No. 16-826L.


United States Court of Federal Claims.

Filed: October 25, 2021.

Attorney(s) appearing for the Case

Matthew S. Owen , Kasdin M. Mitchell , Erin E. Cady , Gavin R. Tisdale , and Neil A. Joseph , Kirkland & Ellis LLP, Washington, DC, for Plaintiff.

Davené D. Walker , Brent Allen , Trial Attorneys, and David A. Harrington , Assistant Chief, Natural Resources Section, Environmental and Natural Resources Division, U.S. Department of Justice, Washington, DC, for Defendant, with whom was Jean E. Williams , Acting Assistant Attorney General.


The Plaintiff in this Fifth Amendment takings case, Ministerio Roca Solida ("the Ministry"), has since late 2006 owned a forty-acre parcel of land in Nevada that lies within the boundaries of the Ash Meadows National Refuge ("Ash Meadows" or "the Refuge"). In 2008 the Ministry opened a church camp and retreat on the northeastern corner of the property. It named the camp the "Patch of Heaven."

Ash Meadows is located within the historic floodplain of the Carson Slough. It is home to many native plants and animals and contains one of the greatest concentrations of endemic species anywhere in the United States. Since at least the mid-1990s, it had been a goal of the United States Fish and Wildlife Service ("FWS") to restore the habitat of two endemic species of fish—the pupfish and the speckled dace. Restoration efforts were needed because private property owners who lived on the land before the Refuge was established in 1984 had diverted spring waters from their natural channels into irrigation ditches that they used for agricultural purposes. Because irrigation ditches do not supply a good habitat for fish, FWS planned a series of projects to fill in irrigation ditches and return spring waters back to their historic paths. One such project was the Fairbanks and Soda Springs restoration project, which was begun in 2009 and substantially completed in 2010.

It is undisputed that as a result of the project, the Patch of Heaven lost one of its most attractive features—the supply of spring water that had flowed through the property in a man-made ditch and which the Ministry had used for recreational purposes and to perform baptisms. The spring water was rerouted out of the man-made ditch that led to the camp and into to a "restoration" channel located on Refuge property that was upstream and to the east of the camp.

To make matters worse, in December 2010, just a few months after the bulk of the work on the restoration channel was completed, the area experienced five or six days of very heavy rainfall. Stormwater coming down from the upper watershed of the Carson Slough flooded the Patch of Heaven, damaging camp buildings, impeding access to the property, and sweeping away grass, trees, and other vegetation. From the Ministry's perspective, the loss of the spring water combined with the destructive flooding in December 2010 transformed its oasis, or "Patch of Heaven," into just another dry desert landscape, covered in silt and devoid of greenery.

In its original complaint, the Ministry asserted that the construction of the restoration channel had resulted in a Fifth Amendment taking of vested water rights. In a November 20, 2019 decision, the Court entered summary judgment in favor of the government as to that claim because it found that the Ministry did not prove that it had vested rights to the beneficial use of the water that flowed through the irrigation ditch and which FWS had rerouted away from its property. See Opinion and Order ("Op. and Order"), ECF No. 56.

The remaining claim before the Court is one for the taking of the Ministry's property by recurrent flooding. The Ministry alleges in its complaint that the construction of the restoration channel caused the destructive flooding that occurred in 2010, as well as subsequent floods that took place in 2015 and 2016. Specifically, it alleges that the berm FWS installed upstream of the Ministry's property to divert spring water into the restoration channel blocked flood waters from taking a natural path, one that would have discharged them on the western side of the Ministry's property, away from the church camp. Instead, according to the Ministry, the floodwaters were rerouted to the east at an elevation above the church camp, causing them to flow into the northeast section of the property where the camp is located. The Ministry further argues that, under the multi-factor analysis set forth in Arkansas Game & Fish Commission v. United States and its progeny, the recurrent flooding allegedly caused by the project constitutes a total physical taking of its property, or at least the taking of a flowage easement, for which it is entitled to just compensation. See 568 U.S. 23, 38-39 (2012).

The Court held a four-day trial on the Ministry's flood-related takings claim in Las Vegas, Nevada, the week of May 11, 2021. At the trial, the Ministry presented the testimony of, among others, Pastor Victor Fuentes, the Ministry's founder, as well as Ronald Matheny, the previous owner of the property. Both Pastor Fuentes and Mr. Matheny testified that, until FWS installed the restoration channel, they had never observed any significant flooding on the section of the property where the Patch of Heaven is located.

The Ministry and the government also presented the testimony of expert witnesses. Each retained a hydrologist to opine on the cause of the floods that occurred at the church camp in 2010, 2015, and 2016. In addition, the government presented the testimony of a meteorologist who offered his opinion about the weather conditions prevailing in the area at the time of the flooding events.

The Court has carefully considered all of the evidence as well as the parties' arguments. For the reasons set forth below, the Court finds that the Ministry failed to shoulder its burden of proving that the restoration project caused significant flooding of the church camp that would not have occurred but for the installation of the restoration channel. The Court therefore enters judgment in favor of the government with regard to the Ministry's takings claim.1


I. The Ministry's Purchase of the Property

Pastor Victor Fuentes and his wife, Mrs. Annette Fuentes, are the founders of Ministerio Roca Solida, a non-profit, non-denominational Christian church incorporated under the laws of the State of Nevada. Trial Tr. ("Tr.") vol. 1, 15:1-5, 54:24-55:3, ECF No. 124; see also Op. and Order at 2. In 2006, after hearing two members of his Bible study group complain of their inability to find a suitable program in the Las Vegas area to help their son who was struggling with alcoholism, Pastor Fuentes decided to search for property nearby that the Ministry could use for a church camp and retreat. Tr. vol. 1, 55:17-56:11. He wanted to find a place "close to nature" where campgoers could "relax," "calm down," and "be ministered." Id. 56:7-11.

To those ends, another member of the Pastor's Bible study group alerted him to a forty-acre parcel of land owned by Ronald Matheny that was within the boundaries of the Ash Meadows National Refuge. Id. 57:3-10; Plaintiff's Ex. ("PX") 269. The property was then, and remains, one of the few privately owned parcels located entirely within the Refuge's boundaries. Joint Stip. of Facts ("JS") no. 5, ECF No. 106; PX 451 (map of Ash Meadows National Refuge).

Mr. Matheny had purchased the land in January 1997 for the price of $60,000. PX 269 (quitclaim deed dated January 28, 1997); tr. vol. 1, 168:18-169:6, 177:6-8. At the time he bought it, the property was "just bare land" and sagebrush. Tr. vol. 1, 170:16-19. Mr. Matheny planted grass and trees and installed fencing, buildings, and other improvements, as well as water and sewer systems. Id. 170:20-171:14. He placed structures and fixtures on the property to replicate an Old West town and he also built a stage for musical entertainment, installed a campfire ring, and sawed down tree stumps that guests could use for seats. Id. 171:7-172:1.

Pastor Fuentes toured the property in 2006 along with several members of the church. Id. 55:10-14, 57:19-22. He found the spring water that ran through two irrigation ditches on the property, as well as its greenery, very appealing. Id. 58:2-16; see also PX 532 (aerial photograph of the property depicting the ditches). He anticipated that he could use the spring water to perform baptisms. Tr. vol. 1, 58:11-16. He also was pleased that the property was surrounded by a federal wildlife refuge. Id. 88:4-9. Because there would be no neighbors, camp visitors could stay up late into the night to "do whatever they want to do," work through "whatever problem they want to lose," and "get in touch with nature" away from city environments like Las Vegas. Id. 88:9-19.

In November 2006 the Ministry purchased the property from Mr. Matheny for $500,000. Joint Ex. ("JX") 5. A member of the Fuentes' Bible study group negotiated the price and donated the funds for the purchase. Tr. vol. 1, 65:17-66:14, 127:19-23.

II. The Construction and Launch of the Church Camp

Between 2007 and 2008, the Ministry and its members transformed the northeastern part of the property into a church camp. Tr. vol. 1, 68:1-6, 69:14-22. They cleaned the irrigation ditches, removed six or seven semi-truck loads of trash, installed electricity in all of the buildings, installed plumbing, painted, added insulation, converted the residential well into a commercial well, and prepared all of the buildings on the property for health and fire inspections. Id. 68:3-5, 69:3-5, 68:7-23. Much of the cleaning and refurbishing was done by members of the church who donated their labor. Id. 69:5-9. In addition, the church hired professional contractors for the more technical work, including the installation of the septic system and the grading of the land, and the excavation of the pond. Id. 69:9-13; 211:7-11 (testimony of Ronnie Murphy, general manager for Ron Murphy Construction). The church spent over $300,000 improving and readying the property, excluding the market value of the volunteer labor of members of the congregation. Id. 80:16-81:21; see PXs 46, 382-410, 415-21, 425-31 (checks, receipts, and invoices for improvements).

By the time the camp was opened in 2008, JS no. 9, it included a chapel, a commercial-grade kitchen, a dining hall, multiple bunkhouses, restrooms, a shower facility, and a snack bar, PXs 531, 532 (aerial photographs); tr. vol. 1, 65:6-14, 196:7-11. The camp could accommodate up to sixty-five guests at one time, tr. vol. 1, 83:24-84:1, and it regularly hosted groups from other churches across the southwest region, id. 82:10-19. In addition, other events were held on the property, including weddings, fundraisers, and retreats for the Ministry's own membership. Id. 82:20-24, 136:14-137:11. The Ministry requested a donation of $45 per person for use of the camp to help pay the expenses of its operation. Id. 83:15-23.

III. The Location of the Property and Attendant Flood Risks

The Ministry's land is located in the SW ¼ of the SW ¼ of Section 21 of Township 17 South, Range 50 East, MDB&M in Nye County, Nevada. JX 5. The property, as noted, lies entirely within the boundaries of the Refuge. JS no. 5; PX 451 (map of Ash Meadows National Wildlife Refuge). As depicted in the surface drainage map below, which is Joint Exhibit 3, both the Refuge (outlined in green in the lower left-hand corner of the image) and the Ministry's property (indicated by a red star) are located at the bottleneck of the historic floodplain of the Carson Slough (outlined in dark blue). JS no. 4; tr. vol. 3, 600:12-15, 724:5-10, ECF No. 126.3

The Carson Slough has a long history of flooding. Tr. vol. 3, 610:10-11, 640:3-9 (testimony of Kevin DesRoberts, the Project Leader of the Desert National Wildlife Refuge Complex since July of 2009); tr. vol. 2, 285:12-16 (testimony of Richard Johnson, former Nye County Floodplain Administrator), ECF No. 125.4 When rain falls in the Slough, it travels downward from northeast to southwest, "collecting and gaining water" along the way. Tr. vol. 3, 727:22-728:9 (testimony of Corey Lee, former Refuge manager). The floodwater ultimately flows through the bottleneck of the Carson Slough that cuts through the upper third of the Refuge, where the Patch of Heaven is located. Id. 696:12-16 (testimony of Robert Andress, former Otis Bay project manager, stating that flood flow runoff in the watershed funnels into the Carson Slough, that the Slough gets very narrow in the area of the Peterson Reservoir and the Ministry's property, and that all of the runoff must go through the narrow zone); see also id. 728:10-19 (Lee testimony, stating that flood flow runoff is funneled "into almost a pinch point" in the upper third of the Refuge); tr. vol. 2, 447:23-25 (testimony of Dr. Blaine Reely, Plaintiff's expert hydrologist, stating that the majority of watershed is upstream of Plaintiff's property).

A significant portion of Ash Meadows, and about two-thirds of the Ministry's property (including the camp location in the northeast corner), have been designated as flood zones by the Federal Emergency Management Agency ("FEMA"). JX 24 (FEMA Flood Insurance Rate Map); Defendant's Ex. ("DX") 47 (FWS document superimposing the boundaries of the Patch of Heaven onto the FEMA-designated flood zone); tr. vol. 3, 803:5-10 (testimony of Dr. David Thompson, government's expert hydrologist); tr. vol. 1, 180:19-23 (Matheny testimony); tr. vol. 2, 272:6-9 (Johnson testimony); tr. vol. 2, 421:2-12 (Reely testimony).

