How Infrastructure Bottlenecks Constrain Peak Flows, Cause Local Flooding

4/3/26 – Engineers design infrastructure to handle certain amounts of rainfall. But:

Rainfall probability estimates increase over time.
Urban growth increases the amount of impervious cover, creating faster, higher peak flows.
Upstream developments encroach on floodplains
Fragmented governance makes it difficult to know how to plan.

Infrastructure bottlenecks reveal themselves during extreme events. This is a universal phenomenon. Bridges, culverts, and channel constrictions control flood elevations behind them.

You see this everywhere around the world. It’s a well studied phenomenon. Let’s look at two areas.

Appalachia

Appalachia has many undersized road crossings. Culverts and bridges have a finite conveyance capacity. Exceeding that capacity creates a partial dam that causes backwater to rise upstream and overtop roadways.

Openings then catch debris floating downstream, further reducing the conveyance of the river. Forested basins supply large wood during storms. Inlet plugging is common. Debris blockage can reduce culvert capacity by 30–80%.

This is especially acute in Appalachia because of steep basins susceptible to flash flooding and the fact that roads usually follow streams. They can have many crossings per mile.

Older culverts and bridges may have been designed in previous eras when engineers anticipated smaller storms. You see scour at outlets where water churns, trying to get under the bridge. You also see jetting on the downstream side. Both lead to channel instability.

In June 2016, West Virginia experienced one of the worst floods in its history. After the event, U.S. Geological Survey, Federal Highway Administration and West Virginia Department of Transportation documented:

Hundreds of culvert and small bridge failures
Widespread road overtopping and washouts
Extensive debris blockage at inlets
Excessive erosion and sedimentation downstream.

*NOAA photo from 2016 West Virginia Flood*

Houston-Area Examples

In the Houston area, we see the same hydraulic constraint issues.

FM1010 Washout

Rapid development of Colony Ridge led to a washout of FM1010 during Harvey.

FM1010 at Plum Grove — *FM1010 culverts could not handle the drainage from new upstream development larger than Manhattan.*

I-69 Bridge Over West Fork

TxDot had to replace part of the I-69 bridge over the San Jacinto West Fork while residents endure massive traffic jams because of scouring under the supports for the southbound lanes.

I-69 repairs — *Replacement took almost a year*.

UPRR Bridge over West Fork

UP Bridge — *Union Pacific’s Bridge over the San Jacinto West Fork was destroyed by Harvey.*

*Trees washing downstream formed a dam that raised the flood level.*

Tree Lane Bridge

Multiple floods backed water up behind the Tree Lane Bridge over Bens Branch in Kingwood. They raised water levels upstream as jetting scoured bridge supports on the downstream side. The bridge has been repaired several times since Harvey.

Given upstream development, the width was insufficient. Engineers ultimately had to widen the opening of the bridge to let high peak flows pass through.

Rustic Elms Bridge

Up to 600 homes flooded twice in Kingwood’s Elm Grove Village in 2019. The bridge below was one of the problems. The culverts convey less water than the open span design of the West Lake Houston Parkway Bridge over Taylor Gully in the background.

Rustling Elms Bridge over Taylor Gully in May 2019 — *Rustling Elms Bridge underwater as school bus tries to cross it in May 2019.*

West Fork Mouth Bar

Eroded sediment washed downstream during Harvey and dropped out of suspension where water slowed as it met Lake Houston. Thousands of homes and businesses flooded behind this sediment dam, which reached more than 10 feet about normal water level.

FM 1960 Bridge

Post-Harvey analyses revealed a significant constriction in the San Jacinto near the headwaters of Lake Houston. The openings in the FM 1960 causeway across the lake are half the width of upstream and downstream bridges.

John Blount, Harris County Engineer at the time of Hurricane Harvey said he noticed a difference in the water surface elevations on the upstream and downstream sides of the bridge. Downstream was lower by 1-2 feet.

Graph from Post-Harvey Analysis by Charles Jones.

Culverts under Kingwood Drive

During Harvey, one hundred and ten homes in Kings Forest flooded behind these culverts half clogged with sediment. The City of Houston cleaned them out in 2025.

Lake Houston Dam

The single biggest blockage in the Lake Houston Area is the Lake Houston Dam. While the dam has a spillway to handle high-water events, it does not have gates with a sufficient release capacity to lower water levels immediately before storms. The City of Houston is designing additional gates that will allow greater coordination with pre-releases from Lake Conroe. And the San Jacinto River Authority is studying ways to coordinate pre-releases.

site of proposed gates for Lake Houston on east side of dam — *Earthen portion of Lake Houston Dam where new gates will be added.*

I-45 Bridge at Cypress Creek

Harvey’s floodwaters in Cypress Creek were so strong that they literally picked up parts of the southbound I-45 lanes and shifted the bridge. The constriction caused by the bridge backed water up into hundreds of homes.

*I-45 southbound feeder over Cypress Creek showing cattywumpus bridge panels*

I could go on. But you get the idea. Bridges and culverts restrict flow in large events, creating dangerous backwater. Those bridges and culverts may have been adequate early in their lives, but upstream growth rendered them inadequate.

For more information about other factors that contribute to flooding, see the Lessons page of this website.

Posted by Bob Rehak on 4/3/26

3139 Days since 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.