Cold November = Cold Winter???

Hey, remember back in July when we I asked and answered the question: does a cooler-than-normal July correspond to a colder-than-normal winter?  (The answer was: no. No, it does not.)  Well, now that we’re heading to a mid-November cold snap, I thought it might be useful to look at whether colder-than-normal November temperatures are followed by colder-than-normal winter temperatures (“winter” here meaning December, January and February).  And since I’m lazy, I’m using exactly the same “Sherlock” memes and gifs to help explain things along the way.

We’re heading into a prolonged stretch of cold weather.  Temperatures for the next 10-14 days look like they’re going to be below normal…but our on-air forecast only goes out to the next 7 days.  Those 7 days are definitely going to be below-average — at least, in terms of temperatures.  Whether or not they’re below-average in terms of life experience is kind of up to you.
Sherlock_GoodDeduction

If we take the data from the first 11 days of the month (about -3.5 degrees compared to “normal” so far), AND if we assume that our forecast for the next 7 days is accurate (don’t laugh), AND then if we optimistically hope that the remaining 12 days of November bring us a return to just-average temperatures, this would still add up to the 9th-coldest November on record.  Exciting.
Sherlock_BarelyContainMyself

Because no one can ever be satisfied just enduring a cold snap while it’s here, I’ve been getting a LOT of questions about what our current cold weather is a harbinger of what’s to come this winter.  (Yes, I used the word “harbinger.”  Don’t judge.)
Sherlock_IQ

My experience as a meteorologist tells me that short-term weather patterns are rarely reliable indicators of long-term climate trends…
Sherlock_UniverseLazy

…but any halfway-decent scientist wants proof.  So, I dove into the record books to look for answers.  First, here are the numbers I’ll be referring back to throughout the rest of this post.  These are the 30 coldest Novembers on record, along with the average December-January-February temperatures for the following winter (and their overall rank in the 143 years of Nashville records):

A few things of interest within those numbers:
1) The “normal” winter temperature over the last 143 years is 40.4 degrees.  The 20 coolest Novembers on record were followed by 12 colder-than-average winters…leaving 8 as above-average.
2) Of the 20 coldest Novembers in Nashville, 4 were followed by Top-20-coldest winters.
3) Of the 30 coldest Novembers, 9 were followed by Top-30-coldest winters.
3) Looking at the flip side, of the 20 coldest Novembers, only 2 subsequent winters ranked in the Top 20 warmest.
5) The Top 30 coldest Novembers were followed by a Top-30-warmest winter only 5 times.

So based on those facts, we can reasonably expect below-normal temperatures this winter, right?
Sherlock_NOPE

Those five facts above are fun bits of trivia, but they’re picked from a small sample size…they’re not the entire picture.  If you want to definitively state that there is a direct correlation between November temperatures and following-winter temperatures, you have to look at all the data, and plot every November temperature against every winter temperature.  If there’s a direct relationship, the chart would look like this:
rsquaredone

Here’s what the chart really looks like for November vs. Winter temperatures:

Quickly glance at that, and you probably notice the line through the dots — that’s the “best-fit” line that attempts to summarize all the data points in a straight line.  “AHA!” you say, “that line goes UP!  Therefore, a cooler July equals a colder winter, and a hotter July equals a warmer winter!”  And Moriarty says…
Sherlock_Ordinary

There’s a way to measure the accuracy of that “best-fit” line…basically, measuring how close the data “dots” correspond to the line.  It’s called the “Coefficient of Determination,” and it’s abbreviated as R-squared.  (Why is it R-squared instead of CD?  Why isn’t it called something simpler?  These are questions I asked my statistics professor in graduate school, and the answers were…unenlightening.)
Sherlock_eyeroll

Anyway…that “Coefficient of Determination” tells us whether the best-fit line does a good job of summarizing the data, or whether it’s just the software doing it’s best to pound a square peg into a round hole.  An R-squared value of exactly 1 means that it’s a perfect fit, while values closer to zero mean that there’s really no relationship between the two things you’re trying to compare.

For November temperatures vs. Winter temperatures, the R-squared value is 0.0447.  That’s really low, and tells us that with almost 150 years of data, there is virtually no predictive relationship between temperatures in November and temperatures the subsequent winter.

