Jump to content

Florida Freeze and Weather Station Data


kinzyjr

Recommended Posts

NOAA makes weather records from the Florida stations available to download for free.  Out of the 1,700+ stations, 358 contain at least some temperature data.  If one were to download all of these records, import them into a database, and use some crafty SQL queries to generate sheets for each of the impact freezes it would provide a really good side-by-side resource to compare the microclimates in each region of Florida.

That is exactly what has been done with the 0000_202011040720_F_SQL_v2.xlsx spreadsheet attached to this post.  Each tab in the spreadsheet contains records for one of our unfortunate cold events.  The records come sorted first by the TAG column, which represents one of the areas in the pictures and is defined by a set of latitude and longitude boundaries.  A second sort is by the station name alphabetically.  This gives you a region of weather stations sorted alphabetically that allows you to see the temperatures in the region beside each other.

The lists are able to be filtered or sorted in any way you choose, so if you are only interested in weather stations in a particular region or set of regions, this is easily accomplished.  For easy viewing, the rows for each region alternate in shading.  This is easy to remove or change if you wish.

If you would like to see the boundaries of a region, the weather stations used with a link to their corresponding Weather Underground station, or a listing of the freezes and some commentary lifted from Florida Citrus Mutual’s website, or a description of the station location, the 202007121300_NOAA_WeatherStations_TemperatureOnly.xlsx sheet will have a plethora of this information.

For anyone who likes to look at the various airport weather stations on Wunderground, 202004292350_AirportWeatherStations.xlsx will give you as complete listing of these stations as I could assemble.

Now for a description of the various TAGs:

PEN = Pensacola area

01_PEN.png

PAN - Panama City Area

02_PAN.png

TAL - Tallahassee Area

03_TAL.png

EPN - Eastern Panhandle

04_EPN.png

CNF - Central North Florida

05_CNF.png

JAX - Jacksonville area

06_JAX.png

NEF - Northeast Florida - Dayona + St. Augustine and surrounding area

07_NEF.png

NWC - Northwest Central Florida

08_NWC.png

SWC - Southwest Central Florida

09_SWC.png

NIC - North Inland Central Florida

10_NIC.png

SIC - South Inland Central Florida

11_SIC.png

ECF - East Central Florida

12_ECF.png

SWF - Southwest Florida

13_SWF.png

SEF - Southeast Florida

14_SEF.png

MUK - Miami and the Upper Keys

15_MUK.png

KEY - The South and Western Florida Keys

16_KEY.png

0000_202011040720_F_SQL_v2.xlsx 202007121300_NOAA_WeatherStations_TemperatureOnly.xlsx 202004292350_AirportWeatherStations.xlsx

  • Like 7
  • Upvote 3

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

This is fantastic - I’ve only searched through SW and SWC Florida, but the data really helps visualize the great microclimate of St. Petersburg. Often the city stays as warm as Naples and Fort Meyers and is consistently the warmest in the SWC designation. Wish there was a station on Anna Maria.  Again, great job on compiling! 

  • Like 2
Link to comment
Share on other sites

  • 5 weeks later...

After seeing @pj_orlando_z9b's breakdown of days below 40F: ( https://www.palmtalk.org/forum/index.php?/topic/66579-florida-freeze-watch-december-2020/&page=6&tab=comments#comment-970076) , I decided I'd generate the results for all of the NOAA weather stations.  Download the attached sheet for the results.

202012281630_DaysUnder40.xlsx

  • Like 3

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

1 hour ago, kinzyjr said:

After seeing @pj_orlando_z9b's breakdown of days below 40F: ( https://www.palmtalk.org/forum/index.php?/topic/66579-florida-freeze-watch-december-2020/&page=6&tab=comments#comment-970076) , I decided I'd generate the results for all of the NOAA weather stations.  Download the attached sheet for the results.

202012281630_DaysUnder40.xlsx 328.37 kB · 0 downloads

Nice work!

I was thinking this might be a good alternative to standard hardiness zones. Particularly since tropical plants don't do well for extended periods below 40f. 

  • Like 3

Westchase | 9b 10a  ◆  Nokomis | 10a  ◆  St. Petersburg | 10a 10b 

Link to comment
Share on other sites

I wish someone a lot more computer savvy than me would do that for South Texas and the Rio Grande Valley, and make a corresponding Climate Zone Map of the area as a result of it.

John

  • Like 1
Link to comment
Share on other sites

1 hour ago, Mr. Coconut Palm said:

I wish someone a lot more computer savvy than me would do that for South Texas and the Rio Grande Valley, and make a corresponding Climate Zone Map of the area as a result of it.

John

If you can get a group of people onboard to help you, downloading the data from NOAA and setting up a database to hold the imports is probably the most important part.  There are over 5,000 stations in Texas.  There are probably at least 1,000 that hold MAX and MIN temperature data. 

Any possibility of enlisting some volunteers from the Palm Society of South Texas, the Houston IPS affiliate, or getting any of the students at the universities to help out?

I can share the tools and the methods I used to import, select, and sanitize the data.  Basically, Microsoft Excel and SQLite database were the predominant tools used.  There's a table that holds all of the stations, using the NOAA label (ex. GHCND:USC00410012) as the primary key (can't be a duplicate).  The temperature data is uploaded with the NOAA label, high, low, and precipitation data.  Since both tables use the NOAA label, you can basically join them together and queries against them.  That's where having a techie onboard comes in handy.

