The Next Step in Inundation Mapping

Several years ago, Harris County Flood Control District (HCFCD) made a giant leap forward with real-time inundation mapping. You could log onto the Harris County Flood Warning System, click on “inundation map,” zoom into your neighborhood, and see how close floodwater was getting in “near real time.”

It was a radical advance over earlier system. But now we need another radical advance – predictive inundation mapping.

No one, as far as I can see, offers such a service. The National Weather Service predicts how high water will get at gages, but not between them where people live.

How Close Will the Floodwater Get…At My Location?

Buyers today have much more information at their fingertips than decades ago, but no one puts it all together.

One of the most useful websites during a flood is the National Weather Service Advanced Hydrologic Prediction Service. There, you can see how fast water is rising, where it will peak, when it will fall, and a hydrologic history of the event. (See layout below.)

*From June 2 at 1:41PM for West Fork San Jacinto at US59*

This is a huge advance over everything before it. But unfortunately, it only predicts flood heights at gages. Most people don’t know how their distance from a gage, elevation changes, and local topography will affect the peak at their home or business. Especially if they are new to a neighborhood, like the woman below.

Debris line on side of house five feet up — *Kingwood resident who moved here in January from a desert, then promptly flooded in May.*

Real-Time Flood Mapping

So, what alternatives do people have? Harris County’s Flood Warning System shows you how far water is from your home in near real time.

The mapping feature is a great enhancement compared to the NWS site in some ways. It shows the extent of flooding between gages.

But unfortunately, the Flood Warning System won’t tell you how close water will get at its peak.

So when you realize you need to get out, you may not have enough time to make thorough preparations.

*Screen capture from Harris County Flood Warning System. It shows current flooding, but not peak flooding.*

Trying to Integrate the Best of Both Approaches

Those with knowledge of prior floods may be able to translate gage readings upstream to their location downstream.

But almost 10% of the American population moves every year. So, five years after a hurricane hit an area, half of the people in a neighborhood would likely have little knowledge or memory of it.

More than real-time inundation mapping, we need predictive inundation mapping.

Legal Concerns over Bad Information

One official I talked to for this article feared such a system. What if the prediction were wrong? What if it implied someone would be safe and then they died while evacuating? Imagine the legal liability!

As a result, officials often err on the side of caution. They may issue orders to evacuate in much wider areas than necessary.

We saw that during the May flood this year. The county issued evacuation maps that showed the extent of flooding during Harvey. As a result, people may be less inclined to evacuate for the next storm, even if it is warranted.

In extreme cases, such over-reaction has resulted in harm. Those who have lived in the Houston area for 20 years or more may remember the disastrous mass evacuation during Hurricane Rita in 2005.

One hundred thirteen people died from Rita in Texas, 107 of which were associated with the evacuation of the Houston metropolitan area.

Need for Better Tools

A former HCFCD employee told me that HCFCD has been working on a predictive inundation mapping system for years. But, says the source, it’s incredibly complex and requires validation.

Validation involves confirming predictions against real-world storms. However, giant, widespread floods don’t just happen every year. And every storm is unique, as we saw in yesterday’s post about the early May storm. Some areas may get feet of rain, while others get inches.

You just can’t generalize. Every storm is unique. And referring to historical analog storms may fail for several reasons:

Massive developments, such as Colony Ridge, may have changed the hydrology of the watershed. Colony Ridge has grown 50% larger than Manhattan in just ten years.
Rainfall intensity may vary in storms, even if rainfall totals don’t.
Unequal rainfall distribution means that flood peaks may combine in different ways at different times.

Potential Ways to Handle Uncertainty

To help handle those contingencies, predictive inundation maps could have buffer zones showing degrees of uncertainty. For instance, “If you’re in Zone A, there’s a 100% chance you will flood. In Zone B, 50%. In Zone C, 25%, etc.”

Plus, the legislature could craft legal protections for agencies making such predictions based on the best available information and science.

We need a public debate. Perhaps an even better approach will emerge from the dialog.

That would be far preferable, in my opinion, to causing immense anxiety among millions.

Posted by Bob Rehak on 6/2/24

2469 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.