Sherlock_Laters

Posted in Uncategorized | 7 Comments

October 13 Severe Weather Threat

(Posted 4:20 p.m., 10/13/2014)

Tornado warnings have started being issued in Hardin, McNairy and Decatur counties west of the Tennessee River.

Stay with 4WARN Weather for updates.

(Posted 3:25pm 10/13/2014)
Just a quick update to let you know that the Storm Prediction Center really hasn’t changed their thinking regarding our threats this evening and tonight — they’re maintaining the Moderate/Slight risk dividing line along the Tennessee River (upper left).  The Moderate Risk corresponds with the highest tornado threat, also some tornadic potential exists across the entire Midstate (upper right).  Within the Slight Risk area, locations west of I-65 have a higher probability of damaging winds occurring within 25 miles (lower left).  All locations have a relatively low risk of large hail (lower right).
SPC_Oct13_20z

Dan Thomas and Lisa Spencer will be on-air this evening during our newscasts and, if necessary, during prime time if/when warnings are issued.

——————————–

(Posted 1:15pm 10/13/2014)
A few updates to our severe weather outlook, now that things are beginning to come into better focus regarding our potential for nasty weather later today…

It looks like there will be plenty of instability available in the atmosphere to help drive severe storms — as of 1pm, temperatures are in the upper 70s, so we’re on track to reach our forecast highs in the low 80s.  The warm temperatures combined with unseasonably high humidity will result in a “buoyant” atmosphere supportive of surface-based thunderstorms (more capable of producing tornadoes, compared to elevated-base thunderstorms).  Even as the atmosphere cools after sunset, strong southerly winds will keep plenty of warm and unstable air in place ahead of the storms.

The precise timing of when the storms will arrive remains the least-confident part of the forecast…until the squall line shows more and consistent signs of actually slowing down, I’m maintaining this estimate of severe storm “windows”:
PAUL STORM TIMING

If anything, I’m erring on the early side — partly to account for pre-squall-line storms, and partly so the storms in the main line don’t catch you off-guard.  The Nashville office of the National Weather Service is still forecasting a later arrival time — they’ve adjusted the previous version of this map to reflect an earlier time frame, but they’re obviously banking on the squall line slowing down significantly.  (To be clear: I’m providing their map not as a “hey, look what THESE guys are saying LOL” but as an example of how reasonable minds are interpreting data in a slightly different manner):
NWS_Oct13

One concerning possibility is that of isolated supercell thunderstorms developing ahead of the main squall line.  As I wrote this morning (below), squall lines are typically associated with a straight-line wind threat — but any discrete storms that develop ahead of the line will have a higher tornadic potential.  I’m not changing my thinking regarding our relative threat levels — winds and flooding are still the biggest concerns, but the tornado potential definitely isn’t zero.  The pre-squall-line storms are the ones developing down in Alabama as of just after 1pm (also, this image gives you an idea of what’s headed our way):
VIDEO VIPIR

The Storm Prediction Center has expanded their “Moderate Risk” outlook a little farther east (not far enough, in my opinion — I still think I-65 should be the Moderate/Slight dividing line).

PAUL SEVERE RISK REGION

I’m still concerned that SPC hasn’t adequately adjusted for the faster-than-expected movement of the entire complex of storms.  Their next update is scheduled for 3pm — I’ll post a quick update with some thoughts at that point.  We’ll likely have a tornado watch issued for our western counties before then — but the storms themselves will still be in West Tennessee.

This morning’s post is below…

———————————-

(Posted 9:25am 10/13/2014)
A potentially dangerous and complicated weather pattern is shaping up over Middle Tennessee and southern Kentucky over the next 24 hours.  So let’s take a look at what we know, what we think we know, and what we’re still figuring out.