Here are some helpful hints for anyone or any group(s) of people wanting to do the data collection and analytics:

To get the data for a state:

  1. Access the NOAA Search Tool currently at this website: https://www.ncdc.noaa.gov/cdo-web/
  2. Click on SEARCH TOOL under the DISCOVER DATA BY section
  3. Fill out the search using the info in Figure 1 (ex. is for TX) and click Search. 
    1. Note the date range should be from the earliest possible date to the latest to get the most data.
  4. When the screen comes up for the state you chose, click on the State name in blue (it's a hyperlink).
    1. See Figure 2
  5. You should come to a screen where it gives you a summary of all of the stations in the data set.
  6. Scroll down and click the Station List (Figure 3).
    1. Yes, there really are 5308 stations
  7. When you find a station that has the data you want, you can click on it and then click ADD TO CART.

Figure 1: Search criteria

image.png.d518d5c658d26cdc1bf244b24a1cab87.png

 

Figure 2: The whole state with the blue link

image.png.1202d9482777c3e0ce82549ba8ce5e05.png

Figure 3: Station List

image.png.71e889781840109536c8570919a6e68d.png

How to tell if a station may have temperature data:

  1. Look at Figure 3 above and ask what is different about the one I put the red box around from the rest.
    1. You should notice that all of the others have decimals.
    2. The ones with decimals typically do not have temperature data.
    3. The ones without numbers or with whole numbers typically DO have temperature data.
      1. Examples:
      2. ABERNATHY, TX US doesn't have decimals or even a number.  It likely has data.
      3. LAKELAND 2, FL, US has a number, but it is a whole number.  It likely has data.
      4. ABILENE 3.3 SW, TX US has a decimal.  It likely does not have data.

What information should I capture about each NOAA station?:

  1. In a word, everything.  You need the Network ID to use as the NOAA_ID in your database.  I recommend capturing the name, latitude and longitude, elevation, start date and end date.  In my table, I also added special notes, a region tag, and a column where I could add a link to the corresponding weather underground station.

image.png.6505b0b7288ffad2b5a8569e31631d0e.png

What information should I capture in each weather record?:

  1. The minimum you'll need to capture are the weather station, date, high, and low.   I also capture the precipitation in my records.  I typically download everything and then just filter out what I don't want.

How do I import data into SQLite:

  1. Online Documentation: https://www.sqlitetutorial.net/sqlite-import-csv/
  2. I'd recommend using the Import a CSV file into a table using SQLite Studio section of this documentation.

Some issues you may run into (and ways to deal with them):

  1. Bogus data:
    1. Ever see -100F in Texas in June?  Yeah, not here in Florida either.  It's difficult to correct negatives as there could have been negative temperatures in areas of the state; so just multiplying by -1 doesn't always change it to the correct value.  You pretty much have to look for all of the negatives, and decide which ones to correct on a case by case or group by group basis.
    2. Blanks: When you query, you can filter these out with a WHERE condition this: WHERE LOW <> ''.  This keeps them from polluting your AVG and MIN functions when finding record lows or average lows.
    3. Ridiculous Numbers: Did you know it was 3F in South Florida one time?  Yeah, me either.  When running queries, the TAG regions I set up allowed me to set constraints for that region on what the low temperature was allowed to be.  That helped significantly in filtering out obviously spurious data.
  2. Missing or Incomplete data:
    1. When you download data for a large metro area, sometimes the reporting station changes several times over a century or so.  Your only option for contiguous data is to merge them together or set date ranges to select from one station if you are getting data from a set of years in a range, but another station with a different date range.  The start and end dates on the NOAA station records can help with this.
  3. Data that fails to import:
    1. 99.9% of the time, it's because the file is formatted wrong.  They need to be saved as CSV for the process above.  Sometimes Excel will save a ton of blank rows under the actual data that will fail any constraints set up on the data.
  4. Database crashes:
    1. Always know where the database file is stored and make copies of it frequently.  Put the copies somewhere safe.  In my case, I have several jump drives with historical copies as well as having copies on mirrored drives.  Redoing work is not a lot of fun.

This is probably enough for anyone or any group that wants to tackle the endeavor for a different state.  If anyone has questions, post them and I'll try to answer them timely.

  • Like 3

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

18 hours ago, kinzyjr said:

If you can get a group of people onboard to help you, downloading the data from NOAA and setting up a database to hold the imports is probably the most important part.  There are over 5,000 stations in Texas.  There are probably at least 1,000 that hold MAX and MIN temperature data. 

Any possibility of enlisting some volunteers from the Palm Society of South Texas, the Houston IPS affiliate, or getting any of the students at the universities to help out?

I can share the tools and the methods I used to import, select, and sanitize the data.  Basically, Microsoft Excel and SQLite database were the predominant tools used.  There's a table that holds all of the stations, using the NOAA label (ex. GHCND:USC00410012) as the primary key (can't be a duplicate).  The temperature data is uploaded with the NOAA label, high, low, and precipitation data.  Since both tables use the NOAA label, you can basically join them together and queries against them.  That's where having a techie onboard comes in handy.

