Why Southeast US Airports Are Often Cold Spots

[Can't think of username]

All my research into why the Tallahassee airport official weather station is such a cold spot has taught me a lot as to why other Southeast US airports are also cold spots. 
To give those who missed those comments some context, the Tallahassee airport is a cold spot that has colder lows than many other places in the Southeast US because deforestation surrounding the airport leads to massively disproportionate soil drainage and ground/sky exposure that are both apt to make readings colder on the clear, calm, nights that are all too common in the Southeast US. 3 studies have been conducted on the matter and I can recommend all of them.
https://www.researchgate.net/publication/239751448_Tallahassee_Florida_Minimum_Temperature_Anomaly_Description_and_Speculations
https://ui.adsabs.harvard.edu/abs/1998JApMe..37..101K/abstract
https://ams.confex.com/ams/pdfpapers/91431.pdf

I've commented and learned plenty about Tallahassee airport at this point, so this thread is going to focus on other Southeast US airports that are the cold spots of their areas. Some examples that come to mind are Wilmington airport, Montgomery airport, Brooksville airport, Valdosta airport, Augusta Bush Field airport, and Charleston/Savannah inland airports, all these are located - like Tallahassee airport - in areas that have sandy soil and thick pine forest but have their weather stations located outside of the forest so the same effect as Tallahassee happens, even to a lesser extent.

Those of you who have been keeping tabs on Southeast US weather for some time may have noticed that during radiational cooling events they end up getting colder than normally-colder places like Atlanta which have clay soil and are outside the pine forest so are not nearly as disproportionately affected by this (some examples are November 2016/2021, and January 2022). It's also the reason behind the surprisingly large diurnal range in places like the above: not only do the lows get colder, but during the day the sky exposure and the dry sandy soil heats up more than it would in its natural pine-covered state, and this isn't what would be expected, at all, from low-latitude, humid climates near sea level.

Unsurprisingly, though, the fact that this is because of forest clearing in a naturally forested landscape means it's artificial and not representative of the actual climate. Obviously there's no way to know what the natural climate of the weather stations would be like without relocating them to proper forested locations, but I think the difference would be pretty substantial: when I brought this topic on City-Data weather I got the following feedback on cleared vs forested land: 

"The effect of forest clearing is pretty major around here, where I'm guessing about 30% of subcanopy vegetation isn't able to grow in the open -mature lowland forest around here is an environment that doesn't see temperatures reaching 0C, while only a hundred metres away, cleared land can see dozens of frosts during winter."
https://www.city-data.com/forum/weather/3366619-what-like-experience-4-distinct-seasons-18.html#post65041669

"Just on my property, the difference between forested and cleared land can be anywhere from 2C-6C"
https://www.city-data.com/forum/weather/3366619-what-like-experience-4-distinct-seasons-19.html#post65043527

Granted I am not really familiar with climates outside of North America, but New Zealand has oceanic climates which probably aren't as prone to radiational cooling as the Southeast US, and probably has moister, less sandy soil too. So the effects described there would probably be even more substantial in the Southeast US.

Let's take Crestview/'The Icebox of Florida' as an example (chosen because in the Tallahassee research papers, the authors suspect it may be experiencing the same massively disproportionate effect Tallahassee does). 
According to Wikipedia the average frost free season goes from March 23 to November 10, the usual number of frosts a year is 38, the usual coldest temperature of the season is -7.7C, and the diurnal range is a high 13.5C over the course of the year. I don't usually monitor it AT ALL outside of years that look like they may repeat the early last frosts in 2012 or 2020 because it performs so poorly for being a climate at 30.45N in Florida, so I can certainly see why the authors suspect so.

If the airport weather station had its natural thick forest covering, it can be conservatively assumed based on what happens in New Zealand that the number of frosts would be in the teens, the usual coldest temperature would be in zone 9b instead of 8b, and that at least a month on both sides could conceivably be added to the average frost free season. The odd winter would probably also pass without frost, going by the capability of other 9b places that have usual annual frost in the teens.
Plus the diurnal range would be average or even low (as would be expected from a warm, humid, low altitude climate). As mentioned, together with the warmer lows there would be cooler highs due to less daytime heating potential.