WHAT WE KNOW
1. Strong/severe thunderstorms are moving eastward through Arkansas, Louisiana and Texas this morning.
2. Computer models have been WAY too slow in how they’re depicting the progression of these storms (most are 3+ hours behind).  Check out the 9am radar image (left) next to Futurecast’s version of what that radar should look like:
VIDEO VIPIR
3. Other than isolated showers/storms, the atmosphere over the Midstate will be warming up and becoming progressively more unstable.  It will also be quite windy — a Wind Advisory is in effect this afternoon through tonight:
PAUL WATCH
4. The ground is already saturated, due to last week’s and this morning’s rain.  Storms moving in later today and tonight will be capable of producing flash flooding in a very short period of time, especially considering that rainfall amounts through early Tuesday are forecast in the 2″-3″ range.  A Flash Flood Watch is in effect until 7pm Tuesday:
PAUL WATCH 2

WHAT WE THINK WE KNOW
1. The environment over Middle Tennessee and southern Kentucky as the storms arrive will be especially favorable for damaging straight-line winds.  There’s a lot of wind energy associated with this system, but a majority of it is associated with “speed shear” (change in wind speed as you ascend in the atmosphere) as opposed to “directional shear” (change in wind direction as you go up).  That means that while tornadoes are still possible, the greater threat is from thunderstorm wind gusts over 60mph.
2. As the atmosphere warms up, there will be enough instability over our heads for severe storms to maintain their strength.  However, it does not look there will be enough instability to support much of a hail threat at this point.
3. The earlier the storms move in, the greater our severe weather potential will be.  Which brings us to…

WHAT WE’RE STILL FIGURING OUT
1. When will the storms get here???  Pretty much the million-dollar question, especially considering that “daylight arrival = nastier storms”.

Factoring in the faster-than-anticipated movement of the storms this morning, here’s my latest estimate of when the strongest storms will arrive in the Midstate:
PAUL STORM TIMING

For posterity’s sake, here the NWS-Nashville’s version of that map:
NWS_Oct13

Obviously some pretty major differences.  I think it’s likely that the storms will decelerate as they cross the Mississippi River, but they’re already 3 hours ahead of this morning’s computer guidance.  Speaking of computer guidance, here’s how Futurecast depicts the radar simulation this evening through tomorrow morning — the hot-off-the-press newest model run is significantly faster that previous versions, and has a more-reasonable estimate of the timing (but I’m thinking it’s still an hour or two too slow):
RPM 5P MON
RPM 7P MON
RPM 9P MON
RPM 11P MON
RPM 1A TUE
RPM 3A TUE
RPM 5A TUERPM 7A TUE
RPM 9A TUE
RPM 11A TUE

Despite its timing challenges, I do think Futurecast has a pretty good handle on what the storm structure will look like — a squall line (technically, a quasi-linear convective system) with embedded supercell features.  The individual storms that comprise the squall line will be moving generally south-to-north within the line, while the line itself will slow down and move from west-to-east throughout the evening and into tonight.  Notice that some re-development of storms is predicted for early Tuesday morning along and east of I-65.

Squall lines are primarily associated with straight-line wind events, but little areas of rotation can spin up within the line and produce isolated tornadoes.  The heavy rain and slowing movement of the line could produce flash-flooding throughout the night…especially with the potential re-development of storms Tuesday morning.  Overall, I’d rank the threats in this order: straight-line winds, flash flooding, tornadoes, hail:
PAUL SEVERE RISK GRAPH 2

The Storm Prediction Center has noted the fast forward movement of the storms so far this morning, but their 8am outlook maintained the position of the severe weather “Moderate Risk” to the west of the Tennessee River, with a “Slight Risk” for Middle Tennessee and southern Kentucky:
PAUL SEVERE RISK REGION

My concern with SPC’s logic is that the storm line may outrace the Moderate Risk area — I think a better Moderate/Slight dividing line would be roughly along I-65, since that’s how far east the storms could get before the sun goes down (instability levels decrease after sunset).  After sunset, and as the storms progress eastward into a less-favorable air mass, the severe threat will diminish…but it won’t go away entirely.  SPC’s next update is due to be posted at 11:30am, so we’ll see what they say at that point.

The biggest wild-card in all of this is the possibility that this morning’s squall line will move too far ahead of its upper-level support and fall apart…if that happens, we’d be playing a waiting game as another squall line develops to take its place.  (In which case the forecast becomes so complicated that you’ll be able to watch my hairline recede before your very eyes as I try to figure out the forecast.)  There are other wild-cards that could also reduce or rearrange our severe weather potential — daytime clouds, non-severe storms developing ahead of the main line, a layer of warm air in the atmosphere that could inhibit storm development…you get the idea.

