“It never rains here. Storms always come in and split north or south around my city.”
If you’ve been in MN for awhile, you’ve likely heard someone utter this phrase. Living in a SW Minneapolis suburb, I know I have said it many times myself! So, I thought I’d dig in and research if there’s some truth to this storm splitting phenomenon.
The Urban Heat Island Effect
Let’s first tackle the concept of a potential “Urban Heat Island” (UHI) effect. Is there such a thing? Can storms really split and go around a city? Is there validation to feeling that storms consistently pass you up for your suburban neighbors to the north or south?
It turns out – what was commonly thought of as folklore decades ago, has actually been studied quite a bit and proven multiple times to be TRUE!
So let’s give some context. The Urban Heat Island (UHI) effect references the ‘heat’ yielded into the atmosphere from metro areas. It has been often linked to studies about climate change (also out of scope for this post). Spoiler: UHI has not been found to be a major contributor to the overall warming of the planet as much as other factors.
Back to storms, UHI (heat) is really just one part of the research on overall “Urban Impact” of storms. This concept of seems to be fairly well agreed upon for large urban areas (metropolitan cities greater than 25km) and has been pretty well debunked for small features like lakes, rivers. Although, massive MN lakes like Mille Lacs or Lake of the Woods are likely to have some impact on storms as well, but that’s out of scope for this summary.
La Porte Effect
One of the first studies of urban impact to storms was in the 1968. Looking back on years of data, Stanley A. Changnon Jr. noted that the small town of La Porte, IN had 31% more precipitation, 38% more thunderstorms, and 246% more hail days than did surrounding stations. Since La Porte is 30 miles east of Chicago, Changnon claimed the (relative) decrease in precipitation conditions in Chicago were due to human-created features (buildings, etc.) and the (relative) increase in precipitation for downwind areas from Chicago were a result of storms merging back together and intensifying. His research was initially met with criticism, but later accepted as fact – becoming foundation for future research.
The image below show a picture of La Porte relative to Chicago (left) and an illustration from Changnon showing the increase in rainfall surrounding La Porte (right).
Urban Modification of Storms
Building upon the work of Changnon, one of the more recent and commonly sited studies was completed by Purdue professor, Dev Niyogi in 2011. He used 10 years of storm data around the Indianapolis metro area to observe how storms changed shape as they approached and departed a metro area. His conclusion? Similar to La Porte, Niyogi found the urban area of Indianapolis impacted the shape and pattern of storms approximately 60% of the time (compared to 25% in rural areas). Niyogi and his team leveraged several data models which consistently showed the urban area was the contributing factor to those storm pattern changes. Most often, storms would split around the metro area, only to merge back again and intensify.
He concluded there were four major factors that contributed to the splitting of storms near metropolitan areas:
Friction from buildings – buildings tend to have sharper angles which prevent winds from freely flowing through. This can have an impact on fronts or low-level winds that would otherwise be the lifting mechanism (one of the key components) for storm development
Urban Heat Island (UHI) – warming from human related activities (buildings, roads, burning of fuels, etc). The warm air rises into the troposphere and can inhibit storm development or act as a ‘bubble’ causing the storm to go around
Topography / Urban Development Shape – the regional terrain and shape of the metro area can have a nudge effect, pushing storms off to the side and away from cities. This isn’t limited to metro areas and can be found in other topographical features away from cities as well
Pollution – the by-products of human activities (aerosols) are concentrated around cities and can interact with the clouds nearby. Clouds are fairly sensitive to the surrounding air and extra variables can mean the difference between rain or not
Bonus: Urban Parks – Not directly part of Changnon or Niyogi’s research, other studies have shown that very large areas of green vegetation in a metro (think New York Central Park) have been found to actually have a reverse “cooling” effect to UHI and counteract some of the effects of UHI specifically.
Bonus: Night Storms – Some research has shown that UHI has more effect during the day and less at night. Often, storm development conditions increase as a result of daytime surface heating. Likewise, severe storms/development frequently dissipates (as does surface heating) when the sun goes down. So, impacts to metro storms patterns should more likely be attributed during the day than at night.
Interested in more of Niyogi’s research? Here’s a great video of Niyogi talking about his (and Changnon’s) research.
Hit or Miss
Bringing all this back to the topic of the post: “Storms always miss me.” The research from Changnon and Niyogi are important because they help us understand macro patterns of storms around a metro area. However, they don’t entirely explain any one person’s city’s experience for any one particular storm. SO MANY factors play a role in storm development to begin with… it’s hard to say if conditions were ripe for a storm otherwise! A few things to consider are: the natural lifecycle of thunderstorms + temp & dewpoint, fronts, pressure features, and winds.