One additional factor that may play into things is low-altitude airport sites that pool cold air on clear and calm nights. I used to think that was the case for many of these because their airports have at least a 2m lower altitude than the cities (the altitude at which 0C can happen when it's 2-3C at 2m up), but now I am not so sure.
This is because, as the Tallahassee research papers have shown, the Tallahassee airport is not actually a low elevation cold hole despite having the lowest elevation in the city.  It is too far away from higher altitude areas to actually pool cold air that sinks down from them, making its colder lows entirely a product of that forest clearing, and I have no way to rule out that this is also the case for the other airports that are colder and lower in elevation than their cities (which is all of the above examples except Charleston and Savannah).

The only one I know for sure that is actually a low elevation cold hole is the Austin-Bergstrom International Airport, because this is explicitly admitted by weather.gov.  This does not have to do with forest clearing either: Austin is inherently not a very forested place and the highs are annually identical to the downtown station (annual average high is 26.9C for both).

"The other ASOS site is located at Austin Bergstrom International Airport. The period of record at Bergstrom goes back to October 1942. This site is located in a low lying area in the drainage basin of the Onion Creek. Because of this, the overnight low temperature in the wintertime under clear skies can sometimes be 10 degrees colder at Austin Bergstrom compared to Austin Camp Mabry. The distance between the ASOS sites is only 10.7 miles."
https://www.weather.gov/media/ewx/climate/ClimateSummary-ewx-Austin.pdf

It's possible a lot of these others also are, but that requires confirmation one way or another.

Overall I think it's a real shame that this is the case for so many Southeast US airports, because official weather sites are artificially different from the natural environment and so have poorer weather monitoring than they otherwise would😔. Wunderground stations often provide a solution to the monitoring problem, but there's nothing that can be done about the official site being artificial problem other than either relocating sites to forest or setting up additional forested sites, both of which are unlikely.
But at least now I can know and understand why the high diurnal ranges and poor monitoring times happen!

Thoughts?

Edited June 5, 2023 by Can't think of username

[flplantguy]

I have noticed this in my research as well into soils and why some areas like Brooksville get so cold vs Tampa.  Dry coarse sand is terrible for heat retention at night and that area is higher hills and sand; as opposed to further south where the cypress swamps and wet soil form. My home, nearly due west of that airport, is (i believe from observing) noticably warmer due to higher moisture content in the soil, the gulf marshes a mile west (and open water 2 miles), and the thick forest in back. That area even at christmas was 30 (F) while the exposed area in front was 27 degrees.  Not a big amount but many times a year brooksville is colder still.  they can frost or freeze many times with only one by me if/when the water chills.  The soil effects the temp in summer too, as the temp drops in the evening (more like plummets) it meets the dewpoint and forces the dewpoint to drop with it for a bit.  That never happens in an established urban heat island here. 

I have seen papers about humans and their effect on the peninsula causing/exacerbating the freezes of the 80s due to wetland and tree destruction, as well as pollution during the time causing the hadley cell to supress and let colder air into the US.  Once the clean air act was passed this changed, and now south sudan is flooding from the hadley cell expansion (again allegedly not enough data yet).  Lots of coincidences we dont have a full handle on yet but a big factor is human activity like you said.  One solution is multiple city stations, that are found to be accurate, to determine the difference but thats not easy in an urban dynamic system.

[Can't think of username]

1 hour ago, flplantguy said:

I have noticed this in my research as well into soils and why some areas like Brooksville get so cold vs Tampa.  Dry coarse sand is terrible for heat retention at night and that area is higher hills and sand; as opposed to further south where the cypress swamps and wet soil form. My home, nearly due west of that airport, is (i believe from observing) noticably warmer due to higher moisture content in the soil, the gulf marshes a mile west (and open water 2 miles), and the thick forest in back. That area even at christmas was 30 (F) while the exposed area in front was 27 degrees.  Not a big amount but many times a year brooksville is colder still.  they can frost or freeze many times with only one by me if/when the water chills.  The soil effects the temp in summer too, as the temp drops in the evening (more like plummets) it meets the dewpoint and forces the dewpoint to drop with it for a bit.  That never happens in an established urban heat island here. 