Either way, plan on staying weather-aware this afternoon and this evening.  I’ll be on Twitter regularly (and Facebook irregularly) with updates, and of course we’re on the air with newscasts at noon and beginning at 4:00 this afternoon.  Watches and warnings are automatically posted to my Twitter account — and allow me to remind you once again that Facebook is not a good source of time-sensitive information like severe weather alerts.

Above all, don’t freak out.  (Easy for me to say after I just dumped a thousand words worth of “this could be bad” on you!)  This is NOT a situation where you should plan on sleeping in your safe place with a helmet on and a weather radio clutched in your shaking hands…even in high-risk severe weather scenarios, the odds of something happening in your immediate vicinity are pretty low — but it’s always best to be prepared!

——————–
Social media links

Twitter: @WSMVweather, @PaulHeggenWSMV, @WSMVLisaSpencer, @WSMVDanThomas, @WSMVNancyVanC, @NWSNashville

Facebook: 4WARN Weather, Paul Heggen WSMV, Lisa Spencer, Dan Thomas WSMV, Nancy Van Camp WSMV, NWS Nashville

Posted in Uncategorized | Tagged , , , | 2 Comments

October 6 Severe Weather Threat

We’re in the middle of a pretty turbulent weather pattern…last Thursday we almost tied a record high temperature (we hit 90), we had severe storms Thursday night, then by Sunday morning our temperatures were down into the upper 30s.  Now the warm air is moving back into the Midstate, and we’re facing another potential severe weather event.

I’ll start with the good news: the weather conditions this afternoon won’t be perfectly aligned for severe weather.  But this is one of those situations where conditions don’t need to be perfect, they just need to be good enough.  Let’s break it down by which ingredients will be favorable, and which ones could help to limit our severe storm potential…

Severe thunderstorms require an environment in which conditions near ground-level are warm and muggy, with cold air in the upper atmosphere — this setup allows storms to explosively develop.  The air over our heads is definitely cold…the factor that’s lacking will be warmth and humidity near the ground.  The temperature in Nashville as of 9am is 59 degrees, and the dew point is 55.  For severe storms to develop, that dew point number needs to be at least 55, and really should be 60+, so we’re kind of borderline there.  Southwesterly winds will be pushing more humidity into the region, but it will be an uphill battle.  Additionally, temperatures today will only reach the low to mid 70s…all in all, the not-quite-warm-enough and not-quite-humid-enough air near the ground is our best hope for dodging severe weather.

Another limiting factor is the lack of a strong front to get the storms started in the first place.  A well-defined cold or warm front is a strong trigger for severe storms, and we’re just not going to see one of those in the area this afternoon.  The trigger for our storms this afternoon will be the approach of a strong upper-level storm system — even though it’s high in the atmosphere, it can still influence the behavior of air near the ground.  The lack of a surface front means that storms that develop this afternoon will be loosely organized (at best) — that means we can’t give you a specific “arrival time” map.  The storms will be clustered in different bunches that may behave erratically.

There will be plenty of wind energy in the atmosphere, so the threat of straight-line winds will be present in any storm this afternoon.  The wind profiles this afternoon aren’t terribly impressive from the perspective of a tornado threat, but at this point we can’t rule out isolated tornadoes.  And those cold temperatures in the upper atmosphere that I mentioned earlier will also help to support the threat of large hail.

One last limiting factor: this morning’s rain, which helps to temporarily stabilize the environment.  The atmosphere looks like it will have time to recover enough energy late this morning and early this afternoon, but that’s largely dependent on how much sun breaks through the lingering clouds.

Add it all up, and we’re talking about an environment in the Midstate this afternoon that will support the intensification of thunderstorms…if those storms get going in the first place.  Yes, it’s complicated.  (Isn’t it always?)  Let’s get to some maps, beginning with how Futurecast depicts the development of storms this afternoon and this evening.
BAMS 2P MON
BAMS 3P MON
BAMS 4P MON
BAMS 5P MON
BAMS 6P MON
BAMS 7P MON

So, does that series of images depict exactly what the radar will look like this afternoon?  No, but it does give us a reasonable approximation of where the greatest severe threat will take shape (along and south of I-40), when the storms will develop (mid- to late-afternoon), and the type of storms we can expect (supercell storms, with gaps in between the individual thunderstorms).