At trial, Richard Johnson, a former Nye County Floodplain Administrator, testified that he visited the church camp around late 2009 or early 2010. Tr. vol. 2, 268:24-269:9. During that visit, he told Pastor and Mrs. Fuentes that the property was in a FEMA-designated AO2 flood zone. Id. 269:8-21, 272:6-9.5 An AO2 designation means that the property is in a "special flood hazard area"—i.e., a high-risk area for flooding. Id. 283:4-6, 17-19.6 Mr. Johnson told the Fuenteses that the County's floodplain ordinance required that buildings in an AO2 flood zone be elevated by at least two feet or flood-proofed up to two feet and anchored. Id. 270:25-271:6. The Ministry did not take any action to bring itself into compliance with this requirement because, as Pastor Fuentes testified, he understood Mr. Johnson to have told him to hold off until the restoration project was completed. Tr. vol. 1, 130:18-131:23; see also tr. vol. 2, 271:5-272:9 (Johnson testimony, stating that, until FEMA did a base flood elevation study of Ash Meadows, he could not tell the Fuenteses by how much the buildings on the property should be elevated).

The record contains a significant amount of historical evidence that reflects flooding on the Ministry's property. Satellite imagery of the property in 1948 shows spring flows to the east of the property, and a flood channel travelling through the property from the northeast corner to the southwest. DX 4; tr. vol. 3, 599:8-21, 601:10-602:4 (DesRoberts testimony). The same natural flood channel running from the northeast corner of the property to the southwest is also visible on satellite images from 1980, see JX 27; tr. vol. 3, 602:5-20, 603:3-10 (DesRoberts testimony, describing Joint Exhibit 27), 1985, and 1989, see JX 14; tr. vol. 3, 604:18-605:5, 606:7-18 (DesRoberts testimony, describing Joint Exhibit 14); see also tr. vol. 3, 697:2-12 (Andress testimony, stating that elevation profiles, patterns of the landscape, and aerial photographs provide evidence of substantial flooding events in the past and that "signs of flow and scour east to west across that part of the upper Carson Slough" are "very visible on the ground and in the imagery").

Despite the foregoing, the previous owner, Mr. Matheny, testified at trial that, during his period of ownership (from early 1997 to late 2006), the northeastern portion of the property had never flooded. Tr. vol. 1, 173:5-174:4, 173:22-175:2. He acknowledged that during heavy storms he had observed flooding on the Refuge outside of his property. Id. 174:5-24. Those floods were "in the vicinity of the turnoff going into Peterson Reservoir," approximately half a mile from his property. Id. 174:5-15.7 But when storm water came onto his property from the northeast, Mr. Matheny testified, it either stayed in the irrigation ditches or else would discharge to the west of the buildings out channels that had been formed naturally by erosion from flooding. Id. 179:17-180:18; 182:1-11.8

Mr. Matheny acknowledged that during periods of heavy rainfall, the water would "puddle up" on his property and that the northeast field would become a "sloppy mess." Id. 173:10-16. However, according to Mr. Matheny, the storm water that ran through the irrigation ditches on the property never overtopped the edge of those ditches nor did it flood the structures he built. Id. 173:17-174:4.

IV. The Fairbanks and Soda Springs Restoration Project

A. Background

The Ministry alleges in this case that the floods that occurred on its property starting in December 2010 were caused by FWS's implementation of its project to restore the outflows of the Fairbanks and Soda Springs ("the restoration project" or "the project"). Those springs lie at the north end of the Refuge and are the first springs to feed into the Carson Slough. JX 20 (Environmental Assessment for Fairbanks Spring and Soda Spring Restoration) at 5, 9; tr. vol. 3, 609:5-12 (DesRoberts testimony).

As noted briefly above, during the twentieth century, and before the Refuge was created, the owners of what was then private property had drained the wetlands to facilitate farming and peat mining. Tr. vol. 3, 609:12-15, 655:6-22. They removed native vegetation, leveled the land, and put in infrastructure for irrigation by digging ditches into which natural spring water was diverted. Id. 655:18-22; 628:18-629:1. One of these ditches, running from north to south and into the Patch of Heaven, appeared sometime between 1985 and 1989. Tr. vol. 3, 605:16-23, 607:7-19, 631:15-632:5 (DesRoberts testimony, discussing Joint Exhibit 36 and observing that spring flows from Fairbanks, Soda, Rogers, and Longstreet Springs were all diverted and combined into a single ditch that headed to the west and then flowed down through the Ministry's property).

These actions, along with other human interventions, had caused "the entire wetland ecosystem [to be] altered and . . . [to] become infested with noxious weeds and aquatic invasive species," which "decrease[d] the viability for endemic species recovery and provide[d] habitat inconsistent with the needs of the species the Refuge was created to protect." JX 20 at 5; see also tr. vol. 3, 655:23-656:4 (DesRoberts testimony, stating that agricultural use had interfered with the periodic flooding of the Slough and the benefits of such flooding to native fish species).

Irrigation ditches do not provide a good habitat for endemic and endangered species, especially fish and other aquatic species. Tr. vol. 3, 594:20-22. FWS therefore planned to restore the "[c]onnectivity of the springs' outflows with the extensive wetland in the Carson Slough" by rerouting spring flow out of the irrigation ditches, removing existing physical barriers, and installing culverts that could be passed by native fish. JX 20 at 5.

B. Preparations for the Project

FWS publicized its intent to restore the Refuge's springs to their historic paths as early as 1995 in one of the Refuge's informational brochures. JX 45; tr. vol. 3, 590:24-591:14, 592:1-5. The brochure stated that FWS was then working to restore the habitat of Ash Meadows and that it expected it would continue its restoration efforts for many years. JX 45 at 2. It explained that the long-term plan for the restoration process included returning "[s]tream channels diverted into concrete irrigation ditches . . . to their natural courses" and removing "[n]onnative plants and animals . . . to allow native plants and animals to thrive without competition." Id.

In 2004, FWS retained environmental and ecological consulting firm Otis Bay Ecological Consultants ("Otis Bay"), tr. vol. 3, 669:22-23, to conduct the Ash Meadows Geomorphic and Biological Assessment, id. 671:7-13. Otis Bay reviewed the then-current status of the habitats and species of the Refuge, evaluated "the restoration potential of the springs, spring channels, and Carson Slough," and "propose[d] restoration recommendations." JX 4 (Ash Meadows Geomorphic and Biological Assessment) at 7. It completed its assessment in 2006. Tr. vol. 3, 671:20-22, 672:8-673:7 (Andress testimony, discussing the objectives of the assessment generally).

In August 2009, FWS issued a Final Comprehensive Conservation Plan ("the CCP" or "the plan") and Environmental Impact Statement for all of the refuges in the Desert National Wildlife Refuge Complex, including Ash Meadows. JX 7 (FWS Final Comprehensive Conservation Plan and Environmental Impact Statement); tr. vol. 3, 612:8-18 (DesRoberts testimony).9 A month later, in September 2009, FWS published its Environmental Assessment ("the EA") for the Fairbanks and Soda Springs restoration project for public comment. JX 20 (the EA). The EA, which was prepared by a private contractor, was posted on the Refuge website, within the Refuge itself, and at public locations in the community. Tr. vol. 3, 617:17-618:3 (DesRoberts testimony).

As described in the EA, the plan for restoring the Fairbanks and Soda Springs involved excavating new stream channels in the vicinity of the historic channels and removing and filling in obstructions to natural flow within those channels such as dams and old irrigation channels. JX 20 at 5; see also tr. vol. 3, 675:9-18 (Andress testimony, stating that the purpose of the project was "to take the[] outflow [of the Fairbanks and Soda Springs] out of [the] old irrigation ditches and . . . fields and focus them into individual stream channels in a more naturalized alignment and condition for the purpose of species recovery" and "ecosystem restoration"). The EA contained a map depicting an overview of the restoration project. JX 20 at 9. It shows the rerouting of the spring waters that then traversed the Ministry's property through the irrigation ditch so that they instead flowed through a new restoration channel to the east of the boundary of the property. Id.; tr. vol. 3, 618:24-619:17, 620:12-15 (DesRoberts testimony).

FWS retained Otis Bay to design and build the restoration project. Tr. vol. 3, 670:17-671:13, 706:21-707:12. Construction began in December 2009. JS no. 10. Otis Bay completed the restoration project in two phases. During Phase I, which took place between December 2009 and April 2010, Otis Bay built the restoration channel north of the berm it later installed to divert the spring water to the east and then south around the Ministry's property. Tr. vol. 3, 646:21-25, 647:18-20. During Phase II, which began in July 2010 and ended January 2011, Otis Bay installed the berm, which it constructed out of "dredged spoils." Gourley Dep. 214:1-6, ECF No. 119-1; tr. vol. 3, 647:2-5, 647:21-23.10

Otis Bay designed and constructed the restoration channel only to carry spring flow. Tr. vol. 3, 713:19-22. Its dimensions were about two to three feet wide and two to three feet deep. Id. 697:15-19. It was built to accommodate a flow of approximately 1,800 gallons per minute, or two to three cubic feet per second ("cfs"). Id. 697:20-24. The restoration channel was not designed or built to carry storm waters during flood events. Id. 713:23-714:7 (Andress testimony); see also id. 654:10-12 (DesRoberts testimony). According to Robert Andress, the Otis Bay project manager who worked on Phase I of the project, id. 676:11-13, 714:19-22, Otis Bay designed the channel with the expectation that any flood flow would travel to the same locations it always had traveled, and it was anticipated that the channel would be overwhelmed in any case if there were floods, id. 714:1-7.

As noted, Phase I of the project involved the construction of partially new channels in the upper section of the spring. Id. 706:21-24. As part of that work, Otis Bay excavated a stream channel and added culverts and a fish barrier. Id. 706:25-707:3. Mr. Andress testified that during the excavation process he used GPS to determine where to drive the stakes into the ground to cut the channels. Id. 708:7-13. He explained, however, that he might adjust the locations of the stakes by up to thirty feet if he concluded that it was necessary "to make it function better." Id. 708:14-25.

Otis Bay did not use the services of a professional engineer in performing its work. Id. 707:16-19, 709:4-6. Nor did it use stamped design plans (meaning plans that had been reviewed and approved by a licensed professional engineer) to complete the construction. Id. 707:16-22, 709:7-8, 866:16-19.

Otis Bay officially completed the project in January 2011. Id. 627:12-14.

V. The Floods

A. 2007 Flood

In the deposition he gave on behalf of Ministerio Roca Solida pursuant to Rule 30(b)(6) of the Rules of the Court of Federal Claims ("RCFC"), portions of which the Court admitted into evidence, see tr. vol. 1, 10:8-12:6, Pastor Fuentes testified that "the first time the property flooded" after the Ministry purchased it was in 2007. Ministerio Roca Solida Dep. ("Ministerio Dep.") 96:12-15, ECF No. 122-1. On that occasion it had rained north of the camp the day of the flooding and the day before. Tr. vol. 1, 143:25-144:10. Pastor Fuentes, on behalf of the Ministry, testified that, as rainfall flowed down the watershed, id. 144:11-13, muddy water slightly overtopped the culvert and the banks of the irrigation ditches, id. 78:18-79:23, 99:11-17, 141:1-5. Floodwater dumped mud and silt into a pond on the property that was fed by one of the irrigation ditches. Id. 78:19-24. Pastor Fuentes had to close the pond because it had filled with silt. Id. 78:24-79:2. He also had to hire a contractor to excavate the pond. Id. 78:5-21.11

B. December 2010 Flood

In late December 2010, the Patch of Heaven suffered significant flood damage after there had been multiple days of heavy rainfall in the upper watershed, with the heaviest rainfall occurring on December 22. Tr. vol. 4, 942:12-20, 974:18-20 (testimony of Jay Rosenthal, the government's expert meteorologist, explaining that there had been heavy rainfall from December 17 through December 22), ECF No. 127. Pastor Fuentes testified that he arrived at the property on December 23 after the peak of the flood flow had receded. Tr. vol. 1, 99:18-23, 147:6-8. Upon arrival, he could not enter the property because the culvert that enabled the restoration channel to pass under the access road to the Patch of Heaven had been blown out by flood water. Id. 100:1-7; PX 481 (photographs of washed-out culvert). Pastor Fuentes was eventually able to get onto the property with Mr. Matheny's help. Tr. vol. 1, 100:7-9. Mr. Matheny laid boards across the restoration channel so that both Pastor and Mrs. Fuentes could cross it and enter the camp. Id. 100:7-9, 102:25-103:3 (Fuentes testimony); id. 187:25-188:7 (Matheny testimony); PX 484 (photographs of boards laid across channel).

