happens."
All I can say is God richly rewarded that trail because, at the end, we
could predict where, when and how and why it failed and why it didn't fail and
the answer was uplift pressures that had been omitted by all investigating teams. .
..
(TT of Bea at 843-44).
Dr. Bea also testified unequivocally that in his analysis, "flow" is where "that molecule
of water on one side of that floodwall moves underneath the floodwall and up to the other side"
and which is a form or element of underseepage. On the other hand, "uplift pressure" is
"pressure transmission underneath the floodwall on the other side, but without the molecule of
water itself having gone under the floodwall." (TT of Bea at 848-49). Thus, Dr. Bea
continued:
". . . both breaches occurred crucially because of these uplift pressures transmitted
from the East Bank Industrial Area, and particularly, the poorly backfilled
excavations that contacts the swamp-marsh layer, which has some very unique
characteristics and properties, to then communicate to the landside of the
floodwall that produces and uplift force that then is the critical horse in this horse
race of instability causing failure defines as I've got a breach and I'm flooding the
Lower Ninth Ward. It's WGI's excavations [that] were a substantial contributing
factor without which the breach would not have occurred."
(Id. at 849-50).
4. SEEP/W ANALYSIS
To prove his theory of pore pressure transmission, Dr. Bea used a computer model
known as SEEP/W which computes flow and pressure based on basic inputs of soil properties
and geometry. Mr. Diego Cobos-Roa performed the analysis for Dr. Bea. Within the program,
a choice is available as to whether to use a "transient flow" analysis or a "steady-state flow"
analysis. (TT of Bea at 911). Dr. Cobos-Roa testified that in running the SEEP/W model, he

28

used transient flow analysis option of the program.8 However, in choosing a value for the
parameter that represents the compressibility of the system due to change in pore water pressure
(denoted as mv), (DX-DM-1006, Marr's Demonstratives at -105) rather than using the
empirically proven value for mv , Dr. Bea used a value that is a 30,0000 times lower or 1 x 10G9
1/psf or zero as his coefficient for compressibility.
Dr. Marr demonstrated this fact in his testimony synopsized in the following chart:
TEST DATA FOR COEFFICIENT FO COMPRESSIBILITY, mv, in ft2/lb
Coefficient of Compressibility
Lab test data-JX-1864, Marr Report

0.00003

Field pumping test (Naymik S=0.002)

0.00003

Normally consolidated alluvial clays and
organic alluvial clays and peats

0.00001 to
0.00007

Dr. Bea used in SEEP/W calculation

0.000000001

Id. Thus, Dr. Bea used a value for the compressibility (10G9) and permeability (10G5) for the
EBIA clay that is analogous to that of sandstone–that is practically incompressible and
impervious. (TT, Silva at 3805). In addition, Dr. Bea testified that based on his field
experience with the piezometer and the "immediate" change in pressure, he sought to obtain that
result using this computer model. The only way the sought-after result occurred was by using

8

At page 70 of 121 of Plaintiffs' Post-Trial Brief (Rec. Doc. 2110), Plaintiffs state that Dr. Bea was not
performing a transient flow analysis. As support for this proposition, they state, "Dr. Bea described the event and
conditions as transient, i.e., "time-dependent," but the analyses were based on steady flow conditions. (JX 01414
and DX 02688 (Bea Rebuttal Rep.) at 17-18 ¶ 20); see also JX- 0193 (Bea Rep. Appx. D at 1) ("the transient
hydraulic conductivity analyses initiate from a steady state condition.") Unfortunately, this "proof" is indirect and
circular in nature and is rejected by the Court. Furthermore, regardless, it is clear that Dr. Bea did not use values for
compressibility in any of these computer models that jive with the actual soil that is present at the EBIA.

29

this extreme value for the compressibility coefficient rather than the actual value found in the
field. (TT, Bea at 932-33).

5. SLOPE/W ANALYSIS–Slope Stability Analysis
Dr. Bea also evaluated the lateral stability (or slope stability) of the floodwall under
maximum demands and capacity using the SLOPE/W program and the pressures generated by
the SEEP/W analyses. (JX 1393 Bea Rep. Appx. D at 21). This inquiry examines the ability of
soil to resist the weight or load being imposed upon it by, in this instance, a storm surge event.
(TT of Brandon at 3159). A core variable used in this analysis is the value used to represent the
strength of soil which is measured in two different ways, depending on the loading condition.
One is undrained and the other is drained. If a scientist determines the soil is undrained, a
total stress analysis should be used. Alternatively, if a scientist finds that the soil is in a
drained condition, then an effective drained analysis should be used.
Whether a soil in geotechnical terms is in a drained or an undrained condition depends on
the speed with which water can move in or out of that soil unit in comparison with the length of
time that the soil is being subjected to some change in the load on that soil. (TT of Bea at 1099).
Drained is the condition under which water is able to flow into or out of a mass of soil in the
length of time that the soil is subjected to some change in load. Conversely, undrained is the
condition under which there is no flow into or out of a mass of soil in the length of time that soil
is subject to some change in load. (TT of Bea at 1100).
The reference to a "loading condition" in the context of the EBIA was the storm surge
caused by Katrina that came up the IHNC and bore on the floodwall, putting stress on the soils at