Here are some helpful hints for anyone or any group(s) of people wanting to do the data collection and analytics:

To get the data for a state:

  1. Access the NOAA Search Tool currently at this website: https://www.ncdc.noaa.gov/cdo-web/
  2. Click on SEARCH TOOL under the DISCOVER DATA BY section
  3. Fill out the search using the info in Figure 1 (ex. is for TX) and click Search. 
    1. Note the date range should be from the earliest possible date to the latest to get the most data.
  4. When the screen comes up for the state you chose, click on the State name in blue (it's a hyperlink).
    1. See Figure 2
  5. You should come to a screen where it gives you a summary of all of the stations in the data set.
  6. Scroll down and click the Station List (Figure 3).
    1. Yes, there really are 5308 stations
  7. When you find a station that has the data you want, you can click on it and then click ADD TO CART.

Figure 1: Search criteria

image.png.d518d5c658d26cdc1bf244b24a1cab87.png

 

Figure 2: The whole state with the blue link

image.png.1202d9482777c3e0ce82549ba8ce5e05.png

Figure 3: Station List

image.png.71e889781840109536c8570919a6e68d.png

How to tell if a station may have temperature data:

  1. Look at Figure 3 above and ask what is different about the one I put the red box around from the rest.
    1. You should notice that all of the others have decimals.
    2. The ones with decimals typically do not have temperature data.
    3. The ones without numbers or with whole numbers typically DO have temperature data.
      1. Examples:
      2. ABERNATHY, TX US doesn't have decimals or even a number.  It likely has data.
      3. LAKELAND 2, FL, US has a number, but it is a whole number.  It likely has data.
      4. ABILENE 3.3 SW, TX US has a decimal.  It likely does not have data.

What information should I capture about each NOAA station?:

  1. In a word, everything.  You need the Network ID to use as the NOAA_ID in your database.  I recommend capturing the name, latitude and longitude, elevation, start date and end date.  In my table, I also added special notes, a region tag, and a column where I could add a link to the corresponding weather underground station.

image.png.6505b0b7288ffad2b5a8569e31631d0e.png

What information should I capture in each weather record?:

  1. The minimum you'll need to capture are the weather station, date, high, and low.   I also capture the precipitation in my records.  I typically download everything and then just filter out what I don't want.

How do I import data into SQLite:

  1. Online Documentation: https://www.sqlitetutorial.net/sqlite-import-csv/
  2. I'd recommend using the Import a CSV file into a table using SQLite Studio section of this documentation.

Some issues you may run into (and ways to deal with them):

  1. Bogus data:
    1. Ever see -100F in Texas in June?  Yeah, not here in Florida either.  It's difficult to correct negatives as there could have been negative temperatures in areas of the state; so just multiplying by -1 doesn't always change it to the correct value.  You pretty much have to look for all of the negatives, and decide which ones to correct on a case by case or group by group basis.
    2. Blanks: When you query, you can filter these out with a WHERE condition this: WHERE LOW <> ''.  This keeps them from polluting your AVG and MIN functions when finding record lows or average lows.
    3. Ridiculous Numbers: Did you know it was 3F in South Florida one time?  Yeah, me either.  When running queries, the TAG regions I set up allowed me to set constraints for that region on what the low temperature was allowed to be.  That helped significantly in filtering out obviously spurious data.
  2. Missing or Incomplete data:
    1. When you download data for a large metro area, sometimes the reporting station changes several times over a century or so.  Your only option for contiguous data is to merge them together or set date ranges to select from one station if you are getting data from a set of years in a range, but another station with a different date range.  The start and end dates on the NOAA station records can help with this.
  3. Data that fails to import:
    1. 99.9% of the time, it's because the file is formatted wrong.  They need to be saved as CSV for the process above.  Sometimes Excel will save a ton of blank rows under the actual data that will fail any constraints set up on the data.
  4. Database crashes:
    1. Always know where the database file is stored and make copies of it frequently.  Put the copies somewhere safe.  In my case, I have several jump drives with historical copies as well as having copies on mirrored drives.  Redoing work is not a lot of fun.

This is probably enough for anyone or any group that wants to tackle the endeavor for a different state.  If anyone has questions, post them and I'll try to answer them timely.

Kinzy,

Thank you.  I will try to find someone over here who can do it.  Your suggestion of getting someone in the Palm Society of South Texas is a good idea if any of them are computer savvy, especially since we are not having meetings now due to the Covid situation.  This would be a good project for someone in the group that would really benefit all of us.

John

  • Like 1
Link to comment
Share on other sites

On 1/2/2021 at 7:29 PM, Mr. Coconut Palm said:

Kinzy,

Thank you.  I will try to find someone over here who can do it.  Your suggestion of getting someone in the Palm Society of South Texas is a good idea if any of them are computer savvy, especially since we are not having meetings now due to the Covid situation.  This would be a good project for someone in the group that would really benefit all of us.

John

Hi John, here is a very rough map of zone 9 and 10 in Texas (based on the numbers only): 

texaszonemap.jpg.8a00da67a6fb3f689ded33b7e6ad8a69.jpg

  • Like 3
  • Upvote 2

Jonathan

Katy, TX (Zone 9a)

Link to comment
Share on other sites

13 minutes ago, Xenon said:

Hi John, here is a very rough map of zone 9 and 10 in Texas (based on the numbers only): 

texaszonemap.jpg.8a00da67a6fb3f689ded33b7e6ad8a69.jpg

Great map Jonathan!  Thanks for developing it.  Very interesting to see the small area of 10a out in Starr and Zapata counties as well as areas in Galveston/Houston.  

  • Like 2

Clay

South Padre Island, Zone 10b until the next vortex.