When they arrived on the property, Pastor Fuentes observed floodwater all around and coming from the direction of the restoration channel. Tr. vol. 1, 102:8-12. It "was rushing down the northeast field" to where the buildings were located, with "lots of water behind the lumber house." Id. The northeast field itself was inundated with water and the garden that once was there was swept away. Id. 103:17-104:1; PX 486 (photograph of northeast field). Floodwater also entered the barbecue area just outside of the snack bar and was several inches deep. Id. 147:9-148:1; see also PXs 506, 508. Mr. Matheny testified that, when he owned the property, he had never seen anything "even close" to the flooding he observed in December 2010. Tr. vol. 1, 189:21-190:4.

Ronnie Murphy, the general manager of Ron Murphy Construction, also provided testimony about the damage caused by the 2010 flood. Id. 207:5-9. He flew over the property in an airplane. Id. 220:23-221:2; see also PX 151 (aerial photograph). He testified that "when the rain came and it was really heavy and the water started flooding the whole property, we knew that something catastrophic happened upstream for that amount of water to be on the property." Tr. vol. 1, 221:3-15. Mr. Murphy was of the opinion that the flooding was caused when a "dam" located to the north of the Ministry's property "broke," id. 221:21, and that once it did "water went everywhere" and it "sheeted across pretty much the whole property, particularly in the area where their bunkhouses and stuff were," id. 222:4-12.12

C. October 2015 Flood

The property flooded again on October 18, 2015, after a series of heavy thunderstorms moved through the area, tr. vol. 1, 105:23-106:4, which on October 18 dumped "tremendous amounts of rain" "throughout the heart of the watershed" in a one-hour period, tr. vol. 4, 959:20-960:12, 963:9-13, 964:7-17 (Rosenthal testimony). Pastor Fuentes again arrived at the property after the peak flooding had passed. Tr. vol. 1, 106:5-10, 152:19-21. He testified that the damage looked "a lot like the 2010 flood." Id. 106:11-14. He explained that, as with the 2010 event, he again observed flood water entering the property from the direction of the restoration channel. Id. 106:13, 106:22-107:1. This time, however, water also came down from the north and flooded the property northwest of the bunkhouses. Id. 106:14-21, 110:6-111:1. A natural erosion channel that runs to the middle and then off to the west of Plaintiff's property also was filled with flood water during this event. Id. 153:18-154:4.

Flood water did not enter any of the buildings in 2015. Id. 154:9-11. Other than some standing water behind the snack bar, most of the water receded off the property within two days of the flood event. Id. 154:9-18, 157:16-23.

D. 2016 Floods

Pastor Fuentes testified regarding two flood events in 2016. Tr. vol. 1, 112:3-4. One occurred in January, when water entered the northeast hill and ran down through the former irrigation ditch and into his front field. Id. 112:5-22. He stated that the flooding "wasn't that bad" on this occasion. Id. 159:18-24. Water did not enter any buildings. Id. 160:6-8.

There was another instance of flooding on July 3, 2016, which Pastor Fuentes stated was "a little bigger" than the January flood. Id. 113:3-5. In the three preceding days, very strong thunderstorms had struck the area. Tr. vol. 4, 970:2-5 (Rosenthal testimony). On this occasion, water overtopped the culvert of the former irrigation ditch and ran through the front of the snack bar area. Tr. vol. 1, 113:2-15 (Fuentes testimony); PXs 298, 300, 301. Again, no water entered the buildings. Tr. vol. 1, 160:22-24.

E. March 2019 Flood

Pastor Fuentes testified that another flooding event occurred on March 9, 2019, a day that he was present on the property. Tr. vol. 1, 114:21-24; Ministerio Dep. 25:2-7. Pastor Fuentes stated that floodwater came from the northeast and flooded the northeast hill of the property. Tr. vol. 1, 115:2-4; Ministerio Dep. 25:8-13, 72:7-10. The flooding was limited to the four or five acres that comprise the northeast fields. Ministerio Dep. 28:4-14. Pastor Fuentes could not estimate the depth of the water on this occasion; he stated only that it was spread widely over the fields, id. 29:24-30:8, and that it then came down to the natural flood channel and flowed out in a southwesterly direction, id. 30:9-12. The water did not damage or enter any of the camp buildings. Id. 27:3-12. No work or repairs were required. Id. 27:13-18.

Mr. Corey Lee, the Ash Meadows Refuge Manager from January 2017 through November 2020, tr. vol. 3, 717:1-13, also offered testimony about the March 2019 flooding event. He observed that the night before the flooding occurred, it had started to rain and it was still raining when he arrived at the Refuge the next morning on March 6. Id. 736:5-18. He estimated that the Refuge received a half an inch of rain in twelve hours, which he characterized as "not a lot of rain" but enough to cause a lot of surface flooding. Id. 736:10-13. Mr. Lee testified that on that day he saw the flooding on Plaintiff's northeast field and determined it was approximately a half to a full inch in most places and moving southwest. Id. 757:4-14.

Mr. Lee visited different points along the Refuge and took photographs of his observations on March 6 and 7, 2019. DXs 193, 197, 198. From an area south of the Patch of Heaven, he observed that floodwaters generally moved in a southwesterly direction. Tr. vol. 3, 739:3-740:4. He observed the same trajectory of floodwater movement while standing north of the Patch of Heaven, in an area between the Fairbanks and Roger Springs. Id. 747:22-748:5, 748:16-19; see also DX 199 (video of floodwaters); DX 209 (photograph from area between the Fairbanks and Roger Springs); PX 451 (map of Refuge).

Mr. Lee also visited the restoration channel itself, at a location about half a mile to the north of Plaintiff's property. Tr. vol. 3, 749:25-750:11; DX 202 (photograph taken from restoration channel facing north); DX 200 (photograph of restoration channel). He testified that here, again, the water flowed from the northeast towards the southwest. Tr. vol. 3, 751:11-13. He stated that he observed that water coming from northeast of the restoration channel would flow into the restoration channel and then flow across the restoration channel and out the west bank. Id. 752:4-9.

F. April 2020 Flood

Pastor Fuentes testified that additional flooding occurred on April 9, 2020. Tr. vol. 1, 116:1-12; Ministerio Dep. 72:16-19. On that day, he witnessed floodwater on the northeast field similar to what he observed in March 2019. Tr. vol. 1, 116:15-18; PXs 703, 705.13 On this occasion again, there was no damage to the buildings. Ministerio Dep. 79:14-22.

G. Cumulative Impact of Floods

The 2010 and 2015 floods were the most severe of those about which the Ministry complains. The 2010 flood damaged the church camp buildings, causing the walls and ceiling to crack in both bunkhouses. Tr. vol. 1, 150:12-14, 150:25-151:8; PX 529 (photograph of water near buildings); PXs 547, 548, 550, 551, 554 (photographs of cracked bunkhouse walls). In addition, those floods (in combination with the subsequent milder ones in 2016, 2019, and 2020) caused significant damage to the landscape. The 2010 flood "washed away" the "vegetation" on the property, and it has not grown back. Tr. vol. 1, 104:4-9. Pastor Fuentes further testified that almost all of the floods left white silt in their wake that lingered on the property long after the floodwaters had receded. Id. 104:4-17 (stating that, after the December 2010 flood, "all the vegetation there died," and the area "was cover[ed] with the silt"); id. 224:15-18 (Murphy testimony, noting that the Ministry's "[p]roperty was completely covered in silt" after the December 2010 flood); id. 109:9-13 (Fuentes testimony, stating that the October 2015 flood left "white silt all over the field"); id. 117:4-11 (explaining that Plaintiff's Exhibit 703 showed "all the white silt due to the flood" after the April 2020 event); PX 596 (photograph of silt following March 2019 flood); tr. vol. 2, 349:4-20 (Reely testimony, describing white silt following flood events generally).

The estimated cost of repairing the buildings and restoring the landscape after the 2010 flood was $86,639. Tr. vol. 1, 227:16-18, 223:14-19 (Murphy testimony); PX 242 (2010 repair estimate from Ron Murphy Construction). The estimate included removing the white silt, bringing in rock to replace the eroded soil, repairing and replacing the irrigation system, and jacking up and releveling the two damaged bunkhouses. Tr. vol. 1, 224:8-18, 225:11-18, 226:19-227:1, 227:16-18 (Murphy testimony).

After the 2015 flood, Ron Murphy Construction estimated that it would cost $222,355 to repair the damage from both floods. Tr. vol. 1, 234:18-21. That estimate included additional costs for regrading and cleaning out the eroded channels, removing additional silt, replacing topsoil, reseeding grass, and regrading and filling erosion near the buildings. Id. 230:17-234:21; PXs 165-67, 592-95 (2015 repair estimates from Ron Murphy Construction).

Pastor Fuentes explained that the church did not have the resources to make these repairs and, even if it did, it would hesitate to invest any money because of the risk of continued flooding. Tr. vol. 1, 123:9-23. To date, the Ministry has only made minimal repairs to the church camp property. Id. 119:22-120:18.

Pastor Fuentes testified that the flood damage and the risk of future floods has hampered the Ministry's ability to fully use the land and operate it as a church camp. See id. 118:6-15. Use of the camp has declined, id. 119:10-20, although some groups continue to use it and pay to do so, Ministerio Dep. 33:24-35:10. One pastor testified at trial that his congregation stopped using the camp for its retreats in 2016 because the land that was once green and lush became like any other desert environment, and also because the water no longer ran through the property. Tr. vol. 1, 201:5-202:10. Pastor Fuentes also testified that he believed potential visitors are afraid to visit the camp because of the risk of flooding. Id. 118:10-11. He stated that during the flood of 2015, he had to help a pastor and his congregants safely exit the flooded property. Id. 118:12-17. After the more significant floods of 2010 and 2015, Pastor Fuentes testified he was not able to use the property at all for three or four weeks after each flood event. Id. 118:19-119:6.

V. Testimony of Plaintiff's Expert Hydrologist

A. Qualifications

Plaintiff's expert in this case was Dr. Blaine T. Reely. Dr. Reely is a civil engineer and hydrologist who has practiced in those fields for nearly forty years. Tr. vol. 2, 295:18-21, 298:13-16. He holds a Ph.D. in civil engineering, with an emphasis in hydrology. Id. 297:2-8. He is a registered professional engineer in a number of states. Id. 300:14-25.

Since 2002, Dr. Reely has owned Monsoon Consultants, where he serves as its principal engineer. Id. 298:17-20, 299:10-12, 299:19-20. The company performs engineering, hydraulic, and hydrologic analyses (including hydraulic modeling) in connection with public and private land-development and water-management projects, such as the construction of dams and flood-control projects. Id. 298:20-299:9 (explaining that Monsoon Consultants' civil engineers and hydrologists work for public and private clients); id. 296:11-16, 301:16-302:11, 305:10-306:5, 306:22-308:24 (describing experience in water resources analysis and engineering, flood control project engineering, hydrologic and hydraulic analysis of dams). Dr. Reely has testified as an expert numerous times. Id. 313:25-314:6. The Court qualified him as an expert in the fields of hydrology and civil engineering. Id. 324:3-4.

B. Dr. Reely's Opinion

At trial, Dr. Reely explained that the Ministry hired him to evaluate the conditions in the Carson Slough near the church camp property as well as upstream and downstream to get an understanding of the hydrology and hydraulics of the surface water system. Tr. vol. 2, 309:3-9. In addition, he was asked to evaluate the effect of the restoration project and the statistical relevance of rainfall events beginning with the 2010 flood. Id. 309:9-15. And ultimately, he testified, his charge was to "develop an opinion as to what . . . is happening and what might be causing some of that flooding—contributing to that flooding." Id. 309:18-21. He was also asked to review and comment on the report of the government's expert hydrologist, Dr. Thompson. Id. 309:22-310:3.

