The March-April edition of WeatherWise magazine has an interesting article in it regarding UHI (Urban Heat Island) effects of enhancing thunderstorm formation in the downwind heat plume. It Stems from this paper (PDF) published in the Bulletin of the American Meteorological Society. I saw a similar study presented in August 2007 when I attended Dr. Roger Pielke’s land use conference presented by Dr. William Cotton on the enhancements modeled in St. Louis, MO. Read that paper here
Excerpts from WeatherWise Magazine:
The Atlanta Thunderstorm Effect
by Mace Bentley, Tony Stallins and Walker Ashley
Although nearly everyone is fascinated by lightning, some of us are terrified, while others are drawn to its elusive beauty. Lightning is one of the most photogenic of all atmospheric phenomena, but also one of the least understood. For all of its beauty, lightning is a major cause of weather-related deaths in the United States and accounts for more deaths than hurricanes and tornadoes combined. Nearly 40 percent of all lightning deaths occur when a person is involved in some form of outdoor recreation.
Now, new evidence suggests that lightning and its parent thunderstorms might actually be enhanced by cities. Urban areas are literally hotbeds for producing heat and lift, two important ingredients for thunderstorm formation. At the same time, throughout the world people are continuing to migrate to cities for employment opportunities and the search for a better life. Eighty percent of the U.S. population now lives in cities. City growth has increased the amount of urbanized land cover in the United States to nearly the size of Ohio! In the United States, many of our cities in the south are growing rapidly due to their location in a more temperate climate. However, a temperate climate also means cities are more prone to thunderstorms. Could all of these factors together combine to increase risk of lightning and other thunderstorm hazards to urban communities around the world?
The Urban Heat Island
The first step in unraveling the question is to understand the interaction between the land and atmosphere. It is currently thought that several processes in this complex interaction are likely at work in altering thunderstorm distributions around cities.
The first is the urban heat island effect, perhaps the most well-known atmospheric phenomenon produced by a city. An urban heat island occurs when the city registers higher temperatures than the surrounding rural areas. Cities heat up because of all the “activity” in them. Cars, air conditioning units, idling engines, and miles of asphalt and concrete all either produce or retain heat. The most notable feature of an urban heat island is the lack of cooling during late afternoon and evening after temperatures normally reach their highest. When compared to the rural countryside, urban corridors have much less area exposed to open air and instead have many warm buildings facing each other. Less heat is lost, and higher nighttime temperatures result. After sunset, city-to-countryside temperature differences grow quickly and can reach, in some cases, more than 10°F. The greatest city-to-countryside temperature differences occur during the long, hot days of summer when daylight is maximized.
…
Caption: Radar climatology illustrating the clustering of strong thunderstorm days directly over and immediately surrounding Atlanta, Georgia.
Caption: Plot of lightning events during periods of westerly flow illustrating downwind enhancement east of Atlanta, Georgia.
The Atlanta Case
The 10-year study of lightning, rainfall, and thunderstorm activity in Atlanta in the summer months showed that enhanced thunderstorm activity was found to shift due to prevailing winds. For example, westerly winds produced a distinct increase in lightning activity east of downtown Atlanta. Evidence suggests that thunderstorms developing over the city center, as well as storms along the periphery, were being directed by the westerly winds to the east side of the city and suburbs. The Atlanta enhancement, particularly for lightning, was well developed for westerly and northwesterly winds that carried thunderstorms east and southeast of downtown. Thunderstorm enhancement can occur in all directions around downtown Atlanta, directed primarily by the prevailing wind direction.
“Pollution can alter how a thunderstorm forms”
“As water droplets collide and freeze onto hail and other ice particles, negative charges are removed from the updraft and added to the downdraft of the thunderstorm”
The rainfall and lightning characteristics of thunderstorms developing in and around Atlanta were also detected when examining radar reflectivity. Over the 10-year study, high radar reflectivity “hotspots” were persistently found along and north of downtown Atlanta and immediately east of the central business district. Towering cumulonimbus clouds containing high concentrations of water droplets and ice crystals reflect significant amounts of microwave radiation back to the radar antenna. On weather radar displays, highly reflective areas are thunderstorms, which are typically color-coded in hot colors (i.e., reds, oranges) to make it easy to identify their size and location. Radar-identified thunderstorms were found to be greatest over the downtown with a general decrease moving outward from the city center. A similar pattern was found over other southern U.S. cities. It appears that the Atlanta urban heat island and associated buildings may combine to produce the downtown thunderstorm radar “hotspot,” while the urban heat island-produced circulations on the fringes of the city lead to increases in suburban thunderstorms, lightning, and rainfall.