Link to comment
Share on other sites

  • 2 weeks later...
On 1/4/2021 at 12:44 AM, Xenon said:

Hi John, here is a very rough map of zone 9 and 10 in Texas (based on the numbers only): 

texaszonemap.jpg.8a00da67a6fb3f689ded33b7e6ad8a69.jpg

Hi Jonathan,

Thanks.   That is great! Great job.  It looks like Zone 10A extends all the way up to the backside of Matagorda Island, which makes sense since I have heard that there are some Red Mangroves growing on the backside (bayside) of Matagorda Island.

John

  • Like 1
Link to comment
Share on other sites

On 1/4/2021 at 12:44 AM, Xenon said:

Hi John, here is a very rough map of zone 9 and 10 in Texas (based on the numbers only): 

texaszonemap.jpg.8a00da67a6fb3f689ded33b7e6ad8a69.jpg

Jonathan,

Did you crunch numbers for the entire state and then only show areas that were zone 9 or higher; or did you crunch numbers from just a portion of the state and show the results of that effort?  I ask because I am a little surprised that there is not a 9a UHI in Austin.  I was noticing a few more tropical things surviving in the downtown area before I moved south.  Just curious.  Thanks for the effort!  Am also curious how possible it is to get better resolution for South Texas.  Does the 10b extend south to Boca Chica an possibly inland a bit? Not much development in that area, so getting data is difficult I expect.  The same can be said for many areas of South Texas.  Probably why I like it.  :)

  • Like 1
  • Upvote 1

Clay

South Padre Island, Zone 10b until the next vortex.

Link to comment
Share on other sites

1 hour ago, Austinpalm said:

Jonathan,

Did you crunch numbers for the entire state and then only show areas that were zone 9 or higher; or did you crunch numbers from just a portion of the state and show the results of that effort?  I ask because I am a little surprised that there is not a 9a UHI in Austin.  I was noticing a few more tropical things surviving in the downtown area before I moved south.  Just curious.  Thanks for the effort!  Am also curious how possible it is to get better resolution for South Texas.  Does the 10b extend south to Boca Chica an possibly inland a bit? Not much development in that area, so getting data is difficult I expect.  The same can be said for many areas of South Texas.  Probably why I like it.  :)

I only looked for where to draw the zone 9 and 10 borders. 

I agree with you about 9a UHI in Austin; I was lazy and mostly used airport data and the stations available from local NWS pages rather than the whole NCDC archive (not a huge difference). Camp Mabry is definitely 9a, southern and southeastern Travis county straddles the 8b/9a border. 

There is very little data available for coastal Cameron county. I think 10b is limited only to SPI and areas with the Laguna Madre directly to the north and northwest like Port Isabel (and probably Brazos Island/Boca Chica though South Bay is shallow even relative to the LM). Bayfront areas without water to the N/NW like Laguna Vista run a bit colder and KPIL sometimes runs even colder than Brownsville being in a rural area despite being fairly close to the coast. Port Mansfield is not very warm either, about on par with Harlingen. 

If I had to pick the "one warmest spot in TX", it would probably be the north end of Brazos Island with the ship channel directly due north and the deepest part of the Laguna Madre (near the causeway) to the northwest. 

Edited by Xenon
  • Like 2

Jonathan

Katy, TX (Zone 9a)

Link to comment
Share on other sites

8 hours ago, Xenon said:

I only looked for where to draw the zone 9 and 10 borders. 

I agree with you about 9a UHI in Austin; I was lazy and mostly used airport data and the stations available from local NWS pages rather than the whole NCDC archive (not a huge difference). Camp Mabry is definitely 9a, southern and southeastern Travis county straddles the 8b/9a border. 

There is very little data available for coastal Cameron county. I think 10b is limited only to SPI and areas with the Laguna Madre directly to the north and northwest like Port Isabel (and probably Brazos Island/Boca Chica though South Bay is shallow even relative to the LM). Bayfront areas without water to the N/NW like Laguna Vista run a bit colder and KPIL sometimes runs even colder than Brownsville being in a rural area despite being fairly close to the coast. Port Mansfield is not very warm either, about on par with Harlingen. 

If I had to pick the "one warmest spot in TX", it would probably be the north end of Brazos Island with the ship channel directly due north and the deepest part of the Laguna Madre (near the causeway) to the northwest. 

That looks pretty accurate. Based on my observations I’d push the zone 10 further up Galveston bay. At least if you’re within a mile of the water I’d say the zone 10 stops at the Fred Hartman bridge. Also the lake front neighborhoods around Clear Lake would be 10a. San Leon, specifically Clifton Beach, has a few weather stations that are often warmer than Galveston. There are some good looking, really tall, Norfolk Island pines in that area. 

Link to comment
Share on other sites

3 hours ago, Jeff985 said:

That looks pretty accurate. Based on my observations I’d push the zone 10 further up Galveston bay. At least if you’re within a mile of the water I’d say the zone 10 stops at the Fred Hartman bridge. Also the lake front neighborhoods around Clear Lake would be 10a. San Leon, specifically Clifton Beach, has a few weather stations that are often warmer than Galveston. There are some good looking, really tall, Norfolk Island pines in that area. 

Yes, I noted the southern shore of Galveston Bay in the map. Kemah and lakefront Clear Lake is where I would place the cutoff though.  Large Norfolk Pine is actually one of the key things I look for in a "numerical 10a" climate for areas that don't have weather stations. 