The Storm Prediction Center has included all of Middle Tennessee and southern Kentucky in their “Slight Risk” outlook for today:
PAUL SEVERE RISK REGION

Because there are so many “if this, then that” factors that could limit the development of severe storms, the percentages are pretty low in terms of the specific wind, hail and tornado threats.  SPC estimates a 15% chance of large hail (1″+ diameter) within 25 miles of any one point in the Midstate…
SPC-oct6-hail

…and also a 15% chance of damaging straight-line winds…
SPC-oct6-wind

…and a 5% chance of a tornado within 25 miles:
SPC-oct6-tor

One method of evaluating forecasts is to look back at similar historical scenarios…one such “analog” forecast found that atmospheric conditions similar to today’s have produced severe weather about 30% of the time:
CIPS-Oct6

The individual severe reports from the 15 most-similar days show why we’re taking this so seriously:
CIPS-Oct6-reports

BOTTOM LINE: Don’t freak out.  You don’t need to hunker down in your safe place for the rest of the day, just stay weather-aware this afternoon and this evening.  There’s a chance that we’ll get lucky and dodge the severe weather this time around, but it’s still something that bears careful watching.  (In fact, I’m skeptical enough about our storm chances this afternoon and evening that I wouldn’t cancel soccer practice, barbecue plans, or other outdoor stuff this evening — just be flexible!)  We’ll have on-air updates at noon and during our early-evening newscasts, and we’ll let you know right away if any warnings are issued.  If you’re away from the TV, follow us on social media (links at the bottom of this post).  As always, any warnings will be automatically posted on my Twitter stream, @PaulHeggenWSMV.  (And I’ll refrain from tweeting non-weather thoughts today.  You’re welcome.)

——————–
Social media links

Twitter: @WSMVweather, @PaulHeggenWSMV, @WSMVLisaSpencer, @WSMVDanThomas, @WSMVNancyVanC, @NWSNashville

Facebook: 4WARN Weather, Paul Heggen WSMV, Lisa Spencer, Dan Thomas WSMV, Nancy Van Camp WSMV, NWS Nashville

Posted in Uncategorized | Tagged , , , | 2 Comments

October 2-3 Severe Weather Threat

It’s been a couple of months since we’ve had enough of a severe weather threat to discuss in this space…but autumn is our secondary severe weather season — think of it as the junior varsity to springtime’s varsity.  Tonight will bring us our first potential severe weather event of the fall — the threat is marginal in our part of the country, but it still bears watching.

Unseasonably warm and breezy conditions will prevail today — the record high temperature for October 2 is 91 degrees, and we’ll be awfully close to that this afternoon.  Strong southerly winds will reinforce the warm air, but also pump in higher levels of humidity…not August-level humidity, but still noticeable.  We could see a few spotty storms in western Middle Tennessee this afternoon, but it’s just a 30% chance and they’ll be non-severe.  Those storms are reflected in both Futurecast simulations for this afternoon (BAMS model on the left and RPM on the right, weather nerds).
BAMS 4P THU

Severe thunderstorms will develop far to our west this afternoon, in Arkansas and Missouri.  They’ll quickly coalesce into a squall line of storms that will march from west to east, eventually crossing the Mississippi River…and approaching the Tennessee River by midnight.
BAMS 12A FRI

The storms will continue their eastward progression throughout the late-night and early-Friday-morning time frame.  Best estimate for their arrival in Metro Nashville is around 3-4am.  Keep in mind that these storms don’t even exist yet!  Trying to pin down their specific arrival times in various parts of the Midstate is very tricky.
BAMS 3A FRI

Those two Futurecast snapshots represent when the storms will pose the greatest severe threat for our neck of the woods.  Most of the Midstate and southern Kentucky are included in the Storm Prediction Center’s “Slight Risk” area for tonight and early Friday morning (specifically, that means a 15% chance of 60+mph wind gusts within 25 miles of any one location in the risk area).
PAUL SEVERE RISK REGION