Dr. Reely formed his opinions primarily on the basis of his physical observations of the property and the surrounding areas, including the berm and the restoration channel. Id. 310:6-21. He visited/viewed the property three times: (1) in October 2013, id. 334:23-24, 443:9-11; (2) in January 2014 (via airplane), id. 336:21-337:7, 443:12-13; and (3) on October 30, 2015, id. 337:19-22. He also reviewed historic aerial photographs and flood data for the area, as well as topographic maps, and FEMA Flood Insurance Rate Maps. Id. 310:21-311:13.

1. Dr. Reely's Opinion as to Causation

Based on his observations, as well as the photographs and maps he reviewed, Dr. Reely concluded that the "predominant" cause of the flooding of the Patch of Heaven was "the construction of the diversion dam and the obstruction of a natural course of Carson Slough." Tr. vol. 2, 427:17-23. According to Dr. Reely, before FWS installed the berm and restoration channel, when floodwaters came down from the upper watershed and the western side of the Carson Slough, they flowed down the "natural trace" of the Carson Slough—a channel that runs along the very western edge of the Slough north of Plaintiff's property. Id. 340:4-10.14 Dr. Reely testified that "under natural conditions, big parts of that flow" would discharge out of the Carson Slough through a series of "off-ramps" on its western bank, thereby avoiding entirely the developed portion of the Ministry's property where the buildings and church camp are located. Id. 340:11-341:7, 342:18-21, 367:8-18, 344:4-14, 355:10-24.

Dr. Reely testified that when FWS placed the diversion berm in the path of the natural trace, it disturbed this historical regime. Id. 341:2-7. Doing so, according to Dr. Reely, not only prevented spring water from flowing through the trace, but it also blocked the path of flood water. Id. 344:17-18, 370:6-8. According to Dr. Reely, the berm kicked flood flow out to the southeast, where the flood waters would "tr[y] to follow the trace of the restoration channel." Id. 344:18-19, 370:8-17. The capacity of the restoration channel, however, was insufficient to contain the volume of floodwater passing through that point of the Carson Slough. Id. 340:19-22. As a result, Dr. Reely opined, the water would "break[] out" of the western bank of the restoration channel "at a number of locations." Id. 344:20-23. After escaping the restoration channel, according to Dr. Reely, the water would enter an intervening and pre-existing system of braided channels ("the braided channels" or the "braided channel network") that drain west and southwest down an elevation of approximately "six to ten feet." Id. 344:24-345:11, 351:8-17.15

The braided channels, which Dr. Reely physically observed, id. 356:1-3, orient generally to the south-southwest, id. 356:4-7. Some track directly towards the Patch of Heaven, while others track to the north side of the property and would exit into the natural drainage system to the west of the property. Id. 356:4-14. The network of braided channels predates the construction of the restoration channel. Id. 356:15-23. As more flooding has occurred, Dr. Reely opined, the channels are getting deeper and wider and the network itself is becoming more defined. Id. 357:2-5. This change in the hydraulic flow regime, testified Dr. Reely, will increase the amount of floodwater that exits the restoration channel and enters the Patch of Heaven via the braided channels. Id. 357:6-15.

2. Supporting Data

According to Dr. Reely, his expert opinion is supported by evidence he observed during his site visits and his analysis of aerial photographs. See tr. vol. 2, 310:6-311:13.

First, Dr. Reely testified, his opinion is supported by aerial photographs. Id. 310:21-24. Specifically, he cited Plaintiff's Exhibits 150 and 151, which depict conditions after the peak flow passed through during the flood of 2015. See PXs 150, 151; tr. vol. 2, 380:21-381:4 (describing Plaintiff's Exhibit 150); tr. vol. 2, 368:21-369:1 (describing Plaintiff's Exhibit 151). Those photographs show milky brown flood water in the restoration channel, but no water at all in the natural trace or the off-ramp channels. Tr. vol. 2, 380:20-382:2 (describing Plaintiff's Exhibit 150); id. 373:23-374:12, 374:23-375:11 (describing Plaintiff's Exhibit 151). Dr. Reely testified that the photographs indicated that floodwater did not overtop the diversion berm and never entered the natural course of the Carson Slough but was instead entirely diverted by the berm as he had opined. Id. 381:15-20.

Second, Dr. Reely testified that he observed silt with "alligator cracking" on Plaintiff's property and in the channels of the braided network. Id. 350:18-25. This cracking, Dr. Reely explained, shows that the restoration channel was carrying floodwater diverted by the berm and that this water entered into the braided channels. Id. 347:20-348:8. As Dr. Reely explained it, floodwaters pick up silt, including fine grain material and clay. Id. 348:2-4. When floodwater leaves the restoration channel and enters the braided channels, the rate of flow gradually declines allowing the silt to drop out of the water. Id. 348:9-19. As a result, a thin layer of "whitish" silt settles on surfaces traveled by the floodwater. Id. 348:19-23. The clay component shrinks as it dries out causing a distinct pattern of cracks called "alligator cracking." Id. 349:12-15. In a dry environment, the alligator cracking pattern can remain for several years following flood events before it starts to break down and be reincorporated into the soil profile. Id. 349:21-350:17.

Third, Dr. Reely testified that he observed features of the natural western discharge channels which indicated that floodwater entered those channels from the side (that is, from the restoration channel) rather than from the north (through the natural trace). Id. 378:9-17. He explained that the western discharge channels are generally oriented north-south. Id. 378:14-15. In his opinion, it was clear that water had cascaded into those channels from the sides because sideways moving water, which has a lot of energy, erodes the stream bank and creates a "headcut." Id. 378:15-25.16 Had the water flowed from the north from the natural trace, explained Dr. Reely, the vegetation in that area would have caused the water to slow down and have less velocity than water "dumping" into the western discharge channel from the side. Id. 379:5-16. If repeated flooding occurs, Dr. Reely concluded, the energy of the sideways moving water will continue eroding the bank until a new drainage system becomes stable. Id. 379:19-380:3.

3. Dr. Reely's Opinion Regarding Rainfall

Dr. Reely, as noted, was also asked to analyze the statistical relevance of the rainfall events that preceded the four flood events cited in the Ministry's complaint. To that end, he gathered twenty-four-hour rain gauge data for: December 22, 2010; December 23, 2010; October 18, 2015; January 8, 2016; and July 3, 2016, using the three rain gauges closest to the Patch of Heaven.17 He compared rain-gauge data collected for twenty-four-hour rainfall events that preceded the flooding to data published in the National Oceanographic and Atmospheric Administration's ("NOAA") precipitation frequency database. Tr. vol. 2, 393:17-22. Based on his comparison, he concluded that the rain events "that led to specifically the peak discharge associated with each of the floods" which he reviewed were not "statistically excessive." Id. 392:14-18.

Specifically, he concluded that the rainfall captured by the Amargosa Farms rain gauge over one twenty-four-hour period equated to a one-year event for December 23, 2010, and fell between a one- and two-year event for December 22. Id. 412:10-25. The Desert Wildlife Refuge gauge for December 23, 2010, showed precipitation frequency approaching a ten-year event. Id. 414:5-9. Last, the Desert Rock Airport gauge for December 23, 2010, showed precipitation frequency between a five- and ten-year event. Id. 415:9-12.

Similarly, rainfall on October 18, 2015, for the Desert Rock Airport rain gauge correlated to something less than a one-year event. Id. 415:21-416:16. Finally, he testified, the NOAA tables revealed that the rainfall on January 8, 2016, for the same gauge was approximately a one-year rainfall event and rainfall on July 3, 2016, was less than a one-year event. Id. 416:24-417:20.

VII. Testimony of the Government's Expert Hydrologist

A. Qualifications

The government's expert, Dr. David Thompson, is a self-employed civil engineer and hydrologist. Tr. vol. 3, 786:23-787:1. Dr. Thompson holds a Ph.D. in civil engineering with an emphasis on hydraulics and hydrology. Id. 787:5-7. He maintains a license in professional engineering in three states and has a certification in hydrology issued by the American Institute of Hydrology. Id. 787:21-788:1. In the past, he has worked for private consulting firms, the United States Geological Survey ("USGS"), and Texas Tech University, where he taught graduate and undergraduate course in hydrology, hydraulics, and numerical methods for fourteen years and directed several sponsored research studies. Id. 789:24-790:6, 788:5-18. Dr. Thompson has participated in more than thirty research studies, published approximately fifty conference proceedings, and written thirty referee papers. Id. 788:22-789:4, 792:7-8.

Dr. Thompson began doing private consulting work in the mid-1970s, id. 789:7-10, and, other than a stint with the USGS, id. 789:24-790:1, he has since continued to consult on a variety of projects requiring his expertise in hydrology and hydraulics, id. 790:12-25. Dr. Thompson has consulted on over fifty projects during his career as a hydrologist. Id. 792:3-7.

The Court qualified Dr. Thompson as an expert in the fields of hydrology and hydraulics. Id. 805:10-11.

B. Dr. Thompson's Opinion

Dr. Thompson testified that his assignment was to determine the cause of flooding for the flood events cited in the Ministry's complaint, which occurred in December 2010, October 2015, January 2016, and July 2016. Tr. vol. 3, 794:10-15. To that end, he made three field visits to the Carson Slough and the Patch of Heaven in January 2014 and in August and December of 2017. Id. 799:17-800:2, 807:4-6. He also reviewed topographic data and aerial photographs. Id. 797:13-17. In addition, he prepared hydraulic and hydrologic models. Id. 795:7-16.

Based on the information drawn from these sources, Dr. Thompson concluded that the capacity of the restoration channel was too small to have caused the flooding that occurred at the Patch of Heaven during each of the four events of 2010, 2015, and 2016. Id. 813:2-15. Instead, he opined, the floodwater that invaded the Patch of Heaven in 2010 and 2015, as well as that which saturated the northeast field in 2016, had flowed from the western to the eastern side of the Carson Slough floodplain north of the berm. Id. 820:1-6. It then flowed down through existing flood channels and over the restoration channel entering the existing braided channel network, which routed it directly onto the Ministry's property. Id. 837:18-23.

Dr. Thompson used two types of modeling software to analyze the causes of the flooding during the four events at issue: the Hydrology Engineering Center's Hydrologic Modeling System ("HEC-HMS"), id., 795:7-8, and the one-dimensional ("1D") version of the Hydrology Engineering Center's River Analysis System ("HEC-RAS"), id. 795:10-11, 795:24, 868:15-18. Both types of software were created by the Army Corps of Engineers and are used by hydrologists worldwide. Id. 795:7-11.

A HEC-HMS (hydrologic) model is "a rainfall runoff model or a watershed model" that hydrologists use to analyze how a watershed "transform[s]" rain "into a flow rate that exits the watershed at the outlet." Id. 795:7-8, 795:19-23. In other words, it calculates the peak discharge for the rain event being examined.

Creating the HEC-HMS model requires the hydrologist to input certain relevant data. Here, Dr. Thompson input topographic data into the model to delineate the watershed. Id. 797:23-24, 798:13-14, 800:3-20. He also input data about the composition of the soils. Id. 798:22-799:3, 800:23-801:18. This data enables the model to estimate "how much rain enters the soil and is lost and how much is left over to produce runoff that then moves across the watershed to the outlet from the watershed." Id. 801:14-18.

After he built the model, Dr. Thompson input rain gauge data "to produce estimates of discharge for the events in question from the Carson Slough watershed at a location near plaintiff's property." Id. 807:15-21. He chose a rain gauge that was within the watershed boundary and that measured rainfall in intervals of less than twenty-four hours. Id. 807:15-17, 826:2-25. Dr. Thompson also performed indirect measurements during his field visits to provide an "independent means of . . . examining the discharge from [the] Carson Slough." Id. 807:23-808:3.