30

the foundation. As stated by Dr. Marr, "If the loading is going to be very slow relative to how
fast water can flow through these materials, we would measure drained shear strength. In the
case of the EBIA, the geotechnical engineers setting up the testing program made an assessment
that these were slow-draining materials, that the load would be very rapid relative to the rate at
which they could drain and, therefore, the appropriate shear strength to use would be the
undrained shear strength." (TT of Marr at 1895-96). The significance of this use is that it
yields a higher shear strength because the soil is in a certain condition, the loading occurs
quickly. The soil cannot change, so that shear strength remains. (Id.)
When one takes the combination of elements of soil and water and then applies a load,
such as the tidal surge which occurred at the EBIA in Katrina, part of the strength of the soil
depends on what happens to the pore pressure inside. If a load is applied rapidly, there is no
time for water to flow in and out of the sample and the water content stays constant. (TT of Marr
at 1896). Clay takes months to years to behave in a drained mode. (Id. at 1898). In Dr. Bea's
analysis of uplift pressures, he used a drained mode in his calculations. (Id. at 1898). Since the
clay soils of the EBIA were loaded in a matter of 30 hours maximum, any stability analyses of
the floodwall involved the strength of the clay should be using undrained shear strength. (Id. at
1898)
In all stability analyses done for purposes other than and prior to this trial, Dr. Bea used
an undrained strength in his Total Stress slope analyses. Clearly, a "total stress" analysis is the
appropriate analysis to be used to measure the clay with undrained strength. (TT of Mar 193132). However, in his original expert report in this trial, he treated the organic clay layer as being
in a "drained" condition, using an effective stress analysis. (TT of Brandon at 3183).

31

In his Rebuttal Report, Dr. Bea opines that regardless of whether the soil is treated as
drained or undrained, the uplift pressures result in factors-of-safety that are "nearly identical"
and are in essence unstable. The Total Stress (undrained) method resulted in a Factor of Safety
of FS=1.1; the Effective Stress (drained) method resulted in a Factor of Safety of FS=0.99. (DX1625, Expert and Rebuttal Reports of Bea, April 11, 2012, ¶¶ 11-14 at -0006-0007; TT of Marr
at 1933). A Factor of Safety of 1 is unstable; thus, Bea posits that regardless of whether the soild
is treated as drained or undrained, the factor of safety hovers at 1 meaning there is instability..
However, Dr. Marr demonstrated these results were unreliable given that it appears that Dr. Bea
used different shear strengths in making his calculations. (TT of Marr at 1933-34).

6. BEA'S NORTH, SOUTH AND NEAR BREACH MODELS
These SEEP/W and SLOPE/W analyses were used to model three different scenarios–the
North Breach (Boland Marine), the South Breach (Saucer Marine) and the Near Breach
(McDonough Marine)–which Plaintiffs contend support their contention that uplift pressure
exacerbated by the WGI excavations was a substantial cause of the failures at the EBIA
floodwall. The summary of these evaluations can be found in Appendix D, "Analysis of Lower
9th Ward Floodwall Performance During Hurricane Katrina" to Dr. Bea's Original Report. (TT
of Bea at 1319; JX-01393).
The Court finds these models unconvincing for a number of reasons. Dr. Bea used
idealized excavations in his models; he did not attempt to recreate with any specificity the actual
excavations at either the North Breach or South Breach. (TT of Bea at 1021-27; 4021, 4060-61).
It was never clearly demonstrated how these idealized excavations comported with or compared