It’s hardly ever possible to completely rule out the possibility of an isolated tornado or some 1″-diameter hail, but both of those threats are very low tonight (2% tornado chance, 5% hail chance).  And since we’ve been so dry lately, the ground will be able to absorb all but the heaviest downpours, so flooding isn’t a major concern either.
PAUL SEVERE RISK GRAPH 2

As the storms continue on to the east of I-65, they’ll weaken significantly.  At this point, it looks like most of the thunderstorms will be moving off the Cumberland Plateau shortly after sunrise on Friday.
BAMS 6A FRI

Smoothing out the differences between the various models and allowing a little “flex time” for Mother Nature’s mood swings, here’s the best estimate of when the strongest thunderstorms will impact various portions of the Midstate.
PAUL STORM TIMING

As I mentioned on TV this morning, this is a very borderline severe weather threat — in general I expect strong-but-not-severe storms, but a few stronger cells could prompt severe thunderstorm warnings.  That means a fair amount of lightning and thunder, some heavy downpours, but not much in the way of damage.  These kinds of marginal straight-line wind threats generally result in some tree damage and some spotty power outages, but not much more than that…unless of course a tree falls on your house!  Still, it’s something you’ll want to stay aware of — we’ll have updates during our regularly scheduled newscasts this evening, and we’ll be here monitoring the storms as they move in overnight.

I automatically tweet all Middle Tennessee and southern Kentucky watches and warnings — follow me at @PaulHeggenWSMV for that.  (I promise, I won’t tweet about baseball or football tonight.  But once the storms have passed through, all bets are off!)  Links to the social media profiles of all of WSMV’s meteorologists are at the bottom of this post.

More storms will pop back up around midday on Friday, but these are expected to be non-severe…they’ll be the result of the atmosphere getting one last squeeze from the approaching cold front.  That front, by the way, will drop temperatures significantly by the weekend — parts of the upper Cumberland Plateau could even see some frost on the rooftops Sunday morning!

——————–
Social media links

Twitter: @WSMVweather, @PaulHeggenWSMV, @WSMVLisaSpencer, @WSMVDanThomas, @WSMVNancyVanC, @NWSNashville

Facebook: 4WARN Weather, Paul Heggen WSMV, Lisa Spencer, Dan Thomas WSMV, Nancy Van Camp WSMV, NWS Nashville

Posted in Uncategorized | Tagged , , | Leave a comment

Storm Spotters Wanted!

Do you find yourself gazing at developing storms, checking radar images frequently and playing weatherman/woman for your friends and family? It sounds like you may be a candidate for the National Weather Service Storm Spotter program. With some training you can report your observations to the National Weather Service and Channel 4 and help warn people and save lives during severe weather outbreaks. You can even get the training from your living room now via your computer. The first on-line class is tonight (Monday, September 22).

0410091243 Benton County

Here are the guidelines from the National Weather Service.

Starting tonight, September 22, NWS Nashville will start the third season of online storm spotter classes. This is in coordination with our live, in-person classes each fall and spring. The online classes give those who have a difficult time making an in-person class the opportunity to become a storm spotter right from the comfort of their own home.

1. If you have never attended a spotter class in the past, OR it has been more than 5 years since your last class, we require you to watch 2 one-hour modules; one on the basic role of the spotter and the other on convective basics.

2. Then, attend one of the scheduled webinars through Go-To Meeting. These are live sessions with one of us here at the NWS that last 45-60 minutes and they get much more specific about what we are looking for during severe weather.

3. While not a requirement, you will need an VOIP microphone on your Mac, PC, iPad, iPhone or Android to actively participate in the webinar.

Go to this address for other dates and the training videos.

http://www.srh.noaa.gov/ohx/?n=onlinebasicspotterclass

Posted in Uncategorized | Leave a comment

Cool July…Cold Winter???