“Radar-identified thunderstorms were found to be the greatest over the downtown”
Although less important, the terrain might also be linked to the lightning and rainfall patterns surrounding Atlanta. Winds from the northeast off the Appalachians and the focus of rainfall and lightning activity on the upwind side of Atlanta suggest that elevation changes across the metro area may interact with the urban heat island circulation and focus lightning and rainfall on the north side of the city. One explanation is that air flowing downhill from the Appalachians will be forced to rise once it encounters the buildings on the northern edge of Atlanta. This is distinct from other prevailing wind directions, where lightning activity was found to intensify over and downwind of the city center.
======================================
Read the complete article at WeatherWise Magazine
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
I realized recently that under the right conditions my own property appears to create a mini-UHI effect. I have a large shop, large house, and associated concrete shop apron and blacktop driveway, probably the largest combination of those for miles around in a sea of dark green forest. On certain cold, calm days a temperature inversion will exist, until a bubble of warm air near the surface bursts through the cold layer. When that happens the ravens detect it, their version of the “Surf’s up!” cry goes out, and they fly into it for a free ride up several thousand feet. When their lift in the bubble tops out, they do formation aerobatics and their grab-ass play on the way back down. By watching where they are circling and where they kick out of the bubble, you can see its location and height. For years I’ve enjoyed watching this directly overhead, and finally made the connection that it’s my own property which may be triggering the bubble to burst through the inversion. Now the question is how do the ravens find it — does one of them have to feel the lift by accidentally flying through it, or can they see an atmospheric optical disturbance caused by the warm air, or do they recognize the suite of conditions, including my mini-UHI, that will create it?
Tony, that RADAR image above is the kind of imagery that is rendered from data generated by the USA’s NEXRAD WSR-88D system.
Basically, the NEXRAD WSR-88D uses an S-band frequency (2.7 to 3.0 GHz band and sometimes described as “10 cm” RADAR), a 28 foot dish which at S-band frequencies yields a beamwidth of under 1 degree with a corresponding gain of 45.5 dB. Maximum range is spec’d at 248 nm.
The transmitter output stage musters 750,000 peak Watts, while average power works out to be only 1,560 Watts on account of the pulse width and duty cycle employed.
‘Rain’ penetration is superior to other RADAR bands normally employed for meteorological purposes (C-band and X-band) and is therefore desirable for the mission and desired accuracy of the WSR-88D RADAR. TV Stations employing their own weather RADAR usually employ a C-band model which for 1 degree angular resolution only requires a 14 foot dish antennna.
The use of S-band for the WSR-88D results in reduced attenuation of both the initial incident RF ‘wave’ impinging on precip deep within a thunderstorm but *also* reduced attenuation of that back-scattered energy as it is reflected off the precip back to the RADAR site.
Contrast this with airborne civilian ‘weather’ RADAR as used by the air transport professionals, an X-band (sometimes referred to as a 3 cm RADAR) which is much more attenuated by precip than C or S but results in a smaller dish thereby retaining some finer angular resolution. These beamwidth can be 3 and more degrees resolution but the overall size of the ‘dish’ is kept small (for aircraft nose-mounting installations).
Nowadays, the FAA’s TDWR (Terminal Doppler Weather RADAR) are also available for RADAR weather imagery on the few dozen plus locale’s where they are installed; this is a C-band RADAR with less range (and half the dish size) than the WSR-88D as the TDWR fulfills a different mission than the WSR-88Ds.
WSR-88D Quick specs: http://www.qsl.net/n9zia/pdf/wsr-88d.pdf
Wx RADAR Wiki: http://en.wikipedia.org/wiki/Weather_radar
Rockwell Collins: WXR-2100 commercial aircraft RADAR
Rockwell Collins: User manual WXR-2100 including section on interpreting RADAR imagery for the WXR-2100
.
.
I’ve discussed this with other Naturalists in Texas.
Our observations point to a collapsing of marginal surface based storms as they enter the UHI of Dallas-Fort Worth. My guess is that the humidity is less in the urban areas as there is less vegetation respiring to recharge the low-levels.
On birds – and updrafts – they can feel it – just like ultra light or glider pilots can feel it.
As a professional civil engineer (now retired), with a career as a transportation engineer, I am of the opinion that the UHI effect of asphalt concrete (AC) is significantly underestimated.