In addition to the HEC-HMS (hydrologic) model, Dr. Thompson created a one-dimensional HEC-RAS (hydraulic) model "to examine how flow moves through the Carson Slough floodplain," which he constructed "with input from HEC-HMS" as well as "whatever supporting calculations were required." Id. 795:10-14, 795:24-796:2. He input topographic data that defined the watershed. Id. 798:14-17, 800:3-5. He also input soils data so the model could take into consideration their ability to shed or absorb water, id. 798:22-799:3, and a "parameter" to estimate the floodplain's "roughness," which is a measure of the friction (usually the amount of vegetation) that water encounters that causes it to lose energy when it passes over land, id. 832:22-833:13. Finally, Dr. Thompson input into the HEC-RAS model the volume of flood flow his HEC-HMS model had calculated that the rainfall generated during the four flood events. Id. 795:24-796:1.

Based on his modeling and physical observations, Dr. Thompson came to three primary conclusions. First, based on the HEC-HMS model, he found that the restoration channel could not have had any meaningful impact on flood flows entering the Patch of Heaven during the events at issue. Specifically, that model revealed that: (1) the peak discharge in the Carson Slough was approximately 3,200 cfs during the December 2010 flood event, 1,570 cfs during the October 2015 event, 1,720 cfs during the January 2016 event, and 1,620 cfs during the July 2016 event, id. 814:19-815:6;18 and (2) that the restoration channel's maximum flow capacity at the upper end of the channel was approximately 75 cfs when Dr. Thompson first visited the site in 2014 and 35 cfs when he returned in 2017, id. 809:19-25, 810:5-6.19 When he looked at the HEC-RAS model he constructed of the restoration channel itself, it showed that if as little as 150 to 200 cfs entered the channel, it would "readily spill[] out," leaving only 35 cfs remaining. Id. 812:15-25. Dr. Thompson therefore concluded that the capacity of the restoration channel (whether 35 or 75 cfs) was far too limited to carry the thousands of cfs of flood flow that came down the watershed for any of the four events. Id. 813:2-813:15 (opining that "[t]here's just no capacity [for the restoration channel] to carry significant amounts of flow").

Second, in an effort to determine what had caused the flooding on the Patch of Heaven, Dr. Thompson built a HEC-RAS model of the Carson Slough floodplain and the Patch of Heaven that did not include the berm and restoration channel. Id. 817:14-16. Dr. Thompson concluded that the Patch of Heaven would have flooded during each of the four events of 2010, 2015, and 2016 even if the restoration channel had never been built. Id. 817:17-818:1, 818:7-17, 870:5-22.

Dr. Thompson ran the 1D HEC-RAS model twice using two volumes of peak flood flow: one based on the HEC-HMS model's estimated discharge of 3,200 cfs to show the approximate depth of flow during the December 2010 flood event; id. 846:21-847:2; and another based on the HEC-HMS model's estimated discharge of 1,600 cfs to approximate the floods of October 2015, January 2016, and July 2016; id. 849:24-850:6.20 The models showed that the Patch of Heaven would flood anytime peak discharge in the Carson Slough reached a volume of 1,600 cfs or more. Id. 817:9-13.

Finally, Dr. Thompson concluded based on physical observations and topographic data that the flooding at the Patch of Heaven was the result of "rainfall-derived flood runoff" that migrated from the western to the eastern side of the Carson Slough floodplain. Id. 814:10-13, 818:7-17, 820:1-6. He explained that there was a "relative . . . topographic high" upstream of the berm, id. 846:3, where, he stated, flood flow would readily cross over from the west to the east side of the Carson Slough and tend to stay on the east side as it coursed down the floodplain, id. 819:6-12.

Dr. Thompson testified that his HEC-RAS model of the Carson Slough validated his conclusions based on his field observations and review of topographic data. It showed that for flood flows of 1,600 cfs or more, floodwaters on the western side of the model would cross over to the eastern side, id. 843:7-11, because the "depth [of the flood flow] is greater than the intervening [topographic] high between the western and the eastern [sides of the] floodplain," id. 847:12-18.21

Once the water was on the eastern side of the floodplain, Dr. Thompson opined, it would "readily bypass the restoration channel and proceed on in a natural direction towards [the] plaintiff's property." Id. 818:7-17. Dr. Thompson's thesis is consistent with two aerial photographs, Plaintiff's Exhibits 108 and 151, which show floodwater to the northeast of the restoration channel crossing over it. Id. 854:18-855:5 (describing Plaintiff's Exhibit 108); id. 810:25-811:2, 814:2-7 (describing Plaintiff's Exhibit 151). According to Dr. Thompson, once floodwater passed over the restoration channel it would enter the braided channel network to the southwest. Id. 836:22-837:4, 837:20-23; see also id. 854:18-855:2 (testifying that the aerial photograph comprising Plaintiff's Exhibit 108 showed water crossing the restoration channel and entering the braided channel network). From there, he testified, floodwater would be conveyed directly to the northeast corner of the Patch of Heaven, where the buildings are located. Id. 837:20-23, 838:11-17. Some of that water would break off into the western discharge channels, but the remainder would run across the church camp, most likely as sheet flow. Id. 843:16-21.

Dr. Thompson's HEC-RAS model showed that flow remaining on the western side of the floodplain (the flow that did not overtop the topographic high between the east and west sides of the floodplain), would still exit through the off-ramp channels along the natural trace. Id. 847:18-20. Dr. Thompson stated that he visited the site of one of the off-ramps and "confirm[ed] that it is actively discharging water whenever there's significant flows within the Carson Slough floodplain." Id. 839:15-23. Dr. Thompson opined therefore that neither the restoration channel nor the berm has a significant impact on whether flow on the western side of the floodplain will be able to enter the natural trace and leave through the off-ramp channels on its western bank. Id. 842:2-17.

VIII. Testimony of the Government's Expert Meteorologist, Jay Rosenthal

At trial, the government presented the testimony of Jay Rosenthal, a meteorologist. Mr. Rosenthal has been working as a professional meteorologist since 1963. Tr. vol. 4, 929:19-21. He has a bachelor's degree in meteorology from New York University and master's degree in meteorology from the University of California, Los Angeles. Id. 929:22-930:1. He spent over thirty-eight years as a civilian meteorologist for the United States Navy until he retired in 2001. Id. 930:19-22. He has also served as a consultant and worked part time as a research leader for Battelle, working on weather issues related to homeland security. Id. 932:15-23. The Court qualified Mr. Rosenthal as an expert in meteorology. Id. 933:7-8.

Mr. Rosenthal was asked to evaluate the flooding events described in the complaint from a meteorological standpoint to determine whether there were any noteworthy weather events that could explain the four episodes. Id. 929:2-6. To do so, he considered "[w]hatever ground-based rain gauge data was available, not just in the immediate area, but in the surrounding area." Id. 934:2-5. As Mr. Rosenthal explained, however, "particularly in a desert region, you don't have the luxury of having as many weather stations as you would, say, in an urban area." Id. 933:23-934:1. Because of the spatial gaps, he also looked at radar data, id. 934:8-10, and satellite imagery to get "the big picture," id. 934:13-16, as well as National Weather Service bulletins and advisories that provide warnings about inclement weather and potential flooding, id. 934:17-23. See also id. 975:13-976:8 (explaining the sources of data a professional meteorologist uses to address the scarcity of rain gauges in an area like the Carson Slough watershed).

Mr. Rosenthal testified that based on all of the data he considered, in his opinion, all four flooding events at the Patch of Heaven (particularly those that occurred in 2010 and 2015) were immediately preceded by "noteworthy" and "significant" weather events. Id. 937:15-18. For example, there was a particularly strong and persistent "atmospheric river" during the period that preceded the flooding at the end of December 2010. Id. 941:15-942:2, 942:21-943:8. This atmospheric river impacted all of the Southwestern states, including Southern Nevada and the Ash Meadows area in particular. Id. 944:14-18. It resulted in six days of rainfall in the area, beginning December 17 and lasting through December 22, the day of the heaviest rainfall. Id. 942:12-20. During that period, Mr. Rosenthal testified, the amount of rainfall "dwarfed the normal amounts expected" in the Carson Slough. Id. 948:7-13. In fact, he stated, for the six- to seven-day period that the rain fell in the region, it averaged three hundred to five hundred percent of the normal amount for the entire month of December. Id. 948:7-18. The rain gauge for Amargosa Farms, which is just to the west of the Carson Slough watershed, showed that during a three-day period, 1.77 inches of rain fell, which was 385% of the average rainfall for the entire month of December. Id. 949:17-18, 950:17-20.22 In addition, bulletins from NOAA as well as the National Weather Service reveal widespread flooding across Southern Nevada. Id. 951:23-952:11.

Mr. Rosenthal testified that, similarly, in October of 2015, the southwestern United States was in the midst of one of the strongest El Niño conditions on record. Id. 957:25-958:2. During that period, storm systems were "really energized," resulting in "periods of thunderstorms over the deserts that dumped a lot of rain." Id. 959:6-8. Mr. Rosenthal testified that the first period of thunderstorms occurred on October 4 and 5, and then another occurred October 15 through 18, just before flooding was observed on the Ministry's property. Id. 959:20-960:12. He observed that satellite imagery from October 18 showed "a whole line of thunderstorm activity moving from south to north right through the watershed area that may or may not be reflected by the few rain gauges" on the watershed's periphery. Id. 963:9-13. Further, radar imagery showed that "tremendous amounts of rain" fell "throughout the heart of the watershed" in a one-hour period on October 18, between 10 and 11 AM. Id. 964:7-17. As a result, the National Weather Service issued flood warnings, including warnings covering the Amargosa Valley. Id. 964:18-966:2.

Turning to the January 8, 2016 flood, Mr. Rosenthal testified that during the three-day period that ended on January 6, there were a series of storms that came through the region, and that, because it was another El Niño year, the storms were very energized. Id. 966:23-967:1. He further stated that satellite imagery showed that on January 5 at around 4:30 PM, Ash Meadows was experiencing moderate to heavy precipitation. Id. 967:9-24. Mr. Rosenthal also noted that during this event there were widespread winter storm advisories regarding the severity of the weather in Southern Nevada. Id. 969:2-16.

Finally, Mr. Rosenthal stated that, just before the flooding that occurred on July 3, 2016, very strong thunderstorms struck the area on June 30, July 1, and July 2. Id. 970:2-5. On June 30, powerful thunderstorms moved up from the south, and radar images show rainfall as much as two inches in an hour in the region. Id. 971:8-13. They also show total rainfall of roughly two-and-one-half inches from June 30 to July 1. Id. 971:25-972:7. As with the October 2015 rain event, there were severe thunderstorm warnings and flood advisories some of which involved the area of interest within Nye County. Id. 972:12-973:4.


I. Fifth Amendment Takings Standards

The Fifth Amendment to the United States Constitution provides that "private property" shall not be "taken for public use, without just compensation." U.S. Const. amend. V. To establish entitlement to compensation under the Takings Clause, a plaintiff must first show that it has "a property interest for purposes of the Fifth Amendment." Members of the Peanut Quota Holders Ass'n v. United States, 421 F.3d 1323, 1330 (Fed. Cir. 2005) (citing Conti v. United States, 291 F.3d 1334, 1339 (Fed. Cir. 2002)). Once the plaintiff has established that it has a property interest, as it is undisputed the Ministry has here, it must show that the government's actions "amounted to a compensable taking of that property interest." Am. Pelagic Fishing Co. v. United States, 379 F.3d 1363, 1372 (Fed. Cir. 2004).

It is well established that "government-induced flooding" of property can constitute a compensable "taking" for purposes of the Fifth Amendment. See Ark. Game & Fish Comm'n v. United States, 568 U.S. 23, 32 (2012) (noting that, "where real estate is actually invaded by superinduced additions of water, earth, sand, or other material . . . so as to effectually destroy or impair its usefulness, it is a taking, within the meaning of the Constitution") (quoting Pumpelly v. Green Bay Co., 80 U.S. 166, 181 (1872)). Further, flooding caused by government action need not be permanent in nature to be treated as a taking. Government action that causes regularly recurring flooding, as is alleged here, may "g[i]ve rise to a takings claim no less valid than the claim of an owner whose land was continuously kept under water." Id. (citing United States v. Cress, 243 U.S. 316, 328-29 (1917) (noting that the government may effect a taking via "intermittent but inevitably recurring overflows" no less than by permanently inundating the land by water)); see also Cedar Point Nursery v. Hassid, 141 S.Ct. 2063, 2071 (2021) (observing that "the government [] effects a physical taking when it occupies property . . . by recurring flooding as a result of building a dam").