We’re in the midst of our third stretch of sustained below-average temperatures this month…in fact, July of 2014 will likely be one of the 20 coolest on record.  (The Nashville record book extends back to 1871, so that’s significant.)  All in all, while there have been some hot and humid days, and some strong thunderstorms here and there, the weather this month has been pretty great.
Sherlock_GoodDeduction
Sherlock_BarelyContainMyself

Because no one can ever be satisfied with enjoying pleasant weather while it’s here, I’ve been getting a LOT of questions about what our current cool weather portends for next winter’s weather.  (Yes, I used the word “portends.”  Don’t judge.)
Sherlock_IQ

My experience a meteorologist tells me that short-term weather patterns are rarely reliable indicators of long-term climate trends…
Sherlock_UniverseLazy

…but any halfway-decent scientist wants proof.  So, I dove into the record books to look for answers.  First, here are the numbers I’ll be referring back to throughout the rest of this post.  These are the 40 coolest Julys on record, along with the average December-January-February temperatures for the following winter (and their overall rank in the 143 years of Nashville records):
July_Winter_Comparison

A few things of interest within those numbers:
1) The “normal” winter temperature over the last 143 years is 40.4 degrees.  The 20 coolest Julys on record were followed by 13 colder-than-average winters…leaving 7 as above-average.
2) Of the 20 coolest Julys in Nashville, 6 were followed by Top-20-coldest winters.
3) Expand that a bit, and 10 were followed by Top-40-coldest winters.
4) Looking at the flip side, of the 20 coolest Julys, only 2 subsequent winters ranked in the Top 20 warmest.
5) The Top 20 coolest Julys were followed by a Top-40-warmest winter only 5 times.

So based on those facts, we can reasonably expect below-normal temperatures next winter, right?
Sherlock_NOPE

Those five facts above are fun bits of trivia, but they’re picked from a small sample size…they’re not the entire picture.  If you want to definitively state that there is a direct correlation between July temperatures and following-winter temperatures, you have to look at all the data, and plot every July temperature against every winter temperature.  If there’s a direct relationship, the chart would look like this:
rsquaredone

Here’s what the chart really looks like for July vs. Winter temperatures:
July_Winter_Graph

Quickly glance at that, and you probably notice the line through the dots — that’s the “best-fit” line that attempts to summarize all the data points in a straight line.  “AHA!” you say, “that line goes UP!  Therefore, a cooler July equals a colder winter, and a hotter July equals a warmer winter!”  And Moriarty says…
Sherlock_Ordinary

There’s a way to measure the accuracy of that “best-fit” line…it’s called the “Coefficient of Determination,” and it’s abbreviated as R-squared.  (Why is it R-squared instead of CD?  Why isn’t it called something simpler?  These are questions I asked my statistics professor in graduate school, and the answers were…unenlightening.)
Sherlock_eyeroll

Anyway…that “Coefficient of Determination” tells us whether the best-fit line does a good job of summarizing the data, or whether it’s just the software doing it’s best to pound a square peg into a round hole.  An R-squared value of exactly 1 means that it’s a perfect fit, while values closer to zero mean that there’s really no relationship between the two things you’re trying to compare.

For July temperatures vs. Winter temperatures, the R-squared value is 0.019.  That’s really really low, and tells us that with almost 150 years of data, there is virtually no predictive relationship between temperatures in July and temperatures the following winter.

Sherlock_Laters

Posted in Uncategorized | Tagged , , , | 14 Comments

July 14 Severe Threat

JUST-AFTER-NOON UPDATE: It’s hot, it’s muggy, and so far we’ve seen limited shower and thunderstorm activity.  Combine those factors with the gradual approach of a cold front, and the pump is still primed for severe weather later today.  But WHEN today?  That part of the equation has, if anything, become even more muddled.  Two short-term models (HRRR and RPM) say we’ll get strong afternoon storms (blowing up by 3-4pm) followed by not-as-strong overnight storms.  One other model (BAMS) says the atmosphere will stay “capped” this afternoon, and the overnight storms will be the main event.  I lean toward a compromise: the strongest storms will develop this afternoon, but they’ll be loosely organized.  The well-organized (and more numerous) storms will arrive tonight with the actual cold front, but the after-sunset arrival would limit their severe potential around here.  But as we’ve seen before (in fact, just six days ago): the atmosphere is gonna do what the atmosphere wants to do.
you-got-me