I have seen papers about humans and their effect on the peninsula causing/exacerbating the freezes of the 80s due to wetland and tree destruction, as well as pollution during the time causing the hadley cell to supress and let colder air into the US.  Once the clean air act was passed this changed, and now south sudan is flooding from the hadley cell expansion (again allegedly not enough data yet).  Lots of coincidences we dont have a full handle on yet but a big factor is human activity like you said.  One solution is multiple city stations, that are found to be accurate, to determine the difference but thats not easy in an urban dynamic system.

Yeah Brooksville airport is a very similar case to Tallahassee airport in being so massively disproportionate. Enormous diurnal ranges and with average/usual monthly/record lows that are far more in line with the I-10 corridor than in the Tampa Bay area, I have seen the phenomenon you are referring to a LOT this past winter when monitoring it vs Wunderground stations.

For what it's worth, there is the possibility that this may be a low altitude cold sink instead of simply being affected by the deforestation, because Brooksville is very hilly and KBKV is 36 meters lower than the downtown. It's important to remember that this needs conformation (after all it's nearly identical to the scenario in Tallahassee that I mentioned), but it can't be ruled out.

They did have another, much higher elevation station on Chinsegut Hill that sadly was shut down in 2012. It was exactly as you describe your house: much warmer lows, much less frost, etc - although I do suspect that it may have been affected by the weather station deforesting because it has had frosts in April and October which seem rather unlikely for ANYWHERE in the Tampa Bay area without some artificial cooling.

Don't know what to say about the human factors you mentioned, it just takes the words out of my mouth. Your proposal of the multiple city stations is a very good one: I wish all weather.gov offices would do it. In the meantime, we have a mix of the occasional office that does do it like downtown Austin, and Wunderground stations set up by companies like WeatherSTEM that the offices use, such as their stations on Tallahassee FSU campus.

Do be aware though that with the WeatherSTEM stations I was told by kinzyjr that they are not direct comparisons to the airport due to being set up to other standards - think that's for better representativity of the environment surrounding the weather station as the WMO outlines.  Accuracy as you say is important, but even if they can't necessarily be compared directly they are useful for now because of the mentioned representativity of the surroundings.

A final thought though is that it's possible for at least minimum temperatures that would be under natural forest cover to be well represented or even underestimated by Southeast US urban heat islands instead of overestimated - I have seen the opposite postulated on City-Data weather so it may be worth a mention here. With some exceptions like the infamous-for-heat-islands-Dallas and New Orleans, I was told Southeast US cities tend to have very little heat island compared to the giants (and that makes sense with regards to listed places like Brooksville or Valdosta, neither measures even close to Dallas and New Orleans).
Much more than can be said for the forest clearing, which of course is very serious as we have been saying. I think there's a very real possibility that in at least some urban stations the readings may still run low compared to hypothetical forested stations.

Edited June 6, 2023 by Can't think of username

[bubba]

Deep, well thought through and appreciated.

[flplantguy]

The low spots in the sandhills here are all over, and highly varied, so your right about that too.  In 2010 it was too cold for some of the plants that are very nearby to have survived and they are in street views, and many queens in the area have bad lower trunk damage where they did survive.  Coastal hernando county, brooksville proper, and Dade city did not have this level of damage (and thats on three sides) but in the "core" of that pocket not a single pygmy date and few queens survived.  Thats shady hills in Pasco county and spring hill, of which parts go below 20 on occasion.  This year other spots in the same area have surviving foxtails but nothing near the airport at all (most other common stuff was minimal damage).  I need to find those studies about urban heat islands and the wetlands in florida.  Tampa has a heat island study already so maybe more are joining in.

[kinzyjr]

33 minutes ago, flplantguy said:

The low spots in the sandhills here are all over, and highly varied, so your right about that too.  In 2010 it was too cold for some of the plants that are very nearby to have survived and they are in street views, and many queens in the area have bad lower trunk damage where they did survive.  Coastal hernando county, brooksville proper, and Dade city did not have this level of damage (and thats on three sides) but in the "core" of that pocket not a single pygmy date and few queens survived.  Thats shady hills in Pasco county and spring hill, of which parts go below 20 on occasion.  This year other spots in the same area have surviving foxtails but nothing near the airport at all (most other common stuff was minimal damage).  I need to find those studies about urban heat islands and the wetlands in florida.  Tampa has a heat island study already so maybe more are joining in.

I posted some recent articles about the UHI here: https://www.palmtalk.org/forum/topic/69728-urban-heat-island-research-and-scholarly-articles/

[flplantguy]

Thanks @kinzyjr thats a lot!  Ill look them over and geek out lol. 

