Weather apps may be lying to you: Study reveals hidden heatwaves in cities

DURHAM, N.C. — Picture this: it's a scorching summer day, and you're checking the weather app on your phone. It says it's 85°F, but as you step outside, it feels more like you've walked into an oven. Well, a new study finds that there's a good chance you're not imagining it. If you live in a less affluent urban area, it's likely the actual temperature really is higher than what your app is telling you.

Researchers from Duke University have discovered a concerning issue when it comes to the weather: the temperatures reported in poorer urban neighborhoods might be severely understated. The study published in the journal Environmental Science & Technology Letters explains that this discrepancy isn't just a matter of comfort — it's a crucial factor in public health and urban planning. Simply put, cities could be wildly underestimating how they prepare for heatwaves due to this faulty app data.

What is an Urban Heat Island?

You might have heard the term “urban heat island” before, but what does it really mean? Basically, it's an area within a city that's noticeably warmer than its surroundings. These heat islands are typically found in more impoverished neighborhoods and are characterized by:

1. More buildings and concrete surfaces
2. Less vegetation and green spaces
3. Higher population density

All these factors combine to create a perfect storm of heat retention. Concrete and asphalt absorb and radiate heat, while the lack of trees and plants means there's less natural cooling. The result? A neighborhood that can be several degrees hotter than wealthier areas just a few blocks away.

So, if we know about urban heat islands, why aren't they accurately represented in our weather reports? The answer lies in how we collect weather data.

Most of us rely on popular weather websites and apps for our daily forecasts. One of the biggest players in this field is Weather Underground, which taps into a vast network of over 250,000 personal weather stations set up by everyday citizens in their backyards.

This crowdsourced approach to weather data seems like a great idea at first glance. After all, more data points should lead to more accurate predictions, right? Unfortunately, there's a catch.

“The poorest areas of a city also tend to have the fewest number of weather stations to pull data from, so if we’re going to rely on their data, we either need to add more ground sensors or try to adjust for the missing data,” says Zach Calhoun, a civil and environmental engineering PhD student at Duke, in a statement. “While having accurate temperature data might be important for residents going on their day-to-day business, it’s especially important for policymakers relying on the data to make well-informed decisions.”

The Connection Between Wealth and Weather

Personal weather stations, while not exorbitantly expensive, still cost several hundred dollars. It's not surprising that they're more commonly found in wealthier neighborhoods. This creates a data gap in precisely the areas that are likely to be the hottest — the urban heat islands in less affluent communities.

“Some of the sensors may be a bit off, but if you put a lot of them together, the aggregate data is pretty reliable,” says Mike Bergin, the Sternberg Family Professor of Civil & Environmental Engineering at Duke. “More data is good data. And less data is bad data.”

Methodology & Results

Faced with this challenge, the Duke researchers came up with an innovative solution. Instead of physically installing more weather stations (which would be ideal but impractical in the short term), they developed a statistical method to correct for the lack of data in underserved areas.

Here's how they did it:

1. They analyzed four years of Weather Underground data for the entire state of North Carolina.
2. They mapped out the locations of personal weather stations and compared them to each area's average annual income.
3. Using this information, they created a statistical model to estimate temperatures in areas with fewer or no weather stations.

To test their method, the team compared their corrected data for Durham, North Carolina, with highly accurate temperature readings taken during a special heat mapping campaign. The results? Their estimates were much closer to reality, especially in the hottest parts of the city.

This isn't just an academic exercise. Accurate temperature data is crucial for public health and urban planning.

“Extreme heat also leads to poor air quality, and the resulting impacts on respiratory and cardiovascular health should be monitored for the benefit of all,” adds Marily Black, a public health scientist for Underwriter’s Laboratory (UL) Chemical Insights Research Institute and co-author of the paper. “This is especially true for the vulnerable in urban areas such as children and those who are economically disadvantaged.”

Knowing the true extent of urban heat islands can help city planners and policymakers make informed decisions about where to plant trees, install cooling centers, or focus heat mitigation efforts. It's also a matter of environmental justice, ensuring that all residents — regardless of their neighborhood's wealth — have access to accurate information about the conditions they're living in.

“Our work shows that you can make corrections and get better estimates of urban heat islands through these types of methods,” Carlson concludes. “It also highlights the need for additional weather stations so we’re aware of just how much hotter the summer is for the poorest members of our community.”

So, the next time you check the weather, remember that the temperature you see might not tell the whole story.