32

to the actual excavation. (See generally, TT of Bea at 1348-1350). No sufficient and clear
explanation was ever given to the Court as to why the actual excavations were not modeled;
furthermore, despite the Court's post-trial request to explain how the idealized excavations
conformed to the actual geometry of the area on the canal side of each breach, no explanation
was given.9
In addition, none of the models was ever run without any excavations which would have
provided the ultimate proof as to what effect the excavations had on the stability of the EBIA
floodwall at all three locations. (TT of Bea at 1337). In response to the Court's specific question
in this regard, Dr. Bea stated that his "control" scenario–the one without any excavation
penetrating the lower organic clay layer–was his "Near Breach" scenario at McDonough Marine.
(TT of Bea at 1292-94). This basis for Bea's contention that the Near Breach serves as a
"control" because those excavations did not pierce that "swampy-marsh" layer is inaccurate. It is
clear that that layer was pierced in reality. (TT of Stark at 3428; of Bea at 4131-32, 4134-35).
Thus, the Court finds that Plaintiffs have not provided the Court with any reliable data to
demonstrate the difference that these excavations made; Plaintiffs ask the Court to make a leap
of faith in this regard and the Court is not prepared to do so. In addition, Dr. Bea's initial
contention concerning a clay layer having been added to the borrow pit was ill-founded and any
assumptions based on that belief are likewise lacking merit. Finally, the Court finds that the
inputs used for these models to achieve the results that Dr. Bea postulated do not jive with the

9

In the Court's post-trial order with respect to issues it wished to be addressed, the Court asked Plaintiffs' to
"discuss whether any specific excavations were used in Dr. Bea’s failure analyses or were the excavations were
idealized. If so, how do the idealized excavations conform to the actual geometry of the area on the canal side of
each breach." (Rec. Doc. 21090 at 4). Plaintiffs did not accept this invitation nor any of the other 10 points as to
causation the Court mentioned.

33

actual scientifically ascertained values for compressibility and undrained soil making these
results suspect as explained in detail above.

E. PLAINTIFFS HAVE FAILED TO PROVE CAUSATION
Thus, the Court finds that Plaintiffs "proof" of a hydraulic connection and the resulting
"uplift pressure" as a substantial cause of the North and South Breaches is unavailing. The Court
found Dr. Bea's testimony not direct and quite circular. The North, South and Near Breach
models presented were unconvincing. Moreover, the Court is not persuaded that the use of
values in the SEEP/W modeling for compressibility that bear no relation to the field conditions
constitutes a valid scientific method.
Furthermore, the genesis for Dr. Bea's determination that uplift pressure was the cause
for the EBIA floodwall failures began with his observation of pressure readings in piezometers at
the 17th Street Canal. However, considering that the soil at the 17th Street Canal is substantially
different than that found at the EBIA renders the analogy suspect. Moreover, to the extent that
the basis for liability at the EBIA is the presence of poorly filled excavations, and in fact, there
were no excavations at the 17th Street Canal, further renders the underlying premise questionable,
i.e., that those excavations were a significant factor in the breaches regardless of whether there
was uplift pressure.
Simply put, Plaintiffs have not proven that it is more probable than not that the United
States' and WGI's remediation, excavations and backfill methods created a "hydrologically
charged" condition such that uplift pressures were transmitted through clay soil without any
appreciable flow of water to destabilize the floodwall causing its demise. Plaintiffs failed to

34

convince the Court that their theory of instantaneous pressure wave transmission or dilatational
wave theory had a valid application in solving levee and floodwall failures during hurricane
storm surge events. The Court accepts Dr. Marr's testimony that the dilatational wave theory
pertains to explosive shock waves that would increase stresses in the ground, both total stress
and pore pressure; however, there would be no change in strength. (TT of Marr at 1936-37).
Likewise, the decoupling of flow from uplift pressure did not seem sufficiently scientifically
supported. (TT of Brandon at 3169). Finally, the Court cannot base a finding of causation using
modeling that has used a value for the compressibility of the EBIA clay which is appropriate to
that of sandstone. Thus, given that Plaintiffs' have failed to prove the first element, causation, in
the duty-risk analysis, there is no need for further inquiry as to duty, breach, scope of liability or
damages.
The Court cannot and will not find as a certainty what exactly caused these breaches.
Clearly, it was not the Ingram Barge. Likewise, it is clear it was not the result of the subject
excavations performed by WGI under the complete supervision of the Corps. In this procedural
posture, the Court is not empowered to find the cause. It is noteworthy that the United States
opines in its Post Trial Brief that the structural defect in the design and construction of the
intersection between the shorter 1969 sheet pile and the longer 1982 sheet pile, which was driven
deeper in ground, was identified by their experts Dr. Marr and Dr. Silva as undesirable and
resulted in placing "far greater stresses on the older, shorter sheet pile than were accounted for in
the original design. (Doc. 2111 at 83-84). Likewise, the existence of a crude field weld at the
connection which had a very, very low resistance to tearing when stress was applied was
likewise identified. These floodwall anomalies created a dangerous condition in the event of a

35

catastrophic storm surge event such as occurred with Katrina. In another trial, the Court was
also shown that at the location of the South Breach, there was a slight kink in the line of the
floodwall, likewise making it more susceptible to stress in the event of a loading condition. In
addition, it is likewise noteworthy that there was no erosion protection on the land of the
floodwall, which guaranteed that with any overtopping, scouring would occur which would
encourage the floodwall's failure. In short, it appears that this floodwall, like the floodwalls at
the 17th Street Canal, the Orleans Street Canal and the London Canal, was a disaster waiting to
happen.