When i have a desktop again that will help. I have tons of cold weather info in folders (some from you on here) but i think you have a lot more here i havent looked into. 

[Can't think of username]

6 minutes ago, kinzyjr said:

I posted some recent articles about the UHI here: https://www.palmtalk.org/forum/topic/69728-urban-heat-island-research-and-scholarly-articles/

I don't know if you know this but it's worth noting that this study doesn't actually measure urban heat island directly. I was told on City-Data weather that what it does is it produces an estimate of what the urban heat island is based on the factors like building height and albedo - and, in this particular case, some estimates may be off.

For reference, this same study estimated the exact same heat island for Gainesville as Orlando and Tampa are estimated to have in that link. Needless to say this doesn't make sense because Gainesville is a much smaller and less populated place than either, and so something went wrong for all of them to have the same estimate.

Alternatively, for at least Orlando, we can compare Orlando International Airport to the near-downtown Executive Airport. The difference between the 2 is at best 2C in January lows, certainly not the 3.3C the study estimates.

Hope this helps!😄

[kinzyjr]

3 minutes ago, Can't think of username said:

I don't know if you know this but it's worth noting that this study doesn't actually measure urban heat island directly. I was told on City-Data weather that what it does is it produces an estimate of what the urban heat island is based on the factors like building height and albedo - and, in this particular case, some estimates may be off.

For reference, this same study estimated the exact same heat island for Gainesville as Orlando and Tampa are estimated to have in that link. Needless to say this doesn't make sense because Gainesville is a much smaller and less populated place than either, and so something went wrong for all of them to have the same estimate.

Alternatively, for at least Orlando, we can compare Orlando International Airport to the near-downtown Executive Airport. The difference between the 2 is at best 2C in January lows, certainly not the 3.3C the study estimates.

Hope this helps!😄

It is a useful insight for the first post on the thread from my point of view.

There are multiple studies listed there.  There is a noticeable difference in what plants survive or the ultimate size they achieve inside of the city limits vs. only a few miles outside of the urban areas.  2C (~3.5F) on the worst 2 days in December, January, and February here is the difference between no damage and replacing a plant.

Probably the best recent local examples are in the January 2022 Freeze Report and December 2022 Christmas Freeze threads.

In my own case, the city I live in is not even as large as Gainesville, but there is a distinct temperature advantage demonstrated by several indicator plants.  The advantage during radiational freezes is significant, illustrated by the January 2022 freeze.

Advective freezes shrink the advantage to between 1F and 3F most of the time, but it does still impact ultimate survival.

[Can't think of username]

8 hours ago, kinzyjr said:

It is a useful insight for the first post on the thread from my point of view.

Most certainly. Even if the formula needs work, they are correctly linking the increase of the urban heat island to factors such as greater albedo, and that is always useful.

8 hours ago, kinzyjr said:

There are multiple studies listed there.

I was referring to the 2021 one by Climate Central that estimated the same for Gainesville as Orlando and Tampa. Some of the other I will definitely have to take a look at, but I deliberately didn't comment about them because I hadn't looked into them as much as Climate Central's.

8 hours ago, kinzyjr said:

There is a noticeable difference in what plants survive or the ultimate size they achieve inside of the city limits vs. only a few miles outside of the urban areas.  2C (~3.5F) on the worst 2 days in December, January, and February here is the difference between no damage and replacing a plant.

Probably the best recent local examples are in the January 2022 Freeze Report and December 2022 Christmas Freeze threads.

In my own case, the city I live in is not even as large as Gainesville, but there is a distinct temperature advantage demonstrated by several indicator plants.  The advantage during radiational freezes is significant, illustrated by the January 2022 freeze.

Advective freezes shrink the advantage to between 1F and 3F most of the time, but it does still impact ultimate survival.

I'm not saying Lakeland doesn't have a distinct heat island, but it can't be ruled out that cold air drainage and/or the disproportionate artificial cold that weather station standards cause in the Southeast US are playing a role in making the urban areas survivable for those indicator plants. Lakeland (as you are no doubt aware) has some hills, an airport station 19m lower in altitude than the actual urban area, and usually plenty of radiational cooling to let all those factors shine during the occasional frost - all of which are or may be conducive to those factors.

