The catastrophic flooding of July 4, 2025 in the Texas hill country has left at least 130 dead. Many news outlets (predictably) have invoked climate change as a contributing cause, for example CNN, NPR, ABC News, The Texas Tribune, et al. As others have documented, there were more than adequate flash flood watches and warnings from the National Weather Service, and there was no shortage of either NWS staff or weather data.
It has long been known that this region is considered “Flash Flood Alley“, where the local topography and little soil covering the rock underneath leads to large amounts of runoff into the Guadeloupe and other nearby rivers in the event of heavy thunderstorms.
Clearly, one weather event is not evidence of climate change. We need to examine long-term weather statistics to evaluate claims that severe weather, of any type, is getting worse. This is especially true of heavy rainfall events, which are notoriously sporadic, with long-term statistics which are not well-behaved.
Flash floods require more than just heavy rainfall; they also require (1) rain to accumulate over a very short period of time, (2) the storms need to stay over the same area, and (3) the geography and hydrology features (little soil, sloping terrain) need to rapidly funnel most of the water into streams and rivers. The Flash Flood Alley region of Texas can deal with, say, 5 inches of rain if it falls steadily over 2 days. But if it falls in only 6 hours, flash flooding is much more likely. As is the case with tornadoes, a catastrophic flash flood requires specific ingredients that seldom occur all at the same time and location. There is a large element of randomness involved.
The Catastrophic Floods of 1978
While the 1987 flood (pictured above) was also severe, it has been nearly 50 years since a flood of similar magnitude to the July 4th flood occurred, and that was in 1978. This San Antonio Express-News article is a fascinating read regarding that event. Like this year’s disaster, the cause was a dissipating tropical storm (Amelia in 1978, T.S. Barry this year). That flood killed 25 people in the Hill Country. What I find especially interesting is that the flood uprooted ancient cypress trees, up to 6 ft. in diameter, estimated to be several hundred years old.
133 Years of Heavy Rainfall Events in the Texas Hill Country
John Christy (UAH) provided me some heavy rainfall statistics for several stations in the Kerrville, TX area. For years John has been going through old daily weather summary forms that were never digitized, extending observational records back into the late 1800s. This is tedious and time-consuming work, but necessary if one has any hope of examining long-term trends in heavy rainfall events.
From those data, here are the last 133 years of the heaviest 2-day rainfall events in each year at Kerrville, TX (since 1893).
Clearly, there has been no long-term change in heavy rain events at Kerrville, Texas in spite of increasing atmospheric CO2 concentrations (also shown in the above plot). Even the EPA shows maps for the 50-year period 1965-2015 that indicate slight downward trends in river flooding magnitude and frequency over most of Texas, as well as the rest of the U.S.
From the Kerrville plot above you might wonder, why did the 2025 flood have so little rainfall at Kerrville? This is because most of the rain occurred upstream of Kerrville. Flash floods can occur at locations where no rain has fallen, as the water flows downstream from where the heavy rainfall occurred. Because of this effect, not all of the heavy rainfall peaks shown in the above plot produced serious flooding, and some of the documented flood events at Kerrville had only modest rainfall amounts.
Instead, the rainfall statistics at Kerrville should be viewed as evidence to address the question, are heavy rainfall events in Texas Hill Country getting worse? At Kerrville, at least, the answer is “no”.
But that’s just one station. John Christy also provided me data for 3 other nearby stations: Boerne, Fredericksburg, and Hondo Texas (not shown). For the same period (1893-2025) the trend lines for those stations are all essentially flat to slightly downward.
Climate Change, Heavy Rainfall, and Flooding: What Is The “Scientific Consensus”?
As we document in our Department of Energy report released yesterday entitled, A Critical Review of Impacts of Greenhouse Gas Emissions on the U.S. Climate, when one looks at rainfall statistics across the U.S. extending back to the mid- to late-1800s, there is little evidence for anything that might be considered related to human-caused climate change.
And don’t take just our word for it.
For flooding, the most recent IPCC report (AR6) said there is “low confidence for observed changes in the magnitude or frequency of floods at the global scale.”
For the U.S., the 4th National Climate Assessment stated that “trends in extreme high values of streamflow are mixed” with both increases and decreases, and there is no “robust evidence” that any trends are attributable to human influences.
So, for the usual suspects trashing our report to the media (Michael Mann, Andy Dessler, and Zeke Hausfather), maybe they should look at what we actually wrote, and the “consensus” sources we relied upon.
The public has been misled on climate science, and we are trying to set the record straight.
This piece originally appeared at DrRoySpencer.com and has been republished here with permission.