As with other physical takings, however, to establish a compensable taking based on recurrent flooding, the plaintiff must show by preponderant evidence that the government's actions were the direct and proximate cause of the flooding. St. Bernard Par. Gov't v. United States, 887 F.3d 1354, 1362 (Fed. Cir. 2018); Sanguinetti v. United States, 264 U.S. 146, 149-50 (1924); Loesch v. United States, 645 F.2d 905, 920 (Ct. Cl. 1981). Put another way, the plaintiff must "establish that government action caused the injury to their properties," i.e., "that the invasion was the `direct, natural, or probable result of [that] authorized activity.'" St. Bernard Par. Gov't, 887 F.3d at 1359-60 (quoting Ridge Line Inc. v. United States, 346 F.3d 1346, 1355 (Fed. Cir. 2003)); see also Cary v. United States, 552 F.3d 1373, 1380 (Fed. Cir. 2009) ("For an injury to be a compensable taking, the court must determine that no break in the chain of causation existed between the suspected government authorized action and the injury."). The plaintiff must also "show that in the ordinary course of events, absent government action, [it] would not have suffered the injury." St. Bernard Par. Gov't, 887 F.3d at 1362. In other words, it must show "`what would have occurred' if the government had not acted." Id. (quoting United States v. Archer, 241 U.S. 119, 132 (1916)).

For the reasons set forth below, the Court finds that the Ministry did not prove that the government's construction of the berm and restoration channel caused the flooding that occurred on its property. Nor did it prove that the flooding would not have occurred but for the project. The Court therefore enters judgment for the government as to the Ministry's takings claim.

II. Causation

A. The Ministry's Allegation that the Camp Property Never Flooded Until After the Government Built the Berm and Restoration Channel

A key allegation underlying the Ministry's causation argument is that the northeast portion of its property—i.e., the part of the property where the camp buildings are located— never flooded until after FWS built the berm and restoration channel or at least that there is "no evidence" that it did. Pl.'s Post-Trial Opening Br. ("Pl.'s Post-Trial Br.") at 5, ECF No. 139. To support its assertions, the Ministry relies largely if not entirely on the testimony of Pastor Fuentes and Ronald Matheny, the prior owner. For the reasons set forth below, however, the Court concludes that preponderant evidence shows that the area where the camp is located has flooded in the past, before the channel was built.

Pastor Fuentes testified at trial that, beginning in late 2006, when the Ministry purchased the property, see JX 5, and up until the project was completed in August 2010, the property had never experienced the kind of severe flooding that later occurred in 2010 and 2015, tr. vol. 1, 99:11-17. But as described above, Pastor Fuentes also acknowledged during his RCFC 30(b)(6) deposition that "the first time the property flooded" while the Ministry owned it was in 2007. Ministerio Dep. 96:12-15. The day of and the day before that flooding event it had rained on the upper watershed. Tr. vol. 1, 143:25-144:10. Pastor Fuentes testified that as rainfall flowed down the watershed, id. 144:11-13, muddy water slightly overtopped the culvert and the banks of the irrigation ditches, id. 78:18-79:23, 99:11-17, 141:1-5. And the floodwater dumped mud and silt into a pond on the property that was fed by one of the irrigation ditches, id. 78:23-79:14, requiring that Pastor Fuentes close it, id. 78:24-79:2, and hire a contractor to excavate it, id. 78:8-24.

The irrigation ditch that dumped floodwater in the pond ran through the northeastern corner of the property, where the church camp was located. The Ministry downplays the significance of this incident as an example of flooding that occurred before the project was built. Pastor Fuentes appears to have believed that the incursion of flood waters was unusual and the result of work FWS happened to be performing upstream. Id. 78:19-24. But FWS was not performing work upstream. JS no. 10 ("Construction of the Fairbanks and Soda Springs Restoration Project began in December 2009."). The 2007 incident thus represented a pre-project example of floodwaters entering the church camp area and causing at least some damage after several days of rainfall in the upper watershed.

Mr. Matheny's testimony also does not persuade the Court that all of the flooding which occurred after the restoration project was completed was dramatically different from what had occurred before. Mr. Matheny owned the property for over nine years, beginning in January 1997. PX 269 (quitclaim deed dated January 28, 1997). When he purchased the property, it was "just bare land, sagebrush, and that's it." Tr. vol. 1, 170:16-19. Mr. Matheny cleared the land, planted grass and trees, and kept the grass watered. Id. 170:20-24, 171:11-14, 179:1-10. He testified that during his period of ownership he had never seen anything comparable to the flooding that occurred in 2010. Id. 173:10-174:4, 189:21-190:4.

The Court does not question Mr. Matheny's observation that he had never seen floods on the property that compared with the 2010 flood. As described above, the flooding that occurred on the property in 2010 was the most severe of the four events cited in the complaint. The flooding that Pastor Fuentes witnessed on December 23 followed six days of rainfall whose amounts Mr. Rosenthal testified "dwarfed the normal amounts expected" in the Carson Slough. Tr. vol. 4, 948:7-13. Indeed, the rainfall during this six- to seven-day period was several times the normal amount of rainfall for the entire month of December. Id. 948:13-18.

The Court found unpersuasive Dr. Reely's testimony that the amount of rainfall that fell in the days before the 2010 flood was unexceptional and no different from the rainfall that occurred during Mr. Matheny's period of ownership. Tr. vol. 2, 392:14-18. Dr. Reely considered only rain gauge data and based his opinion on how much rain fell in a twenty-four-hour period. Id. 393:17-22. In a desert area, there are relatively few rain gauges. Mr. Rosenthal, who, unlike Dr. Reely, is an expert in meteorology, was able to address the spatial gaps by using radar, satellite imagery, and other tools. See tr. vol. 4, 934:8-10, 934:13-23, 975:13-976:8.

Moreover, as Mr. Rosenthal testified, by looking only at an artificial twenty-four-hour period, "you miss the significance of what was going on." Id. 974:15-25. Dr. Reely was "missing a big part of the picture of what actually occurred during each of these four flooding events . . . especially for the December 2010 event, because that was a long duration event that stretched six days." Id. 974:16-20.

Further, the precipitation frequency will be different depending on the duration of a rain event. Id. 974:15-975:8; see also tr. vol. 2, 463:15-464:20 (Reely testimony, agreeing on cross-examination that the precipitation frequency will vary depending on the duration examined). Rainfall of an inch in a twenty-four-hour period, for example, may be a common occurrence. But if an inch of rain falls every day for six days or over the course of an hour, then that could equate to a much lower statistical probability of reoccurrence. Tr. vol. 4, 955:18-25. The Court therefore finds Dr. Reely's use of only twenty-four-hour rain gauge data in his analysis a flawed approach.

The Court credits Mr. Rosenthal's opinion that, even if rain gauge data alone is used, the rainfall that preceded the 2010 flood was likely to occur only once every twenty-five years. Id. 988:2-7.23 The fact that Mr. Matheny never saw any flooding comparable to that which occurred in 2010 therefore does not give rise to an inference that the flooding in 2010 was caused by the restoration project.

Further, the Court does not understand Mr. Matheny's testimony to be that the property never experienced any flooding at all during other heavy rainfall events that occurred while he owned the property. To the contrary, Mr. Matheny acknowledged that during heavy rainstorms the property would "puddle up" and become "a sloppy mess." Tr. vol. 1, 173:10-16. His description of the property as "puddle[d] up" and "a sloppy mess" is imprecise. But it is not inconsistent with the photographs and Pastor Fuentes's description of the state of the northeast portion of the property after several of the other less severe flood events.24 Mr. Matheny did not visit the property during or immediately following any flood event besides the one that occurred in 2010 and therefore could not offer a comparison except as to the 2010 flood.

Equally important, the Ministry's claim that before the restoration project the northeast corner of the Ministry's property never flooded is contrary to other evidence both circumstantial and direct. To begin with, the property is located at the bottleneck of the Carson Slough floodplain. JX 3 (FWS map of the Refuge and Carson Slough indicating the location of the property). It is undisputed that other parts of the Ministry's property to the west of the location of the camp have flooded in the past. Tr. vol. 1, 182:8-11 (Matheny testimony). The Carson Slough has flooded multiple times in recent memory, tr. vol. 2, 285:12-16 (Johnson testimony); tr. vol. 3, 610:10-11, 640:3-9 (Andress testimony), including in 2010 when flood water damaged roads and infrastructure and entirely washed out the culvert that allowed the restoration channel to pass under the Patch of Heaven access road, tr. vol. 3, 642:9-15. Further, the majority of the forty-acre parcel the Ministry owns, including the camp itself, lies in a FEMA AO2 flood zone. See DX 47. Because of the flood risk represented by that designation, Nye County ordinances require property owners to elevate any buildings on their property by at least two feet. Tr. vol. 2, 286:13-287:4 (Johnson testimony).

The Ministry responds that neither the fact that the property is in a floodplain nor the fact that parts of the Refuge have flooded with some regularity proves that the location of the Patch of Heaven is prone to flooding. But the physical evidence also reflects that the specific site where the camp was built has flooded in the past. First, just to the northeast of the developed part of the property is a network of braided channels that both experts agreed existed for many years before the 2010 flood (even if that flood may have caused them to deepen), id. 356:15-23 (Reely testimony); tr. vol. 3, 837:15-17 (Thompson testimony), and tracked flood water right at the church camp's buildings, tr. vol. 2, 356:7-14 (Reely testimony); tr. vol. 3, 837:18-23 (Thompson testimony). Their existence reflects a history of flood waters coming from the northeast, moving southwest, and being discharged directly into the northeast corner of the property that became the Patch of Heaven. Tr. vol. 3, 836:22-838:17 (Thompson testimony).

Second, Mr. DesRoberts explained that certain aerial photographs—dated 1948, 1980, 1985, and 1989—depict Plaintiff's property and the surrounding area, and show that floodwater historically traveled through the developed portion of the property. All four photographs show flood channels tracking directly through the northeast corner of the Patch of Heaven where the buildings sit. Tr. vol. 3, 599:9-21, 601:12-602:4 (Defendant's Exhibit 4) (1948 aerial image); id. 602:5-20, 603:3-10 (Joint Exhibit 27) (1980 aerial image); id. 604:18-605:5, 606:7-18 (Joint Exhibit 14) (1985 and 1989 aerial images).

Finally, the Court notes that even if it were persuaded by the Ministry's argument that the property did not experience significant flooding during the fourteen-year period between 1997 (when Mr. Matheny bought it) and 2010, it could not base a finding of causation on such a "post hoc, ergo propter hoc" analysis. See Loesch, 645 F.2d at 914; see also id. at 913 (rejecting plaintiffs' argument that they established causation by showing that "the erosion on their riverbanks was not a problem until after" the government constructed dams that impounded water); Owen v. United States, 20 Cl. Ct. 574, 584 (1990) (observing that post hoc analysis "can easily lead to wrong inferences if there are other elements of causation at work"). Determining the precise cause or causes of the floods requires consideration of hydrology, hydraulics, meteorology, soil conditions, and other technical subjects. The Court turns, therefore, to the competing opinions of the parties' experts.

B. The Expert Testimony

As described in detail above, the parties each presented expert testimony regarding the extent to which the restoration project caused the flooding of the Patch of Heaven that occurred in 2010, 2015, and 2016. For the reasons set forth below, the Court finds the testimony of the government's expert, Dr. Thompson, more persuasive than that of the Ministry's expert, Dr. Reely.

To briefly recapitulate, Dr. Reely opined that the berm FWS installed to redirect the flow of spring water into the new restoration channel obstructed the path that floodwater coming from the upper watershed had taken in years past. Tr. vol. 2, 344:15-19. Specifically, Dr. Reely explained, until the berm and restoration channel were built, floodwaters would travel down the natural trace and exit to the west of the Ministry's property via certain off-ramp channels, never reaching the Patch of Heaven. Id. 343:12-344:14. Dr. Reely opined that the berm blocked floodwater from going down the natural trace, and instead kicked it to the east where it would then flow south and bend back west in an attempt to follow the trace of the restoration channel. Id. 344:15-19. Dr. Reely further testified that, because the restoration channel had insufficient capacity to handle such large volumes of water, it would pour out of the western bank of the channel and migrate toward the church camp, which he estimated was at a slightly lower elevation by an order of approximately six to ten feet. Id. 344:19-23, 345:12-21, 351:8-17. Flood flow would enter the braided channel network to the northeast of the Patch of Heaven— where it would be directed straight into the church camp. Id. 344:23-345:3, 356:4-14.