I suppose it doesn't really have any impact on your success, vs if it was from a heat island, because the end result is still the same😀. But I think it's important for reference in terms of what heat islands can do because if there comes a point when heat islands notably increase (as there probably will come, of course), the success may increase even more than expected due to the heat island not initially playing that much role.
 

[SubTropicRay]

It's the bonafide heat island in Tampa, often 2-5 degrees warmer than the surrounding area.  

[Can't think of username]

Forgot to mention that some airports are naturally cold spots to an extent (regardless of how much they may be affected by the tree clearing or low altitudes) because they are inland to a coastal downtown, which gives people the wrong impression about the climate. Charleston and Savannah are some of the better known examples, Charleston has an official waterfront station and Savannah has a nonofficial weather-us Tybee Island station that are both warmer in terms of lows and lower in diurnal range than the inland airports.




I think the Tybee Island one is most impressive. On City-Data weather I was told on the Facebook gardening groups that the Sea Island gardeners say most winters the islands like Tybee are 10a, this I certainly can buy given the average 8.4C January lows and I'd presumably find palm grower accounts to back it up too if I searched on here.

There are countless more examples of this. Mobile and Dauphin Island comes to mind for official stations, I was told the Mobile Regional Airport is a cold spot 48km away from the water and that really shows when it's compared to Dauphin Island right on the water with plenty of water to the north.


Wilmington is another example. Sadly no official station exists for the coastal downtown, but the inland airport seems to only rarely get December first frost/February last frosts - for reference it seems to be less common than the ATLANTA airport in the 1991-2020 normals which I find shocking, although Wunderground stations do of course help solve this problem. 

I have been monitoring one this past season that got its first frost on December 18 vs November 18 at the airport, and last frost February 4 vs March 21 - although funny enough, despite how close to normal the frost dates of last season are for the airport, the downtown probably wasn't because it would have a longer frost free season than the OBX if it was. Really wonder what the normals for the city are, perhaps similar to a place like Atlantic Beach or the right-on-the-ocean Myrtle Beach airport?🤔

I imagine for the palm growers there downtown is better than inland around the airport because there's times when it can be 3C warmer than the airport (like this past March 21, -2C vs 1C) that as kinzyjr mentioned can make all the difference: https://www.wunderground.com/dashboard/pws/KNCWILMI428/table/2023-03-18/2023-03-18/monthly

Wish all the coastal places were like Mobile and Charleston in having waterfront stations. Much more representative than inland cold spots if you ask me.

Edited June 6, 2023 by Can't think of username

[Can't think of username]

Here are a couple Southeast US airports that for comparison are not cold spots - or at least much less so. For reference I have seen all of them stay warmer during a radiational cooling event than the discussed Austin and Tallahassee cold spot airports - and they certainly have been so compared to places like Brooksville even if I haven't seen it.

Atlanta. I believe the reason this airport isn't a cold spot is because it's in the Piedmont that apparently has clay soil and so is not going to get the same massively disproportionate artificial cooling from weather station deforestation that airports in sandy soiled pine forest would. This also explains why its usual monthly maximums are modest compared to a similar latitude and sandy soiled place like Columbia or Augusta cold spot airport.
(It's got nothing to do with Atlanta's urban heat island by the way. Hartsfield Jackson Airport is outside downtown, and WeatherSTEM's actual downtown station at Bobby Dodd Stadium is warmer for both highs and lows).


Augusta Daniel Field. No climate box available for this, but it has a longer frost-free season than Charleston/Savannah cold spot airports (late Feb-early Dec vs early Mar-late Nov), a warmer usual seasonal low (-5C vs -5.8-7C), and even warmer average lows/usual monthly minima than they do during the cooler and drier season - quite an impressive feat by the standards of cold spot Southeast US airports for a place 180km inland and 33.5 degrees north. Normals are available on weather.gov/cae.
I believe this is because it has an elevation of 129m that sharply drops in the surroundings, giving it excellent cold air drainage.  For reference I think of it as a 3 degrees further north approximate equivalent to the FSU Tallahassee station on a 65m tall hill that I monitor so much, given the similar elevation advantage and analogous large discrepancies with the cold spot other (Bush FIeld) airport during the cool season.

