Heat, Pollution, and Skyscrapers Make Cities Have More Thunderstorms

Atlanta. Eyebeam Photography/Getty Images

Ok, not all cities. But in regions like the American south, normal urban attributes like heat, pollution, and tall buildings could stir up more storms. New research examined nearly two decades of meteorological data from Georgia and found thunderstorms were slightly more likely to form over Atlanta than the surrounding rural areas. Through modeling and other research, meteorologists have known about the connection between cities and storms for decades, but this is the first time data has shown the phenomena in action.

During the summer, cumulonimbus clouds ripen like peaches over Georgia and Alabama. These moisture-heavy storms appear on radars as dark, pixellated patterns. Alex Haberlie, a geography doctoral student at Northern Illinois University, used a computer program to find these patterns, then crunched 17 years worth of data. In all, the geographers that found Atlanta is 5 percent more likely to initiate a thunderstorm than the surrounding rural area.

“By our count, that’s a couple to three or more storms a year,” Haberlie, the study’s lead author, says. This doesn’t seem like much, but a city’s built infrastructure compounds the effects of any storm. For instance, Atlanta’s catastrophic 2009 floods probably were worsened because all the asphalt and concrete kept the water from seeping into the soil.

“The discovery that urban environments can create their own storms and rainfall: Not new at all,” says J. Marshall Shepherd, a researcher at the University of Georgia who is an expert in urban meteorology. Researchers have long known that cities generally get more rainfall than their surrounding areas. Shepherd says that he and other researchers teased out the relationship between cities and thunderstorms by running models. “Set up studies where you don’t include Atlanta, and some where you do, and lo and behold you can see that taking away the city reduces the rainfall,” he says. Based on research he’s seen, Shepherd says he isn’t surprised by the 5 percent difference between Atlanta and its realm. He expected the number to be upward of 20 percent. But despite his misgivings about the degree of difference shown in the current study’s results, he says what’s really important is that their methods show that these effects are actually happening.

According to Shepherd and Haberlie, these storms are brewed by several factors. First, cities are hotter than surrounding areas. This warm, rising city air creates circulation that mixes with other atmospheric conditions to create thunderheads. Second, tall buildings form a barrier that pushes wind up and around the city. “Upward motion is always good for thunderstorms,” says Haberlie. Finally, pollution particles act like condensation nuclei that water glom onto, creating droplets. In essence, cities get more thunder and rain because they are hot, stale, and dirty.

But you can’t generalize this type of research to every city. Places like New York, Chicago, and Washington D.C. are probably also contributing to their own thunderstorms, but because these cities are so close to large bodies of water, it would be hard to tease this out of radar data the way Haberlie and his co-authors did with Atlanta. Research like this could help city managers plan for bigger influxes of water, either by opening up more reservoirs, or coming up with better strategies for flooding. But perhaps most important, there is no telling if this research will result in a long-awaited new verse to the AC/DC classic, “Thunderstruck.”


Questions:

1. 
   What are 3 ways the built environment increases the likelihood of a thunderstorm occurring?
   

2. 
   Considering that flooding is a problem in Atlanta, and the built environment is contributing to increased rainfall, what can urban planners do to make the city more sustainable and less likely to flood? ( hint: what did L.A do?)
   

1.Which urban structures/ zones contribute

highest to the heat island?

Why?
2. Which structures/zones contribute the

least to the heat island?

Why?

Is this continuous or non-continuous data? When mapped is it best represented by vector or raster data?