As described above, Dr. Reely based his opinion regarding the effect of the restoration project on his observations of the site during three visits as well as his review of topographical maps, aerial photographs, rainfall data, and other reports. Dr. Reely performed no measurements or calculations of the capacity of the restoration channel, never measured the size of the berm, and did not attempt to determine the amount of flood water that came from upstream nor the amount of water allegedly redirected by the berm during any of the events in question. He performed no quantitative assessment of how water moves through the Carson Slough. Nor did he perform any modeling to attempt to verify or substantiate the opinion he reached based on his visual observations. Id. 317:7-318:9. And he provided no response to Dr. Thompson's theory that the water that flooded the Patch of Heaven in 2010, 2015, and beyond likely originated on the eastern side of the Slough.

In the Court's view, Dr. Reely's opinion lacks a sound foundation. Eyeballing the property and examining aerial photographs and maps strikes the Court as an analytic approach that collides with Dr. Reely's own observation that the Carson Slough is "a very complex bit of terrain, and primarily because it's very—topographically very subtle," such that "very subtle changes in this topography will change the way and have changed the way water flows across this portion of the desert, of the watershed." Id. 330:5-10.

Dr. Thompson took a more methodical approach that is consistent with what one would expect from an expert in the field attempting to pinpoint the extent to which the restoration project caused the Ministry's property to flood. He testified that a "hydrologist would typically want to do some kind of computational effort to support the formation of [his] opinions." Tr. vol. 3, 794:19-21. In addition, he stated, "in this particular case and in many others, [a hydrologist] might use . . . a hydrologic model for flow estimation, and then use the output from the hydrologic model to operate a hydraulic model to determine how flow moves through the system in question." Id. 794:21-795:1. "And both of those efforts," Dr. Thompson testified, "are typically supported with some hand calculations or calculations using a spreadsheet or a similar computational tool." Id. 795:1-3. Dr. Reely did not use any of these methods.

Dr. Thompson has extensive experience not only as a practicing hydrologist but also in academia, having taught hydrology at the undergraduate and graduate level for fourteen years, led research projects, participated in numerous research studies, and published some fifty conference proceedings. Id. 788:9-789:4, 789:19-23. The Court credits his testimony regarding the methodology a hydrologist would ordinarily use to tackle an issue of causation like the one presented by this case.

Dr. Reely, as noted, did not support his opinion with any measurements or quantifiable data. Tr. vol. 2, 314:23-315:12. Further, he acknowledged that he could have used models to justify the conclusions that he reached (and also that he uses models routinely in his work). Id. 314:17-22, 318:10-21. But he asserted that it was unnecessary for him to build a model here because he believed the physical evidence was so clear, id. 383:7-22, an opinion obviously not shared by Dr. Thompson.

Because his opinion contains no supporting objective and verifiable data, Dr. Reely did not persuade the Court that the restoration project caused the flooding about which the Ministry complains. Moreover, Dr. Reely actually expressed no opinion—whether based on quantitative data or otherwise—regarding whether the Ministry's property would have flooded during the events in question even absent the berm and restoration channel.

To be clear, the Court does not find that the berm did not divert into the church camp any flood water that otherwise would have gone down the natural trace.25 But Dr. Reely does not explain for any of the flood events how much of the floodwater that would have gone down the natural trace were it not for the berm was instead diverted east, winding up in the braided channels and then flooding the Ministry's property. Dr. Reely agreed that in 2010 the floodwater overtopped the berm and followed its historic path down the natural trace. Tr. vol. 2, 442:24-443:5. And Pastor Fuentes agreed that during the 2015 flood event, he saw "lots of water" coming down the natural trace. Tr. vol. 1, 153:18-154:4. Therefore, the Court finds that during both flood events at least some portion of the flood waters that came from the north continued to follow the natural trace notwithstanding the presence of the berm.26

Equally important, Dr. Reely's opinion does not account for other possible causes of the flooding. Dr. Thompson credibly testified, based on field observation and topographical data, and as supported by his models, that the floodwaters that flowed into the Patch of Heaven likely had flowed downstream on the eastern part of the Slough and therefore would never have flowed into the natural trace anyway. Tr. vol. 3, 819:6-12. Dr. Thompson opined that the waters came down from the eastern part of the upper watershed and flowed to the southwest in the form of sheet flooding. Id. 818:9-13, 843:7-21. According to Dr. Thompson, those waters would briefly enter the restoration channel and then exit out of it on the western bank, entering the braided channels that feed water directly into the camp area. Id. 818:14-17, 837:18-23. Aerial photographs support Dr. Thompson's opinion that sheet flooding occurred in this manner, see PXs 108, 151; tr. vol. 3, 854:18-855:5 (describing Plaintiff's Exhibit 108); tr. vol. 3, 810:25-811:2, 814:2-7 (describing Plaintiff's Exhibit 151), as does the testimony of Refuge Manager Corey Lee, who stated that he witnessed this phenomenon in the wake of the March 2019 flood, tr. vol. 3, 752:4-9. Dr. Reely never addressed the validity of this possible alternative cause of flooding on the Patch of Heaven, and his theory does not challenge Dr. Thompson's conclusions on the impact of water approaching the Patch of Heaven from the northeastern side of the Carson Slough.

As described in some detail above, Dr. Thompson also performed measurements and calculations and used hydraulic and hydrologic modeling to support his conclusions. Using HEC-HMS modeling software, Dr. Thompson estimated that approximately 1,600 to 3,200 cfs of floodwater passed through the watershed during the 2010, 2015, and 2016 flood events, and that the capacity of the restoration channel was only between 35 to 75 cfs. Id. 814:19-815:6, 809:19-25, 810:5-6.27 Given the very limited capacity of the restoration channel when compared to the volume of water coming from the upper watershed, Dr. Thompson concluded, the restoration channel would have been overwhelmed by the floodwaters and would have at most "a minimal impact . . . on flows that are approaching plaintiff's property." Id. 817:22-818:1.

Having concluded that the restoration channel could not have had a significant impact on the movement of flood flows heading to the Ministry's property, Dr. Thompson used a HEC-RAS model to understand what would have occurred during the flood events in question if there were no restoration project. The purpose of the model was to determine the source of the flooding, assuming that it was not the berm or restoration channel. Based on that modeling, Dr. Thompson concluded that the floodwaters that spread down the eastern side of the Carson Slough would have caused flooding of the property even if the restoration channel were not there any time peak discharge reached a volume of 1,600 cfs or greater. Id. 817:9-13, 819:14-820:6.28

HEC-HMS and HEC-RAS models are commonly used tools in the field of hydrology. See tr. vol. 2, 308:8-13 (Reely testimony, stating that the HEC-RAS model is the most common one he uses); tr. vol. 3, 794:19-795:3 (Thompson testimony, stating that a hydrologist would "typically" make "some kind of computational effort," such as using hydrologic and hydraulic modeling). Dr. Reely acknowledged that he has used them "very frequently" in his work. Tr. vol. 2, 308:20-24. Models are, however, imperfect, and their accuracy depends in large part on the assumptions and data inputs a hydrologist uses to build them. Id. 384:10-385:6. It is therefore significant that Dr. Reely never attempted to impeach any of the specific data sets or assumptions Dr. Thompson used to build his HEC-HMS model. See id. 334:7-14 (approving the topographic data Dr. Thompson used to develop his hydraulic models).

Dr. Reely testified that it is important to field validate hydraulic and hydrologic models. Id. 333:10-13. Dr. Thompson did conduct such validation as to the HEC-HMS model's conclusions of the amount of flood runoff during flood events. Tr. vol. 3, 817:3-6. The Court therefore credits Dr. Thompson's opinions, based on his HEC-HMS models, regarding the capacity of the restoration channel and the fact that the restoration channel could not have had a significant impact on flood flows coming down the watershed.

Dr. Reely also expressed some criticisms of Dr. Thompson's HEC-RAS model. For example, Dr. Reely testified that 1D HEC-RAS models, like the one Dr. Thompson employed, have a limited capacity to predict lateral movement of water. Tr. vol. 2, 385:9-15; see also tr. vol. 3, 869:1-19 (Thompson testimony, acknowledging that 1D HEC-RAS models are limited in their ability to show lateral movement of water). Dr. Reely also testified that 1D HEC-RAS models are not good at simulating changes in the momentum of water that may occur where there are tight turns, as with water deflected off of the berm. Tr. vol. 2, 385:16-25. But Dr. Thompson inserted "lateral weirs" that permit the model to move water laterally from one side of the weir to the other. Tr. vol. 3, 812:10-14, 845:20-24. Dr. Reely's testimony does not address the extent to which the use of lateral weirs alleviated these concerns.

Dr. Reely also criticized Dr. Thompson for failing to field validate his HEC-RAS model, tr. vol. 2, 333:10-16, as he did his HEC-HMS model, tr. vol. 3, 817:3-6. The HEC-RAS model Dr. Thompson developed, however, was just one tool he used to help him understand the watershed, and while field validation of the model would have bolstered the opinion he reached based on the HEC-RAS model, the Court found the rest of his testimony regarding causation persuasive, including his conclusions based on visual observations and the HEC-HMS model.

In its brief and on cross-examination, the Ministry also criticizes Dr. Thompson's HEC-RAS model on the grounds that it yielded results that were in some respects inconsistent with evidence of what actually occurred during the 2010 and 2015 flood events. See Pl.'s Post-Trial Br. at 37-41; tr. vol. 3, 891:18-895:23, 878:12-24, 882:11-22, 883:19-884:6. The Court is not sure what to make of this concern because during his testimony Dr. Reely did not address any of the specific alleged inconsistencies. As noted, Dr. Thompson concluded, based on the results of the HEC-HMS modeling, that—given its limited capacity—the restoration channel could not have had any impact on whether or not flooding occurred on the Ministry's property. Dr. Thompson therefore built his HEC-RAS model without the berm or restoration channel in it. It is therefore unclear to the Court what significance to assign to the fact that, for example, his HEC-RAS model of the 2010 flood did not show floodwater in the area where the restoration channel meets the Patch of Heaven access road and where the 2010 flood blew out the culvert under the road. See tr. vol. 3, 878:15-879:25, 895:5-23. For all the Court knows, it may be that it did not show floodwater in that precise location because the model did not include the restoration channel, which meant there was no culvert to be washed away.

A second criticism of Dr. Thompson's model is that it showed water in the natural trace during the October 2015 flood, but that aerial photographs taken shortly after peak flooding did not. Id. 889:4-891:5. Dr. Thompson explained that because these photographs were taken after peak flood flows had passed through, waters in the natural trace would have already discharged out the Carson Slough. Id. 899:9-900:2.

In any event, and most importantly, even if Dr. Thompson's models were less than perfect, it is of little consequence because it was not the government's burden to show causation. It was the Ministry's burden to prove that the project caused its property to flood and also that the floods would not have occurred were it not for the berm and restoration channel. Dr. Reely's testimony did not persuade the Court on either point because it was not supported by quantifiable data, did not address in any way possible alternative causes of the flooding, and presented an inaccurate picture of the extent to which the rainfall that accompanied the 2010 and 2015 floods was exceptional. Further, as noted, Dr. Reely never offered an opinion on whether, even absent the berm and restoration channel, the property still would have flooded as a result of waters coming from the eastern side of the upper watershed.

In short, the Ministry failed to prove that the restoration project caused the flooding of its property. To the contrary, the weight of the evidence shows that the property would have flooded even if the berm and restoration channel had never been built. The Ministry has therefore failed to meet its burden to show causation.


On the basis of the foregoing, the Clerk is directed to enter judgment for the government.