Dallas Love Field. Contrary to popular belief, I believe the reason this airport isn't a cold spot is primarily because it's in some geographic sweet spots - the popular belief is urban heat island but I don't think there's very much of that at all, because Dallas Love Field is quite similar in its coldest month (the one where it can be compared in this regard, more below) to rural stations at similar latitudes, particularly Shreveport Regional Airport (2.9C vs 3.3C).
I think it has the following 3 going for it: being right in the middle of the horse latitudes while on the I-35 corridor (meaning it's going to have rather hot summers for a Southeast US climate, and as a result the rest of the year outside of the coldest month will have to be naturally warmer than a place like Shreveport in order to realistically keep up with that), as well as being located in the wet clay soil of the Trinity River's alluvial plain (which keeps nights warmer during the boatload of radiational cooling North Central Texas gets due to the soil's poor heat loss). For reference about the latter, and proof against the urban heat island theory, it tends to be colder than just about everywhere else (even rural low elevation cold sink KDTO) during a heatwave - when a Dallas size heat island would be realistically warmer - BECAUSE of that clay soil. The same poorer heat loss means poorer heat gain.


There are probably some more that I'm missing, when they come to mind I'll add them too because more context is always better.

[Can't think of username]

@flplantguySomething interesting you might like to know about Brooksville airport. I still don't know whether it has the Tallahassee-style disproportionate artificial cooling, or if it's low enough in elevations compared to surrounding areas to pool cold air, but what I do know is that its airflow is probably exacerbating whatever the problem is.

I couldn't help noticing that KBKV is only 18km from the beach despite seemingly lacking any influence whatsoever - for reference that is more or less the same as Jacksonville airport which is FAR less of a cold spot and has multiple monthly average lows/usual minima/record lows similar to or warmer than KBKV despite being so much further north. Then I figured: Brooksville is at the same latitude as not-a-cold-spot-and-borderline-tropical-Orlando, so perhaps it is getting a significant minority of easterly trade winds in the cool season that block out the western beach influence yet would do nothing to stop either forest clearing or low altitude cold.

Lo and behold.....that's more or less exactly what seems to be happening. The IEM shows quite a significant minority of trades: https://www.mesonet.agron.iastate.edu/sites/windrose.phtml?station=BKV&network=FL_ASOS

So for reference, I think if we had a properly cold air-drained, not disproportionate Brooksville site still going today, it would be more similar to Orlando instead of the I-10 (or even some of I-20) corridor. Eliminate the cold spot factors that the trades can't help, and you get something close to tropical with the lows to match as Orlando is. 

[flplantguy]

@Can't think of usernamethe wind roses match my observations here so far too.  A lot of northeast wind (or calm wind) at night and the west wind during the day kicks in fast. Combine the calm nights and lack of drainage in cold holes and there are a lot of cold spots around.  The airport is at a low spot just east from the veterans exptressway and west of the brooksville ridge.  I think the air movement into the cold air drainage areas is dammed up in low spots and the expressway catches it too somewhat.  It slowly flows towards the water as a land breeze overnight and drops the temp at the coast right at dawn.  Inland is able to hold the cold all night and gets much colder.  If the pressure ridge was not right overhead after a cold front there is always air povement, so only the hours it sits over us calm are the danger zone if they are at night.  That wind is always coming from the coldest spots inland, which are sandy areas that used to run all the way to near Tampa and Tarpon Springs from Ocala and north.  I belive this sand area with little no wetlands is the cause of the colder air spot along the west side of the peninsula.  It matches similar areas inland that crash on radiational nights as well.  The area near that airport is not yet heavily built up, so had a major difference during the 2010 long track freeze vs the freeze this past christmas.  The radiational freezes are worse there and more frequent, so the christmas freeze was like nothing (for the most part) and tampa was only 3 degrees or so warmer than brooksville.  January was colder in brooksville but tampa was fine like usual, since it was radiatonal and the urban heat island is the antithesis or dry sand on a cold night it seems.  The brooksville airport will likely warm too if development continues around it, but not by much i dont think.  Just west a mile or two are surviving foxtails that would have no chance by the airport (those all died last year like every year) and close to the water are happy mango trees and surviving coconuts, so its highly varied too.  Most wont make it but it helps show how the microclimates form.