Jonathan

Katy, TX (Zone 9a)

Link to comment
Share on other sites

6 hours ago, Xenon said:

Yes, I noted the southern shore of Galveston Bay in the map. Kemah and lakefront Clear Lake is where I would place the cutoff though.  Large Norfolk Pine is actually one of the key things I look for in a "numerical 10a" climate for areas that don't have weather stations. 

There’s a decent looking Norfolk in La Porte's Shady River neighborhood, which is about 1/2 mile inland. It does have some overhead canopy. Google street view images are from 2011 and it wasn’t there then, but it is pre 2018. I’ll try to get a picture the next time I drive through that neighborhood. There was also one on Meadowlawn Street in Shoreacres that was damaged in 2018 but survived. Apparently the owner didn’t like the way it looked after the freeze since it was removed about a year later. 

Link to comment
Share on other sites

On 1/17/2021 at 6:18 AM, Jeff985 said:

There’s a decent looking Norfolk in La Porte's Shady River neighborhood, which is about 1/2 mile inland. It does have some overhead canopy. Google street view images are from 2011 and it wasn’t there then, but it is pre 2018. I’ll try to get a picture the next time I drive through that neighborhood. There was also one on Meadowlawn Street in Shoreacres that was damaged in 2018 but survived. Apparently the owner didn’t like the way it looked after the freeze since it was removed about a year later. 

@XenonHere’s the Shady River La Porte Norfolk. I also found one in Morgan’s Point, but I’m not sure that one was there prior to 2018. 

FF2AA307-25E3-49B1-9022-7308415449C8.png

  • Upvote 1
Link to comment
Share on other sites

  • 3 weeks later...

@Austinpalm

I patched together some fragmented data from NCDC and Wunderground (recent years) as well as cross examined the data with readings from Brownsville and Bayview/Port Isabel to get these high-confidence figures (lowest winter extreme min) for South Padre Island for the past ~20 years:

2020-2021: 40F
2019-2020: 37F
2018-2019: 42F
2017-2018: 32F
2016-2017: 31F
2015-2016: 43F
2014-2015: 37F
2013-2014: 32F
2012-2013: 43F
2011-2012: 43F
2010-2011: 28F
2009-2010: 32F
2008-2009: 42F
2007-2008: 43F
2006-2007: 35F
2005-2006: 39F
2004-2005: 28F* missing data, conservative estimate
2003-2004: 42F
2002-2003: 37F
2001-2002: 32F* missing data, conservative estimate 
2000-2001: 41F
1999-2000: 43F

 

  • Like 2

Jonathan

Katy, TX (Zone 9a)

Link to comment
Share on other sites

1998-1999: 36F
1997-1998: 37F
1996-1997: 32F*missing data, conservative estimate
1995-1996: 33F
1994-1995: 42F
1993-1994: 38F*missing data, conservative estimate
1992-1993: 41F

Average of 37.3F over 29 winter seasons 

conservative estimate means I used the figure for Brownsville or Corpus Christi NAS 

Edited by Xenon

Jonathan

Katy, TX (Zone 9a)

Link to comment
Share on other sites

On 2/3/2021 at 2:16 AM, Xenon said:

1998-1999: 36F
1997-1998: 37F
1996-1997: 32F*missing data, conservative estimate
1995-1996: 33F
1994-1995: 42F
1993-1994: 38F*missing data, conservative estimate
1992-1993: 41F

Average of 37.3F over 29 winter seasons 

conservative estimate means I used the figure for Brownsville or Corpus Christi NAS 

Great information Jonathan!

 

Thanks!!

Clay

South Padre Island, Zone 10b until the next vortex.

Link to comment
Share on other sites

  • 1 year later...

Many posters have contributed to the Make Your Own Zone Map! thread, sharing maps based on a variety of data sources and anecdotal observations.  While the USDA Hardiness Zone system of using the average annual low over 30 years (AAL-30) is a good starting point, it has its inherent flaws which have become more visible due to a string of warmer than average winters. 

There have been lively discussions around:

·        Modifications to the definition of a hardiness zone.

o   Is it better to use 50 years or 100 years of data rather than 30 years?

·        Which metric(s) should be considered?

o   Should we consider the number of days under certain temperature thresholds as well as the minimum temperature reached?

o   What about the high temperatures?  Is 70/30 the same as 49/30?

·        The holes in the available data.

o   The number of stations with more than 100 years of individual data are too few to map the entire state. 

o   Is there a way to compensate for the missing data to give a more accurate and holistic picture of the climate of a region?

These are questions where the number of answers based on individual preference varies significantly and could impact what they consider a hardiness zone.  For my own personal map, I used 50-year averages (AAL-50) since it still includes the record freezes of the 1980s.  The data set is easily modifiable and years pass and it will be another decade before the record freezes drop off the books, leaving 2010 as the landscape-defining freeze in Florida.

Some growers have a higher risk tolerance than others.  The following data may be of use to everyone across the spectrum.  For the purpose of this sheet, I defined 315 composite stations that are made up anywhere from 1 to 12 of the original 359 NOAA stations, ordered by priority.  The records for each composite were created by starting with a blank “shell” of station records and updating them in turn with each of the NOAA stations.  In this way, holes in the data are filled using a station nearby similar to what Weather.com does.  The difference is that you know the station it came from because it is recorded for you.

The attached file contains 5 sheets of information.  The first sheet is named StationStatistics_1.  This sheet contains the stations, their primary (first) NOAA station, and a plethora of statistics including the AAL for 30, 50, 100, and All years.  There are also record lows for the entire composite as well as the primary station, and the number of days where the low temperature was below 20, 30, and 40 degrees as well as the number of days where the high was below 30, 40, and 50 degrees.

