USGS Report on Peak Streamflows During Harvey Significantly Revises Flood Probabilities

Hurricane Harvey’s widespread 8-day rainfall, which started on August 25, 2017, exceeded 60 inches in some locations. That’s about 15 inches more than average annual amounts of rainfall for eastern Texas and the Texas coast. The area affected was also much larger than previous events.

New High Water Marks and Record Streamflows

USGS field crews collected 2,123 high-water marks in 22 counties in southeast Texas and three parishes across southwest Louisiana.

Record streamflows were measured at 40 USGS streamgages in Texas that have been in operation at least 15 years. At two streamgage locations, scientists determined that the percent chance for flooding of this magnitude to happen in any given year was 0.2 percent. This probability is also referred to as a 500-year flood. Thirty other USGS streamgages experienced flooding at levels with a 1 percent chance of occurring each year, also known as a 100-year flood.

Check out the “event viewer” noted in the report, especially if you are interested in high water marks in your neighborhood.

How Data Will Be Used

The USGS conducts research on the physical and statistical characteristics of flooding, estimating the probability of flooding at locations around the United States.

The purpose of the study was to check the probability of future occurrences and map the extent of flooding in Texas.

These records will assist officials in updating building codes, planning evacuation routes, creating floodplain management ordinances, providing environmental assessments and planning other community efforts to become more flood-resilient. FEMA will also use this information to revise their Flood Insurance Rate Maps. These maps help identify areas most likely to experience flooding in any given year.

Gages Closest to Lake Houston

The section on the San Jacinto Watershed starts on page 33. The maps for the San Jacinto watershed appear on pages 35 and 36. Use the maps to see the new high water marks in the area and to find the USGS gages nearest you. For most people in the upper Lake Houston Area, it will be one of these gages:

• 08068090 – Grand Parkway and West Fork near Porter
• 08069500 – West Fork and I-69 near Humble/Kingwood
• 08070500 – Caney Creek near Splendora
• 08069000 – Cypress Creek near Westfield
• 08068500 – Spring Creek near Spring
• 08071000 – Peach Creek near Splendora
• 08070200 – East Fork near New Caney
• 08071280 – Luce Bayou above Lake Houston near Huffman

After you locate the gages nearest you, cross reference the numbers of those gages with data at the front of the report. It helps to use the search function in Adobe Acrobat because much of the information is in tables with very small type.

Examples of What You Will Find

Here’s an example of what you can find. For the gage nearest many of the sand mines on the West Fork (08068090), peak streamflow was estimated at 131,000 cubic feet per second. That was the highest of the 33 peaks previously observed at that location (from Table 3 on Page 9).

Now here’s the big news: From the same table, we can see that the Annual Exceedance Probability (AEP) is 2.4.  That’s the likelihood of occurrence of a flood of given size or larger occurring in any one year, expressed as a percentage.

AEP is often expressed as the reciprocal of ARI (Average Recurrence Interval). For instance, A 10-year flood has a 10 percent probability of occurring in any given year, a 50-year event a 2% probabaility, a 100-year event a 1% probability, and a 500-year event a .2% probability.

In this case, a 2.4% AEP would have a likely recurrence interval of 42 years, given the new realities of upstream development, any changes in climate, and pocket calculators with more computing power. This means that the West Fork Gage ((08068090) at the Grand Parkway DID NOT even experience a 100-year flood! Yes, we can expect to see worse in the future.

That’s a far cry from the 1,000-year flood that some talked about earlier and raises real public policy questions about locating sand mines in floodways.

Despite the fact that Harvey was the largest rainfall event in recorded U.S. history, USGS now predicts that it would take even bigger floods to reach the reconfigured 100-year, 200-year and 500-year recurrence intervals: 196,000, 263,000 and 374,000 cfs respectively for West Fork Gage at Grand Parkway (Gage #08068090 from Table 5, Page 14). So the new 500-year flood would have almost triple the volume of Harvey.

Humble Gage Data Missing From Report

Unfortunately, the Humble Gage at I-69 does not show up in the tables even though it is on the map and the cover of the report. This is likely in part due to the fact that the gage stopped reporting during the event due to the excessive streamflow

They may also have not reported the exceedance probability due to the shorter recent record.

For all the other gages, the Annual Exceedance Probabilities translate to new recurrence intervals ranging from 35 to 250 years. The gages at the low end of that range tend to be in the fastest developing neighborhoods.

Implications of New Findings

The report will stimulate public policy debate about development near rivers and the most effective methods of flood mitigation.

After reading this, I believe more than ever that we need more detention, dredging and gates (DDG). We need all three to help us handle the volumes of floodwater that USGS expects at more frequent intervals. Prayer, while advisable, is a less certain option in my mind than including DDG in the flood bond and passing it.

BTW, there was some confusion Tuesday night at the flood bond meeting. A small number of flood control employees incorrectly told residents that dredging would not be possible under the bond. It will be according to Matt Zeve, whom I contacted today.

Posted July 12, 2018 by Bob Rehak

317 Days since Hurricane Harvey