Edited June 20, 2023 by flplantguy
typo

[Can't think of username]

7 hours ago, flplantguy said:

@Can't think of usernamethe wind roses match my observations here so far too.  A lot of northeast wind (or calm wind) at night and the west wind during the day kicks in fast. Combine the calm nights and lack of drainage in cold holes and there are a lot of cold spots around.

RIP💀.  Yeah, I really wish they didn't put the official weather stations at such low altitudes - makes for a very poor monitoring season most years as well as a misleading impression.

8 hours ago, flplantguy said:

The airport is at a low spot just east from the veterans exptressway and west of the brooksville ridge.  I think the air movement into the cold air drainage areas is dammed up in low spots and the expressway catches it too somewhat.  It slowly flows towards the water as a land breeze overnight and drops the temp at the coast right at dawn.  Inland is able to hold the cold all night and gets much colder.  If the pressure ridge was not right overhead after a cold front there is always air povement, so only the hours it sits over us calm are the danger zone if they are at night.

Quite shocking that the expressway exacerbates it, you would expect urbanization of that size to increase the minima somewhat🤯. Easy to see how this creates that strong cold bias, it just sucks so much that they shut down the Chinsegut Hill station which helped alleviate that problem.

8 hours ago, flplantguy said:

That wind is always coming from the coldest spots inland, which are sandy areas that used to run all the way to near Tampa and Tarpon Springs from Ocala and north.  I belive this sand area with little no wetlands is the cause of the colder air spot along the west side of the peninsula.  It matches similar areas inland that crash on radiational nights as well. 

I am fairly familiar with those cold spots you are referring to. They are all variable (probably because they are not all cold spots to the same degree), but here are the ones I have seen aside from Brooksville. All have high diurnal ranges and shockingly bad monitoring at a greater frequency than would be expected and that's a pretty dead ringer for a cold spot.

-Inverness. Surprisingly though, not as prone to cooling as KBKV despite colder average lows - for example, they got a January 17 last frost this season and never got below 2C all March, while KBKV hit 0C in the March 21 radiational cooling event.
-Ocala (as you mentioned). I have always seen it maintain good monitoring performance even with radiational cooling, but there have been times where it dropped the bomb KBKV-style - October 20 1989 and April 1 2003 first/last frosts come to mind. The authors of the Tallahassee studies suspect it's got the same problem as KTLH.
-Cross City. I have seen it really drop the bomb - October 20 2022 first frost on a date with an average low close to 16C. This place is another suspected by the authors of the Tallahassee study to have the same problem and I can most definitely see why, if it's not due to a low altitude site then that's worse than TLH even.
-Mayo. Dropped the bomb March 21 this year and nearly dropped it on October 20 last year.
-Live Oak. Has a record low of -3C(!!!!!) in October and although I can't confirm this, I have seen 1 news article that said it got to 0C on October 20 2022.
-Archbold Biological Station. I have been lucky enough to never watch it drop the bomb for first and last frost monitoring (although it has multiple times in November, March, and even April) but I have seen it get to -9C(!!!!!!!) this past January, for reference the average January low is 8.8C and is at the latitude of the Cfa/tropical transition🤯💀. The cause of this is extremely, extremely, extremely dry Lake Wales ridge soil.

With the natural much wetter ground and forest covering there's no WAY these sites could even dream of their present bad performances, that's for sure.
 

9 hours ago, flplantguy said:

The area near that airport is not yet heavily built up, so had a major difference during the 2010 long track freeze vs the freeze this past christmas.  The radiational freezes are worse there and more frequent, so the christmas freeze was like nothing (for the most part) and tampa was only 3 degrees or so warmer than brooksville.  January was colder in brooksville but tampa was fine like usual, since it was radiatonal and the urban heat island is the antithesis or dry sand on a cold night it seems.  The brooksville airport will likely warm too if development continues around it, but not by much i dont think.  Just west a mile or two are surviving foxtails that would have no chance by the airport (those all died last year like every year) and close to the water are happy mango trees and surviving coconuts, so its highly varied too.  Most wont make it but it helps show how the microclimates form.

Spot on about the first part. As normal on Christmas as Brooksville may have been compared to Tampa, it was hit hard again with the exact same -4C low in January that Tampa did not get any -1C like Christmas due to as you mention the radiation.

