11/30/24 – On November 26, 2024, the Court of Appeals for the First District of Texas dismissed homeowners’ claims in the long-running SJRA Takings Case. The case arose from Hurricane Harvey flood damage.

Downstream homeowners alleged that San Jacinto River Authority (SJRA) releases from Lake Conroe significantly exacerbated their flooding. The resulting damage, they claimed, amounted to an illegal taking of their property by the government.

The Texas Constitution provides that “no person’s property shall be taken, damaged, or destroyed for or applied to public use without adequate compensation.” (Texas Constitution, Article I, Par. 17(a)).

Sadly, the alleged “facts” cited in the Appellate Court’s ruling apparently went unchallenged. And some of those so-called facts may mislead.

Speed of Floodwaters a Major Deciding Factor

The appellate court judges based much of their ruling on an argument related to the speed of floodwaters.

SJRA had produced an expert report by Mark E. Forest. The appellate judges said on page 24 of their ruling that “…the timing [emphasis added] of the water releases and flooding supported the conclusion that the River Authority’s water releases did not cause the homeowner’s flooding.”

The judges continued, “Forest explained that the water from Lake Conroe would take about 30 hours to travel the 38 miles downstream to reach the Humble and Kingwood areas, where most of the homeowners’ properties were located.”

Then, on the same page, comes this pivotal sentence. The judges say,

“About 30 hours after the River Authority first released water from Lake Conroe, the Humble and Kingwood areas had already experienced ‘major flooding,’ as the river levels had already risen from 41 feet above sea level to 62 feet.”

Page 24, SJRA v. Medina

Then the coup de grâce. “This major flooding occurred ‘without any contribution from Lake Conroe since those contributions had not yet arrived’ Forest explained.”

The appellate judges in the SJRA Takings Case evidently bought that argument. At the bottom of Page 24, they said, “Therefore, the River Authority produced evidence that its water releases from Lake Conroe did not cause or exacerbate the flooding of the homeowners’ properties.”

In their conclusion on page 36, the judges also state, “The homeowners’ evidence does not raise a fact issue to refute the River Authority’s evidence. The homeowners have not met their burden to provide evidence showing there is a material fact issue as to the causation of their inverse condemnation claim.”

Basis for SJRA Speed Estimate Unclear

I have asked SJRA repeatedly how they arrived at 30 hours and never gotten an answer. That made me suspicious. The SJRA claims floodwaters move only1.27 MPH (38 miles in 30 hours). An average person can walk 3-4 miles per hour!

Plus, I’ve measured (with my drone) logs floating downriver in lesser floods at 5-6 MPH.

The speed of floodwaters determine their arrival time downstream. Claiming a 30 hour travel time that easily could have been as brief as five or six hours could cause people to focus on the wrong part of the flooding bell curve. And it appears that may be what happened in this case.

Judges Cite Flooding at Less-Than-50-Year Level

I’m not sure what data Forest cited; the judges don’t specify. But let’s assume it came from the gage at the West Fork and US59.

At that location, 62 feet is less than a 50-year flood. Most structures are built at least two feet above the 100-year floodplain. The 100-year flood level there is 64.8 feet – almost three feet higher.

Importance of Speed in Determining Proximate Cause

If you assume a higher rate of speed and correlate that to when floodwaters reached the 100-year level where most homes begin to flood, it’s easier to see a possible connection between the Lake Conroe release and structural flooding downstream.

Floodwaters traveling 5-6 MPH would reach US59 in five to six hours, not 30 hours.

According to the Harris County Flood Warning System, the West Fork reached a 100-year-flood level at US59 at 8:42 PM on 8/28/2017.

And according to an SJRA affidavit supplied during the original trial:

• 30 hours earlier, SJRA was releasing 10,946 CFS.
• 6 hours earlier, SJRA released 78,885 CFS.

So, between the 30 hours claimed by the SJRA and the six hours indicated by drone measurement, the SJRA increased its release rate by almost 8X.

If the drone measurements are accurate and representative…

SJRA releases increased dramatically shortly before structures near US59 flooded.

This suggests a different conclusion. SJRA releases may have been a far more proximate cause of downstream flooding than SJRA-supplied data implied.

Would homes and businesses have flooded eventually anyway? Certainly, that’s true for some. But it might not be true for others.

No one would expect a home to flood in Kingwood when you release 11,000 CFS from Lake Conroe.

But how about an additional 79,000 CFS when homes were already on the verge of flooding? That’s a very different story – especially considering that at the peak, Lake Conroe releases comprised one third of the total water coming down the West Fork. Many homes at the periphery of the flooding might have stayed dry had the timing and volume of releases been different.

Too Late Now

The plaintiffs’ expert consultant in the SJRA Takings Case did not conduct his own hydrological modeling (Page 26). That’s unfortunate.

Even if the plaintiffs in this case succeed in getting the Texas Supreme Court to hear an appeal, additional evidence would likely be inadmissible.

In general, appellate courts do not consider new evidence. Their primary role is to review the record from lower courts to determine whether legal errors were made that significantly affected the outcome of the case. Appellate courts focus on issues of law rather than fact.

Posted by Bob Rehak on 11/29/24

2650 Days since Hurricane Harvey

The thoughts expressed in this post represent opinions on matters of public concern and safety. They are protected by the First Amendment of the US Constitution and the Anti-SLAPP Statute of the Great State of Texas.