This is somewhat of an extreme, but circa 1970, the temperature (in August, mid afternoon) of a recently constructed section of state highway near Tucson, AZ was measured by the state of AZ highway engineering department at 167 degrees F. Many municipal jurisdictions cover the AC streets every 5 or so years with a mixture of fine aggregate and asphalt, returning them to a very low albedo once again, so that is not so extreme a temperature after all, particularly in cities here in the southwest.
Just got me a new toy a couple of days ago, a non-contact temperature measuring device, so just gave it a try. 2:00 PM here in Phoenix, AZ,
March 18, 2010, air temperature (nearby weather station per Weather Underground) was 80 degrees F. Sidewalk (Portland cement concrete, PCC) was 110 degrees F. AC pavement (somewhat weathered) was 117 degrees F, and last years crack seal material was 121 degrees F.
AC attains a significantly higher temperature in sunlight (thus storing lots more heat) than other construction materials commonly used. When the sun “goes down”, the AC continues to warm the much cooler air due to conduction (at the surface of the AC), and because of convection keeps warming new cooler air, and radiates IR to other structures in line of sight.
For doubters, here is a great physical experiment to do this summer: At mid afternoon on a sunny day, just place your bare foot first on PCC curb/gutter or sidewalk, then place it on the AC pavement (you won’t be keeping your bare foot on the AC all that long, as it will be quite painful).
Asphalt concrete combined with sunlight is the “gift” of heat that keeps on giving – long after sunset, as well as during daylight hours.
I haven’t seen anything written up about it, but there’s another urban effect for towns on the Great Plains, including those along the Colorado Front Range. This area was virtually a desert when people first began living here. The only trees in Colorado Springs were those along the several ridgelines in town (almost exclusively evergreens) and cottonwoods and willows along the few year-round creeks. Today the city has more than two million trees, including a majority of deciduous varieties. Add in green lawns, and you have a serious increase in humidity – and moderation of temperatures – in the city and downwind. I’m sure the thunderstorm frequency on the plains east of Denver is also greater, with both a large urban heat-island effect and an increase in general moisture content of the ambient air.
Jim – enjoyed your comment on RADAR. You are exactly right.
That’s not really the sense we storm spotters who are often glued to RADAR and satellite imagery during thunderstorm events have. In fact, given the surface ‘flow’ of moisture from from the GOM (Gulf of Mexico) Ft Worth is less in that moisture flow than Dallas is (check a map, drop straight south from Ft. Worth).
I have literally witnessed initiation of a ‘squall line’ late in the afternoon along the dryline along I-35E in Dallas. In fact, being a ‘naturalist’, you should note the change in plant life basically east of I-35 owing to increased seasonal rainfall amounts (this is also very apparent in which Corp of Engineers reservoirs take longer to fill up, etc, in the Ft. Worth area vs Dallas) … and becomes especially apparent by the time you reach east Texas …
For this ‘sense’ you guys have, is it based on any observations (a time or two) … or more speculation? It would help to know, one could go back and look at other factors that may have intervened.
Like – take into consideration meteorological factors such as powerful thunderstorm-inhibiting factors, like as measured by the CIN or CInh (Convective Inhibition) or more informally ‘the cap’ (the capping inversion); it is the primary factor that determines, 9.9 times out of 10, when and if we will have severe thunderstorm activity here in North Central Texas esp during the spring and summer thunderstorm season. The flow at this level is a little more variable than the boundary layer (which is predominantly S in spring/summer).
THEN there is the location of the ‘jet’ or any other upper level low-pressure activity (SW trough, etc) that can influence T-storm initiation and progress and lifetime, before ‘cap’ effects begin to dominate, but I digress …
See ‘cap’: http://www.srh.noaa.gov/jetstream//append/glossary_c.htm
.
.
Any compensation for (surface) emissivity (of the test subject)? That seems to be the biggest problem with non-contact thermal measurements; EVERYONE trusts that miraculous little digital display, whether or not it is anywhrere close!
It would be nice to double-check those surfaces with a contact temperature measurement device, just to establish the correlation.
.
.
” Tony (07:05:27) :
That is an interesting image of a thunderstorm. As an airline pilot flying in the tropical Asia, I see many many thunderstorms and our weather radars typically paint a completely different image to the one presented above. Especially when the precipitation is heavy our radars often suffer severe attenuation. It would be interesting to know how that image was actually formed.”
Tony, the additional information on the display indicates that the WSR88D is operating in VCP mode 212 (Volume coverage pattern 212). It doesn’t give you the specific nexrad product, but its most likely “base reflectivity .05&0176;” or, the lowest tilt, as this is the usual tilt for looking for low level tornados in reflectivity data.