——————–

Let’s start with the good news: today’s severe thunderstorm potential isn’t an off-the-charts, hide-in-your-safe-place-until-Tuesday type of scenario.
bb-high-five

Hang on there, Walt.  (You too, Jesse.)  Even a marginal risk of severe weather is still a risk of severe weather.  This morning’s weather-balloon data shows plenty of moisture in the lowest levels of the atmosphere, which is the fuel for strong thunderstorms.  Whether or not we get strong/severe storms this afternoon depends on whether we get storms this morning.
what

Exactly.  Stay with me here: think of the atmosphere as an iPhone battery.  The amount of moisture in the atmosphere this morning means that the battery is fully charged, at 100%.  This morning’s storms are like playing Candy Crush non-stop for an hour or so — not enough to drain the battery completely, but enough to drop the charge to 50%.  Severe thunderstorms are like using Google maps to navigate a turn-by-turn route for a long drive while downloading a podcast on a roaming network…you need a full 100% charge to make that happen.  So, in areas that are getting non-severe storms this morning, you’re likely not going to have enough “charge” left in the atmosphere for severe weather to be as much of a threat this afternoon.  If you stay dry this morning, the atmosphere is going to be ready to go later today.

(In related news, I may be slightly frustrated with my phone’s battery life.)
shaking_head_breaking_bad

The problem from a forecasting perspective is that even the short-term, high-resolution weather models don’t do a very good job of handling early-morning storms…which means their opinion regarding storm development later in the day are highly suspect.  That said, this series of images (from the HRRR model, for you weather nerds) does a pretty good job of simulating both the timing and placement of this morning’s non-severe storms, which lends it a little more credence regarding the evolution of stronger storms this afternoon:
HRRR 10AM
HRRR 1PM
HRRR 3PM
HRRR 6PM
HRRR 8PM
yeah-science

Keep in mind: these models simulate what the atmosphere is going to do, they’re not a guarantee of specific future timing, placement or strength.  The overall pattern is what we’re looking at, and in this case the overall pattern is:
1) Morning storms in northern Middle Tennessee diminish
2) Strong (possibly severe) storms develop along and north of I-40 this afternoon, and they’ll gradually move farther south
3) More storms will develop in West Tennessee this evening and move into southern Middle Tennessee, increasing flood concerns in that part of the Midstate.

We are included in the Storm Prediction Center’s “Slight Risk” outlook for severe thunderstorms today…here’s that outlook, along with the specific threats for us today:
PAUL SEVERE RISK REGION
PAUL SEVERE RISK GRAPH 2

The problem with this type of weather pattern is that one tiny change in this morning’s weather will have a huge impact on this afternoon’s and evening’s weather.  Returning to the iPhone analogy…if you leave about 20 apps running in the background, there are 20 different ways your battery can be run down without you being aware of it.  Now try to predict exactly which app will cause the battery to crash, and at exactly what time…that’s the “fun” part of trying to predict summertime severe weather in this part of the country.

Overall, I’ll return to where I started: this is a pretty low-end risk of severe weather.  Odds are, in your neighborhood you’ll probably just get some garden-variety storms that produce some heavy rain and lots of lightning and thunder.  (Some of us may avoid even that!)  But the severe risk isn’t zero, and it’s high enough to warrant staying “weather-aware” today, especially in the afternoon and evening.  The best-case scenario is that the morning storms produce enough lingering cloud cover to limit this afternoon’s severe potential…the worst-case scenario is lots of storms producing widespread damaging straight-line winds.  (As usual, the safest bet is “somewhere in between.”)

We’ll keep you updated during our regular newscasts at noon, and beginning again at 4pm.  We’ll also keep you “4WARNed” (branding!) on social media (links below).

——————–
Social media links

Twitter: @WSMVweather, @PaulHeggenWSMV, @WSMVLisaSpencer, @WSMVDanThomas, @WSMVNancyVanC, @NWSNashville

Facebook: 4WARN Weather, Paul Heggen WSMV, Lisa Spencer, Dan Thomas WSMV, Nancy Van Camp WSMV, NWS Nashville

Posted in Uncategorized | Tagged , , , , | 2 Comments