Which Past Was Hers, 42nd Day of Summer, 525 M.E. (Betelgeuse): My heart breaks reading about the 78-and-counting deaths in Texas from flooding on the Guadalupe River. I can't imagine the anguish of the parents who learn their children were swept away from a summer camp by the floodwaters. It's beyond comprehension and well beyond tragic.
But in these times we live in, everything becomes political and everything is about finger-pointing and blaming others. The left is blaming climate change and government budget cuts, the right is blaming the National Weather Service. As a hydrogeologist, I can't let this incident pass without my own two cents (spoiler alert: I don't blame either side).
Texas Gov. Greg Abbott declared today a “day of prayer” for the victims, both surviving and deceased. If prayer helps you get through this, then fine, go ahead and pray, but don't expect it to change anything.
The director of the Texas Division of Emergency Management has faulted the NWS for not predicting “the amount of rain we saw,” even though alerts were issued beforehand and as it became clear the region was facing an emergency. A flash-flood watch was issued Thursday afternoon that noted Kerr County, where much of the flooding began early Friday morning, was a particularly vulnerable area, along with more urgent flash-flood emergency alerts in the overnight hours as the disaster unfolded. The Emergency Management director should be asking why his agency didn't do more to heed the NWS' warnings and advisories, and work proactively to mitigate the situation by evacuating people in the flood zone while it was still possible.
Also, the NWS was one of several federal agencies targeted by the Stable Genius' DOGE boys, which had laid off nearly 600 employees, around the same amount of staffers the service lost in the 15 previous years.
But the staffing shortage wasn't the issue, and Monday-morning quarterbacking Emergency Management's decisions isn't the solution. It's not the cause, but the underlaying issue here is climate change, as little as some people want to hear that. To be clear, the flood wasn't "caused" by climate change. This morning, I heard Chris Christie say on This Week that no one will ever convince him that Hurricane Sandy was caused by climate change. Governor, I agree. If someone tells you it was caused by climate change, they don't know what they're talking about and you can safely ignore them.
Climate change doesn't cause rain, hurricanes, floods, wildfires, or droughts. They're caused by meteorological and atmospheric processes, but climate change does increase the severity and frequency of storms, hurricanes, floods, wildfires, and droughts. And it also impedes scientists' ability to forecast extreme weather events.
Flood forecasting is basically a statistical exercise. The USGS has been monitoring the Guadalupe River since October 1941, and has developed statistics on the frequency and intensity of flood events based on the 84-year record. Based on the monitoring history, scientists can calculate the average flow rate (53 ft³/sec) and the maximum flow rate (599 ft³/sec, recorded in 2002). As a 599 ft³/sec flood occurred once in 84 years, one can state that the chance for a flood of that magnitude happening in any given year is 1-in-84, or a 1.2% chance. FYI, the 100-year flood is the flood that has a 1% chance of happening in any given year.
Today, the river is flowing at 872 ft³/sec, shattering the 23-year-old record of 599 ft³/sec. To give you an idea of the volume, today's flow would fill an Olympic-size swimming pool every 90 seconds.
The available, online data shows the river has flooded eight times between October 2007 and June 2025, meaning the gage level was above 10 feet, and two of those floods were "moderate," with the gage over 18 feet. A flood is considered "major" if the gage exceeds 22 feet, and today the river level topped 37.5 feet, another record.
From the monitoring data, one can calculate a "flood frequency," i.e., the river floods x number of times per year, and extreme floods occur once every x years. Of course, the frequency alone can't predict when the flood will actually occur, but coupled with rainfall and meteorological data, there is sufficient information to issue flood watches and flood warnings when appropriate.
Climate change, however, throws a monkey wrench into the calculations. With conditions changing so rapidly, the past is no longer a useful indicator of the future, or even of the present. The x number of years extreme floods occur is increasing and increasing rapidly, and if we only consider, say, the last 10 years to calculate flood frequencies, well, that's not enough data to be meaningful. Statistically, the data-point population isn't significant enough for the calculated frequencies to be meaningful. Nowadays, we have so-called "100-year floods" occurring annually or every few years, with each flood more severe than the one before.
Computer models can simulate the "new normal" and can provide useful statistics of the expected flood frequencies and flood levels in our new climate, but due to climate skepticism and denial, the results aren't trusted and the effort is being rapidly defunded by the government. So the old tools don't work, the new tools aren't trusted, half the country's underwater, and the other half is on fire.
But Gov. Abbott's gonna pray the flood away, so I guess we'll be alright.