1985 Upper San Jacinto Flood Control Study Prophetic, But Largely Unheeded
This morning, I came across a 1985 study by Wayne Smith and Associates for the Texas Water Development Board and the San Jacinto River Authority. It’s called the San Jacinto Upper Watershed Drainage Improvement and Flood Control Planning Study.
Recommendations of specific projects aside, the principal recommendations are as valid today as they were then. Had only someone acted on them.
Make sure you at least read Chapter 5: Conclusions and Chapter 6: Examination and Recommendation of Basic Design Criteria for Watershed. Together, they total just five pages.
Purpose of Upper San Jacinto Study
The Upper San Jacinto study had four main goals:
- Develop a comprehensive stormwater drainage plan
- Recommend specific improvements
- Evaluate/compare alternatives
- Provide drainage authorities with information necessary to control flooding.
Problems of Rapid Development in Flat Areas
The study begins with a discussion of the problems of rapid development in flat areas. The Upper San Jacinto Watershed covers 1200 square miles. It includes all of Montgomery County and parts of Walker, Grimes, Waller, San Jacinto, and Liberty Counties. For the purposes of this study, the Harris/Montgomery County line formed the southernmost boundary.
Seven major streams comprise the watershed: the West Fork, Lake Creek, Spring Creek, East Fork, Caney Creek, Peach Creek, Luce Bayou and Tarkington Bayou.
The topography changes from rolling hills in the north and west to flat coastal plains in the south and east. The lack of slope in the southern and eastern regions seriously affects the ability of streams to drain stormwater.
The authors warned that as development would move northward, hydraulic “improvements” would alter natural stream patterns by increasing flow velocities and reducing ponding.
Even before urban development, they said, channels in the Upper San Jacinto Watershed did not have adequate capacity to transport runoff from large storms.
In 1985, at the time of the report, less than 5% of the land area in the watershed was developed. The Woodlands was relatively new and still building out. The report warned that because of development, increases in impervious cover “will require a more efficient drainage system to collect and transport runoff.”
The report lauded the type of development in The Woodlands, where, “discharges are no higher today than they were years ago in the undeveloped stages.” However, the report also cautioned that “…with most of the current development in the southern and eastern extremities of Montgomery County, watershed flooding problems may be greatly enhanced by urbanization.”
The report even prophesied ever greater amounts of subsidence moving north with urbanization.
The chapter which discussed planning said, “Right of way and reservoir land acquisition should occur while the land is open and available.” Sadly, with the exception of Lake Conroe, which had already been built, none of that happened.
Benefit/Cost Ratios of Regional Detention in Undeveloped Areas
The last advice sounds so simple, one wonders why no one acted on it. However, as I read through the economic analyses of alternatives (reservoirs, channel improvements, etc.), the reason became blindingly clear.
So few people lived in undeveloped areas in the Upper San Jacinto Watershed in 1985 that the annual flood damages are minuscule. For instance, there were only 39 structures in four Lake Creek floodplain areas that the authors examined. Annual damages totaled only $9,600. That made the Benefit/Cost Ratios (BCRs) for the various mitigation alternatives that they developed come out to less than .001 in some cases and .09 at most. Benefits equal costs at 1.0. So FEMA usually demands BCRs exceed 1.
But compare the cost of a reservoir then and now. In 1985, the authors estimated the total cost of a Walnut Creek reservoir (a tributary to Spring Creek) to be only $41,000,000. Today, the cost would be $132 million – more than 3X. But it would take many more homes out of the floodplain. So the BCR today could be 1.04 making the project doable (see page 44)…although much more expensive and much to late to help those who flooded recently.
It’s instructive to compare the project costs in the 1985 plan to those in the San Jacinto River Basin Master Drainage Study released last December. Reliance on the BCR in this case seems to dis-incentivize future planning and cost reduction. There’s a major opportunity for improvement.
To get around this problem, the Harris County Flood Control District started its Frontier Program. The program buys up land for regional detention ponds (those that serve multiple developments), and then resells detention capacity back to developers for future use. Because regional detention is usually more efficient than developers building individual detention ponds on their own, it can actually lower developers’ costs while protecting the public and conserving money long term.
Most High-Level Recommendations Still Valid
Page 43 of the 1985 report makes six high-level recommendations (apart from specific projects) that are as valid today as they were then.
- Create a central agency to control, monitor, remedy and finance flood control for the entire watershed.
- Control development within the 100-year floodplain and prohibit it in the floodway with laws and regulations.
- Establish minimum building slab elevations in flood-prone areas.
- Limit fill in the floodplain.
- Develop procedures to follow when allowing floodplain development, i.e., not obstructing 100-year floods.
- Develop specific criteria, procedures and requirements for downstream impact analysis to compare Development A with Development B, and to analyze their combined effects.
Regular readers of this site have heard many of these recommendations before. The surprise, if there is one, is that we haven’t adopted them all already or that we haven’t adopted them consistently. Even where recommendations have been adopted, they are enforced inconsistently.
For future reference, the 1985 report can also be found on the reports page under the SJRA tab.
Posted by Bob Rehak on 9/19/2021
1482 Days since Hurricane Harvey and 731 since Imelda