At 7:23 p.m. on Monday, April 13, an EF-2 tornado formed a few miles west of Ottawa, Kansas. It tracked east through town with peak winds of 125 miles per hour over a path of 7.3 miles. The same supercell produced a second EF-2 near Hillsdale and an EF-1 near Mound City. Three people were hurt in Ottawa. Nobody died.

By Thursday morning, NBC News had a story explaining what happened. The headline: "'We are missing data': NWS weather balloon changes scrutinized as tornados hit Midwest." The piece drew a causal chain from Trump administration cuts, through staggered weather-balloon launches, to a forecast that did not anticipate a tornado threat for the Kansas City area.

That story has a problem. The Storm Prediction Center's outlooks and forecaster discussions for April 13 are public, time-stamped, and addressed to the Kansas forecast offices by name. Read them, and the NBC piece falls apart.

NBC's argument is specific. Many Great Plains forecasting offices shifted their 7 a.m. weather-balloon launches to noon because of staffing issues. A Wisconsin mesonet meteorologist, Chris Vagasky, told NBC, "We are missing data at the normal times." A retired National Severe Storms Laboratory director, Alan Gerard, thought the schedule change followed from continuing staffing shortages. The piece cited the Trump administration's buyouts and firings of probationary workers as the driver.

The piece also ran the NWS response. A spokesperson for the weather service, Erica Grow Cei, said the changing cadence of weather balloons has not affected forecasts. NBC quoted her, then wrote around her for the rest of the article.

A working journalist would have opened the Storm Prediction Center archive before deciding which source to believe. This one didn't.

Three outlooks NBC could have read

At 8:00 a.m. Central time on Monday, SPC issued the day's 1300 UTC Day 1 Convective Outlook. The headline category placed the main Enhanced Risk over southern Minnesota and parts of Wisconsin. For the southern Plains, the forecasters wrote that whether sustained storms would even develop was "a key question." Eastern Kansas did not appear in any tornado probability tier. The outlook explicitly said a Slight Risk for the southern Plains might be drawn in later updates.

The 11:36 a.m. update added Slight Risk to the southern Plains and kept hedging. The forecasters wrote that confidence had "incrementally increased" but that "specifics regarding the extent of convective initiation are still uncertain." Eastern Kansas still sat outside the meaningful tornado probability area.

The 2:56 p.m. update extended severe probabilities into eastern Kansas. The discussion said guidance and observational trends showed "an increased likelihood of a storm or two this afternoon." Four and a half hours before Ottawa, SPC was already updating the risk area.

Eight mesoscale discussions and the one NBC missed

Between 2:38 p.m. and 5:09 p.m., SPC issued eight Mesoscale Discussions. These are the real-time forecaster notes that speak directly to local Weather Forecast Offices. Seven of them aimed at other states. MD #401 at 2:59 p.m. pinned a 95 percent Tornado Watch probability on Iowa and Minnesota and said a couple of strong tornadoes were possible there. SPC was not looking the wrong way for lack of interest. It was looking where the environment looked dangerous.

MD #404 went out at 3:53 p.m. It covered eastern Kansas. It addressed the Springfield, Kansas City, Topeka, and Wichita forecast offices by their four-letter IDs. Watch probability: 20 percent. The summary read, "Watch issuance appears unlikely at this time owing to uncertainty regarding the occurrence of convective initiation."

Inside MD #404, the forecasters cited a specific balloon by name: the 1 p.m. Central sounding from Lamont, Oklahoma. The text called it a "special sounding," meaning an off-cycle launch ordered to capture extra data on a severe weather day. Lamont is a Department of Energy research facility. It launches balloons four times a day on its own schedule, regardless of any NWS staffing decision.

The NWS upper-air network supports extra balloon releases for critical weather "as needed," on top of its standard twice-daily schedule. The closest NWS balloon sites to eastern Kansas are at Topeka, Dodge City, and Norman. On April 13, SPC was reading sounding data from across the region. That data included a launch NBC's story cannot explain away. The balloon data NBC said was missing sat in the text of the Kansas discussion.

MD #404 went out at 3:53 p.m. The Ottawa tornado touched down at 7:23 p.m. Three and a half hours of real-time forecaster attention, with the afternoon sounding data already cited in the text, ended in a 20 percent watch probability and a supercell that rolled through anyway. This was a forecasting miss. It was also the kind of miss mesoscale meteorologists have been writing about for decades. The ingredients were present. The trigger location and strength outran any tool in current use.

That is the problem of mesoscale science. Sometimes you're wrong. It is not a balloon schedule.

The buried counter-frame in NBC's own piece

Brad Temeyer, a meteorologist at the Kansas City NWS office, told NBC something the piece did not feature. He said the atmosphere was primed for a possible non-event. "There was a pretty strong possibility there would be no showers or thunderstorms at all," he told NBC. "It was a low-probability event of it occurring, but given that it did occur, it had high impact."

NBC also wrote, "Weather service offices did issue warnings when tornadoes were imminent, however." The local forecasting offices in Topeka and Kansas City sent out several warnings later. That is the warning chain doing its job. NBC's own reporting suggests the headline point is wrong. The piece ignored it.

Had NBC done real reporting, the story would have read differently. It could have said, straightforwardly, that this was a forecasting miss that happened in Kansas. It could have argued, as a separate piece, that cuts to the NWS are unwise given the public service the agency performs on alerts like the ones that got Ottawa residents indoors. Both of those would be defensible pieces to write. Pinning the cuts story to this specific tornado is not. This was a forecasting miss. It likely would have been a forecasting miss without any cuts. It was a forecast-science failure ahead of the event. The models and the humans reading them could not pin down where the cap would break. The warning chain then corrected in the last minutes when radar finally showed what was coming.

We have seen this movie before

Nine months ago, after the July 4, 2025 Guadalupe River flooding killed at least 135 people, including 27 at Camp Mystic, major outlets ran the same reflex. Staffing cuts and NWS leadership turnover became the explanatory frame. The ongoing wrongful-death litigation has told a different story. NWS issued its first flash flood warning for Kerr County at 1:14 a.m. Camp Mystic did not begin evacuations until nearly two hours later. The camp's director testified under oath three days ago that he did not see the early warnings. Texas Rangers are now assisting a state investigation of the camp. The framing has quietly collapsed. The sequel is the Kansas piece.

NBC picked a villain and worked backwards. What happened on April 13 is messier than that, and it is worse for the press than for the agency. This was a forecasting miss. The morning outlook missed. The 3:53 p.m. mesoscale discussion for the Kansas offices said watch unlikely. The supercell that formed over the eastern Kansas dryline ran bigger than the atmosphere told anyone at 8 a.m. or 4 p.m. that it would.

What saved Ottawa in the last minutes was the warning chain doing exactly what it is supposed to do. Tornado warnings went out. NWS Topeka's local office got residents indoors. Three injuries and no deaths is the signature of a warning system that worked under stress.

SPC missed on April 13. Two forecasters signed the Kansas discussion — Chalmers and Hart — and their names are on the archive along with the miss. The agency will post-mortem the case. The verification record will note the bust. The next forecast will carry whatever April 13 taught the system. That is the ordinary loop of an institution accountable to its own archive.

The question is whether NBC will do the same. The Camp Mystic template did not chasten the newsrooms that ran it; Kansas is the sequel. Meteorologists have a hard job, but they're performing it under pressure. Journalists have a comparatively easy job, and aren't doing it, and the weather service archives prove the point.