The second sheet is named StationDefinition_1 and defines each composite as a collection of NOAA stations in order by priority.  This sheet also gives you a breakdown of the number and percentage of records taken from each NOAA station to form the composite station.

The third sheet is name CALs, which is short for Composite Annual Lows.  This will give you the year-by-year breakdown of the low for the year and the station that contributed it to the composite station.  The fourth sheet give the ExpandedZones definition I used for my map.  This breaks zones into 5 subzones (a-1, a-2, a-b, b-1, b-2)  vs. the standard two (a, b).  The last sheet is the definition of the TAG regions shown by the original post and is included for reference.

This sheet makes it easier to create your own Google map with the data points you prefer, as well as viewing the data I used to create my own.

2021_FloridaWeatherAlmanac_upl.xlsx

  • Upvote 1

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

This post is an update to the previous post.  Unfortunately, the spreadsheet attached to the previous post contains some errant data.  The attachment to this post corrects those errors and includes another data set based on the individual NOAA stations.  Please delete the old sheet and use this one for more accurate data.

There are 10 sheets of data contained in this file.  The StationStatistics_NOAA and StationStatistics_COMP sheets contain statistics for the individual NOAA stations and the composites stations I created from combinations of NOAA stations, respectively.

The StationYearCompletion_NOAA and StationYearCompletion_COMP sheets contain a breakdown of the number of daily records for each station vs. the maximum (366 for leap year, 365 otherwise) along with a  percentage showing how complete the daily records are for that station in the given year.

The NALs and CALs sheets contain the annual lows for each station sorted by the station name and year.

The StationDefinition_COMP sheet gives you the priority of the station combinations used to generate the composite station records.  For example, if a composite is made up of three stations, the one with the lowest priority number would add their records first and the second station would fill in any blanks left by the first.  The third station would then fill in any remaining blanks.  This sheet also shows a breakdown of the percentage of records for the composite station given by each individual station.

The Expanded_Zones tab was modified to 3 decimal places as the calculations for zones would leave gaps without that level of precision.  The TAGs sheet contains the latitude and longitude boundaries for each of the TAG regions used to compare stations within a geographical region.

There is also a template sheet at the end used to create additional sheets with the same formatting.

2021_FloridaWeatherAlmanac.xlsx

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

  • 4 weeks later...

An update that includes:

  • the available Jan. 2022 data
  • multiple new NOAA stations, including one for Captiva Island
  • a manufactured station for the Zephyrhills airport (KZPH) using scraped data
  • Updated metrics for the composite Lakeland stations using WUnderground data

2022_FloridaWeatherAlmanac.xlsx

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

For those who prefer the Google maps:

Florida Impact Freezes 00 (1835-1977): https://www.google.com/maps/d/edit?mid=18H1cufNcjuFbcjZEkhaOPAxwazmWVGc&usp=sharing
Florida Impact Freezes 01 (1981-2022): https://www.google.com/maps/d/edit?mid=1h2lbBtHdxlGNuyDTVo5A6xdrn4McSCQ&usp=sharing

  • Like 1

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

  • 1 month later...
On 5/22/2022 at 12:14 AM, kinzyjr said:

Many posters have contributed to the Make Your Own Zone Map! thread, sharing maps based on a variety of data sources and anecdotal observations.  While the USDA Hardiness Zone system of using the average annual low over 30 years (AAL-30) is a good starting point, it has its inherent flaws which have become more visible due to a string of warmer than average winters. 

 

There have been lively discussions around:

 

·        Modifications to the definition of a hardiness zone.

 

o   Is it better to use 50 years or 100 years of data rather than 30 years?

 

·        Which metric(s) should be considered?

 

o   Should we consider the number of days under certain temperature thresholds as well as the minimum temperature reached?

 

o   What about the high temperatures?  Is 70/30 the same as 49/30?

 

·        The holes in the available data.

 

o   The number of stations with more than 100 years of individual data are too few to map the entire state. 

 

o   Is there a way to compensate for the missing data to give a more accurate and holistic picture of the climate of a region?

I've got an idea for a metric hardiness zone system that is also a partial climate indicator, but not too complicated with too many superscripts and abbreviations that are translated from another language like German in the case of the Köppen system. 

The letters could change depending on language as they should with the Köppen system as well, since other than the first primary temperature gradient letter (A thru E), the dry summer/winter or fully humid middle letter should change in different languages, since it's just an abbreviation more than a code. 

For example the f in the Köppen system Cfa doesn't really mean "fully" humid it just means humid, translated from the German word for humid:  feuct. 

Edited by Aceraceae
Link to comment
Share on other sites

22 hours ago, Aceraceae said:

I've got an idea for a metric hardiness zone system that is also a partial climate indicator, but not too complicated with too many superscripts and abbreviations that are translated from another language like German in the case of the Köppen system. 

The letters could change depending on language as they should with the Köppen system as well, since other than the first primary temperature gradient letter (A thru E), the dry summer/winter or fully humid middle letter should change in different languages, since it's just an abbreviation more than a code. 

For example the f in the Köppen system Cfa doesn't really mean "fully" humid it just means humid, translated from the German word for humid:  feuct. 

