Science1 distinct publisher3 min readPublished
A CU Denver team matched hourly boardings at 7,283 RTD stops to satellite readings of heat, cold, rain, smoke and ozone, and the clearest response came from the conditions riders feel on their skin. Free-fare months muted it, but only in summer.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
Divide 44.3 million rider experiences by 7,283 stops and the average stop in this dataset carried about 6,083 boardings over the study window [3][20]. Across sixteen months of hourly records [2][21], that is roughly 380 a month, a dozen or so a day at the average stop [23]. This is not a flaw in the analysis, which pools across the whole system, but it fixes the unit of observation: a count at a pole, not a person. The study did not track individual riders or their reasons for boarding [10].
The exposure side is the design's strongest move. Rather than air temperature, the team used the Universal Thermal Climate Index, which folds in humidity, wind and sunlight [11], alongside satellite data on heat and cold stress, rain, smoke and ozone from NASA's Research Opportunities in Space and Earth Science program [4]. If the question is how people respond to conditions they can feel, felt temperature is the right variable to regress on.
Hourly resolution earns its keep on rain. The drop is concentrated in the first several hours, which the authors read as trips postponed or rerouted rather than abandoned [7]. A daily total would have shown a smaller, blurrier effect and told you something wrong about what riders actually did.
The fare result is where I would slow down. Zero Fare for Better Air ran in August 2022 and in July and August 2023 [12], three months out of sixteen, every one of them summer [22]. So the comparison behind "weather matters less when the bus is free" [13] is summer with fares against summer without them, and the largest effect in the paper, cold, which begins within hours and holds all day [6], never once falls inside a free-fare month. The authors state that limit themselves [14]. A calendar period also carries along everything else that happened to be true of those months.
Heat runs in two directions at once: lower ridership over the course of a very hot day, more boardings in the hottest hours [8], with the whole heat relationship shifting depending on whether a stop has a shelter [15]. The researchers offer one explanation, that some people who would otherwise walk or bike take an air-conditioned bus instead [10]. Plausible, and unresolved, because shelters are sited by individual municipalities rather than placed where need is greatest [16]. Sheltered and unsheltered stops therefore differ in more than shelters, and any of those differences could carry the effect.
The weak smoke and ozone signal [5] is the finding most likely to be misread. It says ridership barely moved, which is not the same as saying the air was tolerable, or that riders knew a warning had been issued. As Elizabeth Dzwonczyk, who worked on the project as a CU Denver doctoral student, put it, the lived experience is that you take the bus on a bad air pollution day because your life responsibilities still exist [18]. Revealed preference only reveals choices somebody was free to make.
Priyanka deSouza, the lead author and an assistant professor of urban and regional planning at CU Denver [17], and her colleagues end with a list: real-time rider alerts, service adjustments in bad weather, fare strategies, and shaded, thermally protective stops [19]. That list is testable in a way this study was not. Put shelters up at some stops before others, and the heat question stops being an association.
Ranked by verification strength, evidence, and original report placement.
New research led by the University of Colorado Denver was published in Environmental Research as "Weathering the ride: Associations of heat, smoke, precipitation, and ozone with bus ridership in Colorado."
The study analyzed hourly bus ridership across the Denver metropolitan area from June 2022 through September 2023.
Researchers examined nearly 44.3 million rider experiences from 7,283 Regional Transportation District (RTD) bus stops.
Data on heat and cold stress, rain, wildfire smoke and ozone came from NASA Research Opportunities in Space and Earth Science satellites.
Denver-area bus riders changed their commuting patterns when faced with cold weather or rain, but not as much with wildfire smoke or ozone warnings.
Cold stood out as the weather condition with the strongest and most consistent connection to lower ridership; the decline began within hours and continued throughout the day.
Distinct publishers with included, body-backed reporting in this cluster.
phys.org
1 article · September 2, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
science
A ban is not a retirement: BDE-209 keeps concentrating in Europe's sediment and sludge1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Big dataset, no numbers shown
The base is real: a peer-reviewed paper in Environmental Research, 44.3 million boardings, satellite inputs, and a thermal comfort index instead of a thermometer. What never arrives is a number attached to a finding. "Strongest and most consistent" comes without a coefficient, an interval or a percentage, and it comes through the university's own announcement, so the ranking that gives this story its headline is the one thing a reader cannot check.
The only experiment came first
The single intervention that actually happened here is RTD's, and it happened before the research: fares came off in August 2022 and again in July and August 2023, which is why the fare question could be asked at all. Alerts, weather-triggered service changes and shaded stops remain a list, shelters are still handed out municipality by municipality, and nothing in this reporting shows an agency responding to the paper.
Understated, and candid about why
Unusually, the announcement argues against itself: it volunteers that no rider was tracked individually, that the walk-or-bike explanation is a guess, and that three summer months of free fares cannot be set beside winter cold. Put that restraint next to a 44-million-boarding dataset with no magnitudes published and the work reads smaller than it is. The one place the framing leans forward is the "when the bus is free, weather matters less" subhead, which asks three summer months to carry a year-round claim.
A university making its case
This starts in a communications office, and it shows in shape more than in substance: CU Denver gets the credit, three of its faculty are quoted, and the closing list of alerts, fare strategies and thermally protective shelters doubles neatly as a funding argument. No product, no vendor and no money turns on publication, which keeps the pressure modest rather than absent.
One institutional source, honestly hedged
Moderate, and capped by arithmetic: one publisher, one upstream author, no independent reading of the paper, no word from the agency whose stops and fares are the subject. What keeps the figure from sinking is that the account hedges itself in the right places and its internal numbers hold together.