Lifecycle of a Thunderstorm
NOAA has a number of great resources for learning about storm development, so I won’t duplicate that information here. But it’s good to remember that the normal lifecycle of a thunderstorm last around 30-60 minutes. That means if a storm is traveling 30kts, it might only travel 20-35 miles over the course of it’s entire life (including immature and diminishing phases) before it dies off. Typically, another storm would have to develop and replace it, and so on again, for a system to travel across the entire state.
Invisible conditions: Temperatures/Dewpoints/Fronts/Pressure Features/Winds
Seeing a huge storm 40 miles to the west and watching it fizzle right as it gets to you is flat out disappointing (for someone who loves severe weather – or anyone that just wants rain)! One of the biggest mistakes I see is people looking at radar from miles away to predict rainfall. Radar is a view of what is happening right now. With a loop of history, it is possible to see some of the storms’ behaviors…however, there are so many factors that need to be right for storm development (and can equally cause storms to suffocate)…radar just doesn’t give you a view into the necessary parameters to predict rain. It’s exciting to watch, but it can set you up for disappointment quite easily.
Things to consider…
Is there a front or jet stream parked over your city? How strong and which ways are winds blowing at the surface and aloft? Are temps and dewpoints different around you? How moist is the air at the surface? Does it change with altitude? Is there a shortwave, bulge, or depression (differences in pressure) nearby?
A slight change in any one of those factors could mean the difference between storms for hours or absolutely nothing. Conditions can change quickly over short distances and have a massive impact whether or not you have blue skies or right environment for storm development.
Just a few days ago, I watched a massive storm system develop directly over my house in Prior Lake, MN. To the west were blue skies and overhead clouds and no rain. But just 10miles to the east were 70mph winds, downpour, and 2″ hail. As the storm ripped through multiple areas of the metro north and east of us – it was clear a cold front/wind shear initiated the storm- which didn’t appear impacted at all by the metro area. Just ‘bad luck’ for me.
What does the data say?
So far, much of this has been about theory and probability, One should ask, what does the data say about actual rainfall? This is going to surprise a lot of people.
The chart below was pulled on July 30, 2023 looking back 365 days of observed precipitation. I’ll have to admit, I was surprised myself. Here’s my interpretation of the data:
-Increased rainfall as you approach the Mpls/St. Paul area from the west and continue east
-Relatively small / insignificant decrease (~5in.) in precip just west of the metro, but much further west than I would have thought (St. Cloud, Willmar, etc)
-Relatively small / insignificant increase (~5in.) in precip just east of the metro
-Medium significant increase (~10in.) in pockets of western Wisconsin
-Overall “C” shape in precip groups around the metro area, but not so much in WI
Anticipating that someone will point out the fact that this includes winter/snow, I generated the same report, but only for the last 90 days — which should safely be referred to as “July 30, 2023 YTD Rainfall”.
An interesting observation….if urban impact was a contributor to differences in rainfall, it disproportionally affected the northern suburbs more than southern and also St. Paul seems less affected than Minneapolis!
Taking this one last step…below are historical June rainfalls for 2020-2023. It doesn’t really matter which is which, it seems fairly consistent that areas east (especially south east) of the metro have seen more rain. It also appears that areas west of the metro have seen relatively less rain. But being so far west, is that from the urban impact? You can clearly see the metro area itself has seen less rain, but it’s not a clear urban boundary that defines that.
 |  |
 |  |
Summary
At the time of writing this (2023), we’ve seen a lot of isolated storm systems the past few years. These storm systems by nature are ‘isolated’ to small areas which means MOST of the metro area will not see any action at all. Instead of feeling left out, take comfort that you are likely in the majority by not have seen any rain. Again, just bad luck for your neighborhood.
The data above tells an interesting story. I was surprised by the impact on the west side of the metro being so much further west and quite surprised by the difference on the east side of the metro/state. I will definitely be prioritizing Wisconsin chases in the future!
No part of this post is intending to dispute the idea of urban impact to storms for any area of the metro – but just to illustrate there are MANY parameters that need to align first. Urban impact, whether heating (UHI) or other, plays a role for sure. How much?! More research would be needed to determine for sure.
You may be on the west metro / upwind side and see fewer storms or the east metro / downwind side and see more! In either case, urban impact is a multiplier to the odds already stacked against you!