F. CONCLUSION
Since this opinion will likely be the last significant one issued by the Court regarding
Hurricane Katrina, I will engage in a bit of judicial license.
I have been the judge presiding over this hydra-like "Katrina Umbrella" litigation for
almost eight years. There are presently more than 21,166 entries in the consolidated docket,
which for pretrial purposes combined nine categories that have comprised the In re Katrina
Canal Breaches Consolidated Litigation. There have been many issues which have been vexing
and unique. Throughout these proceedings, I have been greatly assisted by the excellent lawyers
on all sides, as well as the outstanding work of Magistrate Judge Jay Wilkinson, Jr. in
attempting to resolve in as timely a fashion as possible the legal aspects of the monumental
disaster wrought by the failure of the levees surrounding New Orleans and vicinity as a
consequence of Hurricane Katrina.

36

One central theme has been painfully obvious throughout this entire process; many of
the levees protecting New Orleans and the surrounding area were tragically flawed. This fact
has been documented and detailed in several of my opinions10.
However, lamentably, there has been no judicial relief for the hundreds of thousands of
people and tens of thousands of businesses impacted by these defalcations. The Flood Control
Act of 1928 as interpreted over the years gives the United States Army Corps of Engineers
virtually absolute immunity, no matter how negligent it might have been in designing and
overseeing the construction of the levees. This Court did grant relief in the MRGO litigation
(Doc. 19415, In re Katrina Canal Breaches Consolidated Litigation, 647 F. Supp. 644 (E.D.La.
2009)), because I found the Flood Control Act of 1928 and the discretionary function exception
in the FTCA were not applicable in that instance.
In the trial of that case, it was proven being cavil that the Corps was aware that the
drastic widening of the MRGO, due to the Corps' failure to maintain it, endangered the levees
protecting St. Bernard Parish and the Lower Ninth Ward. Indeed, one memorandum from an
agency of the Army Corps of Engineers stated that the resulting losses in the event of a major
hurricane could be catastrophic. Therefore, this very real possibility was known by the Corps for
almost 20 years prior to Katrina and the Corps did nothing to rectify the problem, nor did the
Corps issue any specific warning to Congress or the public. Notably, the United States did not
appeal this Court's findings of fact of its negligent actions in that case.

10

For example, Rec. Doc. 10984, In re Katrina Canal Breaches Consolidated Litigation, 533 F. Supp.2d
615 (E.D.La. 2008) dismissing Plaintiffs' claims against the Corps with respect to the floodwall breaches at the 17th
St. Canal and other outfall canals.

37

A representative of the Corps admitted in trial that it was not "policy" for the Corps to
blithely stand by and take no action in the face of such danger to human life and property.
However, a higher Court in reversing itself and this Court found that the Corps was immune
because such decisions were "susceptible to policy considerations" applying the Discretionary
Function Exception to the Federal Torts Claim Act.
In the instant case, this Court has found the actions of the Corps in supervising the
remediation project executed by WGI along the EBIA did not substantially cause the North and
South breaches. As this Court has already noted that the floodwall that did not fail at the
northern most end was 13 feet longer, therefore deeper, than the floodwall that did fail. Of
course, if the levee was designed improperly the Corps is absolutely immune.
I feel obligated to note that the bureaucratic behemoth that is the Army Corps of
Engineers is virtually unaccountable to the citizens it protects despite the Federal Tort Claims
Act. The public fisc will very possibly be more jeopardized by a lack of accountability than a
rare judgment granting relief. The untold billions of dollars of damage incurred by the Greater
New Orleans area as a result of the LPV levee failures during Katrina speak eloquently to that
point.
I take note that the Corps of Engineers has many excellent and dedicated engineers,
supervisors, and staff. I also note that if individuals, corporations, and bureaucracies are never
brought to task for substantial negligence, each will be much less assiduous in discharging their
respective duties.
Accordingly, and based on the foregoing Findings of Fact and Conclusions of Law,

38

IT IS ORDERED, ADJUDGED AND DECREED that judgment be entered in favor of
United States and the Washington Group International, Inc. and against Kenneth and Jeannine
Armstrong, Fred Holmes, the Succession of Ethel Coats, Alvin Livers, and Clifford Washington,
with each party to bear its/his/her own costs.
New Orleans, Louisiana, this 12th day of April 2013.

STANWOOD R. DUVAL, JR.
UNITED STATES DISTRICT COURT JUDGE

39