Could you give an example of how a zoning system would work?  For example, what zone assignment for the following cities in Florida:

  • Tallahasee
  • Jacksonville
  • Gainesville
  • Daytona Beach
  • Tampa
  • Orlando
  • Saint Petersburg
  • Miami
  • Fort Myers
  • Like 1

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

Miami would be 2a-FD

@kinzyjr

The scale is every 5c (9f) or 2.5 c (4.5f) for the letter. Slightly more specific than the 5/10f USDA rating, but nowhere near as specific as some of the maps you've made that break down Fahrenheit beyond 5° ab zones into 2.5f (1.5c) zones. 

The numbers are plus and minus around the freezing point.

The F indicates that there has been a frost or freeze even though the Miami rating is plus two (mean min over 5c 41f). Another option would be to put the zone number of the record cold in place of the F, which would be 2a,-1aD since the all time low is -3c 27 f in the less famous 1917 Florida freeze. 

The D indicates that it's a mostly dry winter. 

Summer "hardiness" is not included as it would really only be relevant for extremely hot deserts or cold summer oceanic Cfc or equatorial highland mountains. This is a winter hardiness and partial climate indicator for cold season concerns. 

This kind of gives an all-in-one indication of the advantage of dry climates that have caused a misunderstanding of zone 6 palm/yucca/etc hardiness where in a Boston zone 7 (wet), needle palm and trachy die is much more harsh than an Albuquerque zone 5 (dry winter) where even washingtonia survive extremely brief overnight dips to zero f or -20c.  

Edited by Aceraceae
  • Like 1
Link to comment
Share on other sites

There are a lot of weather records from the Frost Warning Service and Florida Citrus Mutual available in the book A History of Florida Citrus Freezes.  The information includes duration below various temperature thresholds in some cases.  The attached sheet is a quick capture of the tabular information presented.

20220815_AHFCF_tables.xlsx

  • Upvote 1

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

  • 1 month later...

One of my biggest interests for this winter is to see how downtown Tallahassee's urban heat island compares to the official weather station at the cold hole airport (first and last freeze, number of freezes, coldest temperature of the season, etc).
According to Wikipedia, the heat island is an average of +3.2 Celsius and is especially pronounced in winter, so the differences ought to be very interesting

Edited by Can't think of username
  • Like 1
Link to comment
Share on other sites

  • 5 months later...

For those who would like a complete list of the NOAA Stations: https://www.ncei.noaa.gov/access/homr/reports

In my case, the MSHR Standard Version works well.

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

  • 4 weeks later...
On 7/2/2022 at 9:02 PM, kinzyjr said:

For those who prefer the Google maps:

Florida Impact Freezes 00 (1835-1977): https://www.google.com/maps/d/edit?mid=18H1cufNcjuFbcjZEkhaOPAxwazmWVGc&usp=sharing
Florida Impact Freezes 01 (1981-2022): https://www.google.com/maps/d/edit?mid=1h2lbBtHdxlGNuyDTVo5A6xdrn4McSCQ&usp=sharing

Currently limited to 10 layers, so the currently available statistics for the December 2022 Cold Snap are mapped here:

https://www.google.com/maps/d/edit?mid=1QwsArS8xqOQ1Bbw_4LqRrurwD6cf3Q8&usp=sharing

There may be a few more airport stations added in the future, and some of those stations may also find a place on some of the previous freeze maps back to1996.

There are also some nodes on this map with some suspect data.  The low of 39F at Mountain Lake in Lake Wales, the low of 24F at the Tarpon Springs Water Plant, and the two Plant City stations being 7F apart (25F vs. 32F).  As far as the overall data set goes, this was one of the worst datasets to sanitize out of all of the datasets I've generated.

  • Like 1
  • Upvote 2

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

On 1/16/2021 at 10:03 AM, Austinpalm said:

Jonathan,

Did you crunch numbers for the entire state and then only show areas that were zone 9 or higher; or did you crunch numbers from just a portion of the state and show the results of that effort?  I ask because I am a little surprised that there is not a 9a UHI in Austin.  I was noticing a few more tropical things surviving in the downtown area before I moved south.  Just curious.  Thanks for the effort!  Am also curious how possible it is to get better resolution for South Texas.  Does the 10b extend south to Boca Chica an possibly inland a bit? Not much development in that area, so getting data is difficult I expect.  The same can be said for many areas of South Texas.  Probably why I like it.  :)

Hello Jonathan,

 

Just curious how much the last 2 winters have affected the general zone boundaries.  Not trying to give you any more work.  Just curious if SPI/PI are still falling out as zone 10b or if they have dropped?

 

Thanks,

Clay

Clay

South Padre Island, Zone 10b until the next vortex.

Link to comment
Share on other sites

39 minutes ago, Austinpalm said:

Hello Jonathan,

 

Just curious how much the last 2 winters have affected the general zone boundaries.  Not trying to give you any more work.  Just curious if SPI/PI are still falling out as zone 10b or if they have dropped?

 

Thanks,

Clay

Hi Clay, 

I don't think anything has changed, Brownsville is still averaging around 33F for the last 30 years. SPI average is pulled up by all of the mild years above 40F (no advective freezes). North Padre is probably 10a. 

Hope the crappy two winters haven't dampened your gusto too much. Would love to see what you are planting (maybe on the main forum). 

Best,

Jonathan 

 

Jonathan

Katy, TX (Zone 9a)

Link to comment
Share on other sites

On 4/17/2023 at 6:31 PM, kinzyjr said:

Currently limited to 10 layers, so the currently available statistics for the December 2022 Cold Snap are mapped here:

https://www.google.com/maps/d/edit?mid=1QwsArS8xqOQ1Bbw_4LqRrurwD6cf3Q8&usp=sharing

There may be a few more airport stations added in the future, and some of those stations may also find a place on some of the previous freeze maps back to1996.