VCP 212 is a scan protocol that steps through 14 tilt angles (5 deg – 19.5) in 4 1/2 minutes. Its a fast scan used for rapidly evolving, widespread severe convective storms. Since it is likely .5 degree tilt, you’re looking at no more than about the first 10K feet of vertical storm (above ground). In this storm there is likely another 40Kfeet of stuff above! The pattern presented therefore represents the structure for only the lower part of the storm. In the 4 1/2 minute scan the radar is also doing doppler wind and outputting a bunch of derived products too, and the operator or data feed recipient can also step through the other tilts to see additional structure or visualize the storm in 3D. The pattern shown is characteristic of an HP supercell.
For an awesome multicell Dallas/Ft. Worth event, I’ll leave you guys with these two animations:
DFW composite reflectivity
and the same as “Echo tops” – measuring vertical height
And since _Jim mentioned dryline development, here is a convective dryline being transformed in Kansas: Topeka
An unemphasized aspect of this report is the value and quality of fieldwork in observation and analysis of real weather. This sort of work was the main driver of operational research through the first three-quarters of the 20th Century. In recent years much of the work has moved indoors to the laboratories of academic institutions where the understanding of weather relies on remote sensing, often retrospectively, and the modeling of weather systems from the macroscale through the mesoscale to the microscale has blurred the detailed and often chaotic character of actual weather systems to produce overly smoothed depictions so beloved by many modelers who use insufficient resolution to capture the fine details that could provide for more insightful hypothecation.
On the subject of UHI effects on convective storm systems, field studies of these systems were carried out in great detail from the late 1940s and early 1950s. These studies were undertaken shortly after the end of WWII when radars became available for real time observations of such storms in three dimensions. Although such work is still going on today, namely at places like OU where storm chasing has been elevated to a genuinely academically involved study, it has not been as well supported by the percentage of funding being doled out to the meteorological research community that it commanded back in the good old days when field observations reigned supreme.
A good example of the kind of work that was being done in those early days can be found in a comprehensive report on the activities of the Illinois Water Survey covering the latter half of the 20th Century. People like Stan Changnon, Floyd Huff, Dick Semonin, Ken Kunkel among many others, produced a stream of papers on the studies of weather and hydrology with emphasis on the impact to crops and local flooding. The report can found here. Note especially the extensive bibliography of the papers they authored including numerous studies of the UHI created by the city of St Louis and its effect on the inadvertent modification of convective storm systems over and downwind of the city. Much of this work was done between 1968 and 1979 with Changnon the lead author.
The principal difference between their studies and a lot of the stuff that’s being done today was that the observational data came first. The theories and conclusions were derived after the data was collected and analyzed. Today we see many examples of the theories or more properly the hypotheses being determined in advance before a search was undertaken to find data to support them.
For a near current illustration of all that is wrong with today’s climate science, it’s worth reading Steve McIntyre’s account of going to Colorado on a working vacation to actually drill a few cores from a grove of bristlecone pines where several trees had been used in paleometeorological studies and trying to get some of the actual dendroclimatologists from Boulder to take a couple of hours’ drive to join him in the effort. They wouldn’t leave the comfort of their ivory tower to take him up on that, even when Steve pointed out that there was a Starbucks within easy range of the groves in question. His reports can be found here.
Hi Austin, me again, on another point this time.
Actually, I think the contrary is true –
– we now have a) reservoirs and b) landscape/yard irrigation c) a lot of humans (they exhale CO2 and H2O) and d) a lots of hydro-carbon combustion (products of combustion CO2 and H2O)
Also, our ‘flow’ is up from the GOM, so a _lot_ of what we generate locally is going to wind up downwind from us, which is going to be the Red River valley and into Oklahoma (speaking of the source as the Dallas area now).
Current surface conditions … and on Friday they will indicate winds from the south for the Dallas area (given the prognostications)
In fact, lets take a look at the Convective Outlook from by the SPC for Friday into Saturday here in NCTexas where they discuss moisture:
ALLOWING MOISTURE TO ADVECT NWD is the money quote, and that moisture comes from the Gulf on this occasion …
Storm Prediction Center outlooks: http://www.spc.noaa.gov/products/outlook/
.
.
I couldn’t agree with this more (anecdotally speaking).
I was stationed east of Cleveland at a Nike missile base HQ around 1970, back before the Clean Air & Water Act had a chance to have an effect.