1. Because the Court concludes that the Ministry has not met its burden of proving causation, it does not address the other factors the Supreme Court has identified as relevant to determine whether recurrent flooding caused by government action rises to the level of a taking. See Ark. Game & Fish Comm'n, 568 U.S. at 38-39 (explaining that the court should consider: (1) the duration or frequency of the flooding; (2) whether the flooding is intended or is the foreseeable result of the government action; (3) character of the land; (4) reasonable investment-backed expectations; and (5) severity of the interference with property interests). In addition, the Court's factual findings will be limited to those that are relevant to the issue of causation.
2. This section sets forth the Court's principal findings of fact pursuant to Rule 52(a) of the Rules of the Court of Federal Claims ("RCFC"). Other findings of fact and rulings on questions of mixed fact and law are set out in the Discussion section.
3. The outer boundaries of the Slough are high points of the area (e.g., mountain ridges). Tr. vol. 3, 726:1-7. The lighter blue lines are "drainages," which usually are dry, but when flooding occurs, water feeds through them. Id. 727:16-20, 797:2-6.
4. "A floodplain is an area around a river or a stream or a wash that is subjected to flow whenever a flood passes through [a] particular breach of a spring." Tr. vol. 3, 796:18-21. It is "generally a relatively wide area," and once "flow gets out of the channel itself," it moves "in a similar fashion to sheet flow." Id. 796:21-24. Sheet flow is flooding over land that spreads out and slows down as it moves downhill over a patch of land. Id. 729:16-730:7.
5. An AO2 designation means that during a "100-year flood" event, the property could be inundated by up to two feet of water. Tr. vol. 2, 269:18-21, 271:14-18.
6. The former owner of the Ministry's property, Ronald Matheny, was aware that the property was located in a FEMA-designated flood zone when he sold it to the Ministry. Tr. vol. 1, 180:19-181:2. He testified that he did not disclose that information to the Ministry at the time of sale because he believed (incorrectly) that he had "built the town where the flood zone area isn't." Id. 181:3-9.
7. On one occasion, Mr. Matheny testified, the water in the flooded area of the Refuge was deep enough to come up to the headlights of his "Chevrolet 1 ton" "[w]ork truck." Tr. vol. 1, 174:16-21.
8. The Court understands Mr. Matheny to be referring to a complex of flood channels that had been naturally eroded out from repeated flooding (that is, the naturally-created discharge channels west of the property). Tr. vol. 1, 182:7-16; tr. vol. 2, 440:1-11, 444:19-445:3 (Reely testimony); tr. vol. 3, 829:20-830:2, 830:17-19 (Thompson testimony).
9. FWS was required by the Refuge Improvement Act of 1997 to create the CCP. JX 7 at 15; see also tr. vol. 3, 612:8-11. Planning for it formally began in the fall of 2001. JX 9 at 2.
10. Throughout these proceedings the parties have engaged in a somewhat semantic debate about whether to call this structure a "dam" as the Ministry would prefer, or a "berm" as the government urges. While the structure may technically constitute a "dam" in that it does block the flow of spring water to at least some degree, in the Court's view it is more precise to call it a berm, which Merriam Webster defines as including "a mound or wall of earth or sand." Merriam-Webster's Collegiate Dictionary 115 (11th ed. 2020).
11. Pastor Fuentes testified that he believed that the flooding that resulted in the damage to his pond was caused by work that FWS was performing upstream, which he believed had caused "something" to "break loose from the top" and "rush the water down to the property." Tr. vol. 1, 144:11-19; 145:3-146:2. Mr. DesRoberts testified, however, that there was no work being done by FWS at the time that the flood occurred in 2007. Tr. vol. 3, 639:18-640:2. The Court credits Mr. DesRoberts' testimony and finds that the flooding that occurred in 2007 was not the result of actions taken by FWS.
12. There is no evidence in the record to show that a dam located north of the property broke or that the flooding that occurred was caused by such a break.
13. Pastor Fuentes testified that from his perspective, for each event, he could see that the flood water came at the property from the direction of the restoration channel to the northeast. Tr. vol. 1, 100:11-15 (December 2010 flood); id. 106:22-107:1 (October 2015 flood); id. 112:20-24 (January 2016 flood); id. 113:12-15 (July 2016 flood); id. 116:19-22 (April 2020 flood).
14. Plaintiff's Exhibit 77 is an aerial photograph that depicts the "natural trace." See PX 77. The right side of the photograph is north. The dark colored north-south channel that runs parallel and to the west of the straight manmade channel (irrigation ditch) is the natural trace, which is usually dry but carries floodwater during flood events. Tr. vol. 2, 340:4-8. This channel is at the very edge of the Carson Slough. Various "off-ramp" channels flow out of the natural trace to the west ("off-ramp channels"). Id. 340:11-12, 342:18-20. The off-ramp channels are heavily vegetated, and the vegetation that surrounds the channels is visible in the photograph. See PX 77; tr. vol. 2, 342:9, 353:12-16, 354:2-5.
15. The braided channels are also visible in Plaintiff's Exhibit 77. They are in the bottom left quadrant of the image and are located directly to the southwest of the restoration channel. See PX 77.
16. A headcut is an erosional feature of some streams with an abrupt vertical drop. Regional Stream Stewardship and Recovery Handbook, Commonly Used Terms, (last visited September 8, 2021).
17. The three gauges were: (1) Desert Rock Airport rain gauge located in the upper northeastern quadrant of the watershed, tr. vol. 2, 396:5-17; (2) the Desert Wildlife Refuge rain gauge located just south of Plaintiff's property in the watershed, id. 396:19-397:4; and (3) the Amargosa Farms rain gauge located northwest of Plaintiff's property just beyond the western boundary of the watershed, id. 397:6-12.
18. Dr. Thompson testified that because the HEC-HMS model provides only estimates of flood flow, he checked those estimates by taking "indirect measurements." Tr. vol. 3, 807:23-808:3. Specifically, he examined high-water marks observed in an area upstream of Plaintiff's property and used them along with the geometry of the channel to "make an estimate of flow rate." Id. 799:4-11, 815:15-20. The high-water marks allowed Dr. Thompson to measure the width and depth of flow at that location. Id. 815:25-816:6. Then, using "computations based on open channel hydraulics," he estimated the flow that had caused the high-water mark. Id. 816:6-9. Dr. Thompson's indirect measurements led him to conclude that the upper and lower bounds of the peak discharge in recent flood events were 4,500 cfs and 1,400 cfs, respectively. Id. 815:25-816:9, 816:17-20. These estimates gave Dr. Thompson confidence that the outputs of the HEC-HMS model were "reasonable." Id. 816:23-817:6; see also id. 833:18-835:25 (testifying as to how high-water marks are determined and used to estimate depth and area of flow).
19. The restoration channel's capacity decreased, he explained, because the amount of vegetation in and along the restoration channel increased in the ensuing years, which decreased the capacity of the channel. Tr. vol. 3, 810:1-7.
20. Dr. Thompson's 1D HEC-RAS model combined four different 1D HEC-RAS models into one. Tr. vol. 3, 843:24-844:1. Two of the models, Dr. Thompson explained—specifically his models of the east and west sides of the Carson Slough floodplain—were particularly important and allowed him to determine approximately how much flow would be on the east side of the floodplain. Id. 844:1-7. The other two models, which focused on areas on or near Plaintiff's property, allowed him to estimate the depth of flow on Plaintiff's property and the discharge passing through the western discharge channels. Id. 844:7-11.
21. As a 1D HEC-RAS model is limited in its ability to show lateral movement of water, tr. vol. 3, 869:17-19, Dr. Thompson inserted a "lateral weir" in the model at the topographic high between the east and west sides of the floodplain to account for side-to-side movement, id. 845:18-846:8; see also id. 869:6-19 (agreeing that a 1D HEC-RAS is "limited in its ability to . . . make computations with energy in the lateral direction").
22. The Amargosa Farms gauge is one of two or three stations upstream and within 15 miles of the Ministry's property. Tr. vol. 4, 982:9-20.
23. This is according to the Desert Wildlife Refuge rain gauge in the watershed that Dr. Reely also relied on. See tr. vol. 4, 976:21-977:1 (Rosenthal testimony); tr. vol. 2, 396:19-397:4 (Reely testimony).
24. The floods in 2016, 2019, and 2020 did not flood the area where the buildings are located and were primarily concentrated in areas away from the church camp's buildings. Tr. vol. 1, 112:5-19, 160:6-8 (Fuentes testimony as to January 2016 flood); id. 160:14-24 (as to July 2016 flood); id. 115:2-14 (as to March 2019 flood); id. 116:1-18 (as to April 2020 flood); see also PXs 298, 300, 301, 306 (photographs of July 2016 event, tr. vol. 1, 113:2-114:3) (depicting areas of land away from the buildings inundated with water); PXs 596 598, 608, 611, 620 (photographs of March 2019 event, tr. vol. 1, 115:7-14) (depicting scattered puddles); PXs 703, 705 (photographs of March 2020 flood, tr. vol. 1, 116:1-117:20) (depicting flooding in the northeast field only).
25. It bears noting, however, that physical evidence Dr. Reely relies on to show that the berm caused floodwaters to be diverted into the church camp—i.e., the alligator cracking, tr. vol. 2, 349:4-350:25, and the erosion of the natural channels on the property, id. 378:5-379:4—is also consistent with Dr. Thompson's theory, discussed below, that floodwaters on the church camp did not come originally from the west but instead originated on the eastern side of the Slough and passed over the restoration channel in their southwesterly traverse down to the Ministry's property.
26. Plaintiff introduced aerial photographs into evidence that were taken after the October 2015 flood which show floodwater in the restoration channel and north of the berm, but not in the natural trace of the Carson Slough. PXs 150, 151. Dr. Reely testified that these photographs verify that floodwater never overtopped the berm in 2015 but were rather redirected by the berm to the east—otherwise the photographs would have shown floodwater in the natural trace. Tr. vol. 2, 373:24-374:4, 375:3-11. Dr. Thompson agreed that these photographs revealed no floodwater south of the berm in the natural trace of the Carson Slough. Tr. vol. 3, 889:1-2, 890:19-891:17. But Dr. Reely explained that the photographs were taken after the peak flood discharge had already passed through the area. Tr. vol. 2, 380:20-381:4. Dr. Thompson similarly stated that because peak flood flows had passed, floodwater that overtopped the berm and went into the natural trace would have already discharged down the trace and to the southwest of the Carson Slough. Tr. vol. 3, 889:1-2, 890:19-891:5, 899:9-25. He further testified that the flood water in those photographs in and around the restoration channel was residual flood flow that had not yet discharged out the Carson Slough. Id. 899:25-900:2.
27. Dr. Reely agreed that Dr. Thompson's calculations regarding the capacity of the restoration channel were reasonable and estimated himself that the capacity is "probably somewhere in the 10 to 35 cfs range." Tr. vol. 2, 441:4-16. Dr. Reely did, however, criticize Dr. Thompson's exclusion of the Amargosa Farms rain gauge in his analysis (a rain gauge that is close to the Patch of Heaven but outside of the watershed) and stated that, had Dr. Thompson included that rain gauge, "he may or may not have decided to use a slightly lower rainfall precipitation amount than he used in his analysis." Id. 331:10-17. But Dr. Thompson explained that it was his "professional preference" to use rain gauges in his hydrologic modeling that are within the boundaries of the watershed. Tr. vol. 3, 826:15-20. Additionally, Dr. Reely did not himself calculate the amount of runoff in the Carson Slough using rain gauge data, so did not and could not offer to the Court an alternative conclusion on the amount of flood flow. Tr. vol. 2, 450:13-451:3.
28. Dr. Thompson's HEC-HMS model showed that the floods of 2010 and 2016 would have exceeded 1,600 cfs of flow and the flood flow of October 2015 would have been 1,570 cfs. Tr. vol. 3, 814:19-815:6.


1000 Characters Remaining reserves the right to edit or remove comments but is under no obligation to do so, or to explain individual moderation decisions.

User Comments

Listed below are the cases that are cited in this Featured Case. Click the citation to see the full text of the cited case. Citations are also linked in the body of the Featured Case.

Cited Cases

  • No Cases Found

Listed below are those cases in which this Featured Case is cited. Click on the case name to see the full text of the citing case.

Citing Cases