There are also some nodes on this map with some suspect data.  The low of 39F at Mountain Lake in Lake Wales, the low of 24F at the Tarpon Springs Water Plant, and the two Plant City stations being 7F apart (25F vs. 32F).  As far as the overall data set goes, this was one of the worst datasets to sanitize out of all of the datasets I've generated.

I’m pretty sure that 24f reading in Tarpon Springs is wrong. It should be at least as warm as my area and I had 30f.

Practically speaking, I’m not sure what to do to correct that. 

Edited by RedRabbit
  • Upvote 1

Westchase | 9b 10a  ◆  Nokomis | 10a  ◆  St. Petersburg | 10a 10b 

Link to comment
Share on other sites

7 minutes ago, RedRabbit said:

I’m pretty sure that 24f reading in Tarpon Springs is wrong. It should be at least as warm as my area and I had 30f.

Practically speaking, I’m not sure what to do to correct that. 

I think one of a few things is true.  Either that station is in a "cold spot" or it is just reading way too low.  I've seen it happen here as well.  The best thing you can do is attempt to do the Blockchain method and come up with a number by consensus - usually take the nearest and best-sited Wunderground stations around it and grab their numbers.  Then toss out the highest and lowest readings and average the rest.

This happened back in 2020 as well.
https://www.palmtalk.org/forum/topic/66736-new-climate-normals-1991-2020/?do=findComment&comment=999898

I'd imagine that area was realistically between 30F-32F.  You had 30F and Clearwater Air Park had 32F, so that is how I formed my estimate.  I'm working on condensing the data for the airports that do not store records any longer and adding them to the map.  Perhaps one of these airports (or a consensus between them) will guide us to a more appropriate measure.

  • Like 1
  • Upvote 1

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

December 2022 (and next 9 events): https://www.google.com/maps/d/edit?mid=1QwsArS8xqOQ1Bbw_4LqRrurwD6cf3Q8&usp=sharing

A few updates to this particular map:

  • Added the highs and lows parsed from the NWS html files uploaded to the N. America Winter 2022-2023 thread in ZIP format.
    • Note that some of these were added using a USZ######### NOAA ID if they did not have a NOAA station number at any point.  Ex. Zephyrhills Municipal = USZ0000KZPH
    • For the stations that had an old ID but stopped keeping records, I resurrected their ID for this map. Ex. Tampa Executive (Vandenburg) = USW00092802
    • Since there are multiple station IDs for the Lakeland Airport (KLAL), I used one to capture the low of 28F in the hourly NWS records (USW00012883)
  • The spreadsheet used to generate the Google Map is available here as: 202304192315_2022_DecemberColdSnap.xlsx
  • The spreadsheet used to parse the NWS records is available here as: 202304180400_NWS_Parse.xlsx
  • The records gleaned from the parsing of the NWS are available here as: 202304192335_NWS_dump.xlsx
  • The addition of the New Smyrna Airport gives us a better idea of the low temperature than the incomplete records at the New Smyrna Beach Marine Discovery Center
  • The Crystal River Airport agrees with the reports I received of 27F-28F in the area.
  • Of interest to @RedRabbit: The Clearwater Air Park and Vandenburg Exec both recorded 32F, leading me to believe Tarpon Springs was likely closer to 32F than to 24F.
  • The addition of the Shuttle Landing Facility backs up @Jimbean's map on the screenshot thread in regard to the cutoff of where the climate changes in the Cape Canaveral area.
  • The numbers for the Tampa WFO in Ruskin came in backwards and were manually corrected.

Being south of 27.5o North Latitude really helped in this one, primarily due to the advective nature and the increased cloud cover.

202304192335_NWS_dump.xlsx 202304192315_2022_DecemberColdSnap.xlsx 202304180400_NWS_Parse.xlsx

  • Like 1
  • Upvote 1

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

  • 7 months later...

Updated a few of the Impact Freeze Maps with additional scraped data.  The scraped stations begin with USZ :

Each of the sheets also includes some local WUnderground data.  This gives some insight into how much impact microclimate variables and/or sensor differences can impact readings.

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

Florida, being a peninsula, is a very interesting area for weather.  I find the Sandhill region particularly interesting.  The vegetation is spectacular there.  It's a very rugged ecosystem (for Florida).  I'd recommend checking out Bok Tower Gardens, Highlands Hammock State Park and Kissimmee Prairie Preserve State Park.

Edited by RFun
Link to comment
Share on other sites

2 hours ago, RFun said:

Florida, being a peninsula, is a very interesting area for weather.  I find the Sandhill region particularly interesting.  The vegetation is spectacular there.  It's a very rugged ecosystem (for Florida).  I'd recommend checking out Bok Tower Gardens, Highlands Hammock State Park and Kissimmee Prairie Preserve State Park.

If you like those areas, you'd like Lakeland Highlands Scrub at the end of Lakeland Highlands Rd.  I've been to each of the other areas and enjoyed those as well.

Lakeland, FL

USDA Zone (2012): 9b | Sunset Zone: 26 | Record Low: 20F/-6.67C (1985, 1962) | Record Low USDA Zone: 9a | 30-Year Avg. Low: 30F | 30-year Min: 24F

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...