It was in the heart of an area called “The Snow Belt.” And the base was also on top of the highest land in the county, which contributed some cooling (and thus condensation was accentuated). But there was all kinds of pollution – especially as compared to today.
The freaking rain storms were just ridiculous, filling up the wastewater ditches usually inside of five minutes. I still marvel at the downpours there, and just shake my head.
It was explained to me by someone stationed with me about the weather coming across Lake Erie and the water condensing on particulates in the air from the heavy industry and the pre-catalytic converter cars. And I believed him; the evidence was right there. It was the only place in the US I’ve ever compared to monsoons.
Yeah, it is one of those “Don’t you know the difference between weather and climate?” things, but I will forever remember the army buildings getting flooded as often as not. I have a hugely funny weather story about one of the dumbest Warrant Officers in the history of the world. Think Frank Burns… ROFL just thinking about him.
Good memories!
And all about rain.
Oh, snow, too. SNOW. MANY intense, intense snowfalls there.
This isn’t surprising. If you watch time-loop sat cloud views of the tropics, islands are preferred Tstorm locations compared to open ocean. This is because the land heats more than the ocean during the day, and a contrast is setup between the two. Such localized contrasts are what help Tstorms to develop. Night-time Tstorms can also develop just offshore of coasts or islands for the opposite reason — land cooling off more than the adjacent ocean during the night.
A large city surrounded by cooler forest or farmland can be just like an island.
Kind of continuing to demonstrate how ‘weather’ comes together across an area I am familiar with (the DFW area), here’s an excerpt from the Day 1 Convective Outlook from this morning:
(Emphasis mine))
And indeed winds are winding up a little with nice flow off the GOM and dew points in the upper 50’s (50 deg. F) just onshore in Texas and 40’s (40 deg. F) further into Texas and into Oklahoma:
Surface obs Friday, 2-19-2010 1433 UTC (9:33 AM CDT) (This link is a ‘capture’ of the surface obs image from rap.ucar.edu/weather/surface and locked in time via an image saved on tinypic.com)
.
.
Continuing to look at the developing wx system … this is a fairly typical scenario setting up prior to a front coming through with warm, somewhat moist air coming off the Gulf (GOM); note that the wind has picked up quite noticeably from earlier (many of the wind ‘flags’ have 1 1/2 to two barbs showing – see the nice SE wind swoop off the Gulf then N into TX and OK):
Surface obs Friday, 2-19-2010 2033 UTC (3:33 PM CDT) (Link goes to a ‘capture’ of the surface obs image from rap.ucar.edu/weather/surface locked in time via tinypic.com image)
The meat of the size-up from the SPC (Storm Prediction Center) Day 1 Outlook:
SEVERE THUNDERSTORM POTENTIAL WITH THIS SYSTEM REMAINS
MARGINAL THRU THE FORECAST PERIOD…PRIMARILY DUE TO SUCH A MEAGER RETURN OF GULF MOISTURE IN THE WARM SECTOR.
DEWPOINTS ONLY INTO THE LOW/MID 50S ARE ABOUT THE BEST THAT CAN BE EXPECTED SRN PLAINS PRIOR TO FROPA.
HOWEVER A TYPICAL EARLY SPRINGTIME EML IS SPREADING EWD ACROSS THE SRN PLAINS AND COMBINED WITH SUFFICIENT HEATING SHOULD LEAD TO A REMOVAL OF MUCH OF THE CINH BY MID AFTERNOON. SURFACE BASED THUNDERSTORMS ARE EXPECTED TO DEVELOP AFTER 21Z ALONG AND JUST AHEAD OF THE SEWD MOVING COLD FRONT SWRN OK INTO NWRN TX.
The “EARLY SPRINGTIME EML’ I had not seen before in Convective Outlooks, EML standing for “Elevated Mixed Layer”.
For more on EML see: The Elevated Mixed Layer
and
A Synoptic Climatology of the Elevated Mixed-Layer Inversion over the Southern Great Plains in Spring. Part I: Structure, Dynamics, and Seasonal Evolution
.
.
And – to tie the ribbon on it … we today in N. Central Texas on Sunday the 21st of March have received approximately 4.5 inches of snow from the wx system that ultimately resulted from this southerly flow a day or so earlier … here is the unscientific snow measurement taken on the top of my car this AM:
4.5 inch snowfall in Collin County (north of Dallas) 3-21-2010
As we had gusty winds with this system, we had smallish drifts here and there, and initially a lot of the snow on concrete/road surfaces melt … snow depth on the sidewalk measured 1 inch, the road surface proper about 1/2 inch.
.
.
1285296