I think the fact that the radiational cooling is a consistent mainstay that FAR outweights advective cold means planting at high altitudes/moist ground/with plenty of other vegetative cover, and relying on wunderground stations (WeatherSTEM, etc) that have the appropriate altitude at least to not be cold spots is the way to go. 2010 is a testament to this, I looked at the data and KBKV literally reached -10C compared to -6C at Chinsegut!!!!!!!🤯💀

[flplantguy]

I use all the local stations for my job as well.  Those and street views can show how stark the changes are. One home can grow apples peaches and butias that look good, while a mile away its spindles and happy queen palms with more tropical plants that would die near the airport.  I have people wanting mango trees in those spots when they really should grow things you see in the panhandle.  I think its opportunity for the area to have markets for different things, and the spotty native trees from both north and south florida in the area show what can be done.  No one has caught on to that so you dont see local grown plants much.  It has been noted before how florida used to be phoenix palms and all cold tolerant species even further south because of this generalization.  Its too varied to map out but the center of shady hills will get frost from october to march every year and spots are colder than the airport even.  Then Holiday and NPR has mangoes with fruit most years.  Makes for a fun job when you tell someone their stunning adonidia will not survive long where they are and they should grow something they saw up north.  It doesnt go over well lol

[flplantguy]

The article below is another cool bit of information on the postion of the anticyclone and advective freezes vs radiational.  the 1962 event that slammed the west coast is slightly different from the others in position of the high (southern louisiana) bringing in the coldest air to that area.  outside of that most of the worst freezes for the rest of the state land in texas first as a large 1045mb high (or higher) then move northeast as a baroclinic low forms to pull it away.  the exceptions are long track events, and the high sits just north of the state for days and gives us that cold air inflow from the north and northeast.  if the low forms in just the right spot the cold makes to to south florida after getting whipped around the pressure centers, but the trajectory needs to be precise. Not so once you reach north florida and those same positions lead to the wind flow patterns seen in freezes that regularly slam the northwest peninsula. The last two in 2021 and last december landed in texas, but the low formation was not near the southeast, and the ridge in the atlantic was stronger, so the cold was blocked and/or moderated.  If the hadley cell is expanding this could be the new norm but more data is needed.  i will try to find the study about the hadley cell and pollution effects on its position next since i think that contributes as well.  

Rogers, Jeffrey C., and Robert V. Rohli. “Florida Citrus Freezes and Polar Anticyclones in the Great Plains.” Journal of Climate, vol. 4, no. 11, 1991, pp. 1103–13. JSTOR, http://www.jstor.org/stable/26196403. Accessed 20 June 2023.

[flplantguy]

https://www.researchgate.net/publication/258787600_Anthropogenic_sulfate_aerosol_and_the_southward_shift_of_tropical_precipitation_in_the_late_20th_century

This is the article about the tropical rain belt.  The coincidence with the pattern and 80s freezes in florida got me digging deeper.  Its many variables that factor in for freezes, but this made me think about the other patterns discussed here and how the different factors can add up to show the wild fluctuations we see.  What it means for the future i dont know, and i doubt many places will benefit, but i think the pattern will change slightly with the temperature readings showing a similar pattern shift.  If that happens it would lend weight to the theory about the 80s freezes being an oddity.  There is evidence that the 1800s had a similar shift but the data is missing as to why (i need to find that info too).  Soot from coal maybe from europe and the northeast, but it seems dicy and is all coincidence.  The change in air quality after the world wars could be the space of time in between, less soot and cooling with improved industrial tech, but not much data to show anything.  And at this point does that era matter after we have changed so much land?

[Can't think of username]

@flplantguySorry for the late response. I've been thinking long and hard about the tropical rain belt article and I still don't know what to say. Just filled with uncertainty for my weather monitoring.

As for your Jstor comment, all of this together with the Appalachians sometimes diverting intense cold to their west seems to play into some hella wacky cold distribution during very advective cold fronts. Christmas last year comes to mind.
Houston and Gainesville, at the same latitude, got clobbered very differently. Houston (Bush Airport) got to -9C (which is 6C colder than its usual coldest low), while Gainesville airport got to -5C (which is pretty much exactly as cold as expected for the coldest of the season, usual is -5.1C), in spite of the latitude necessitating equal cold all else being equal. And Wilmington airport got to only -7C despite the high latitude🤯, that is 1C warmer than the usual low and even warmer than Houston Hobby (-8C).

Mind blown what those factors do.