2020 Global Model Runs Discussion (Out thru day 16)

This is the general tropical discussion area. Anyone can take their shot at predicting a storms path.

Moderator: S2k Moderators

Forum rules

The posts in this forum are NOT official forecasts and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or STORM2K. For official information, please refer to products from the National Hurricane Center and National Weather Service.

Help Support Storm2K
Message
Author
User avatar
cheezyWXguy
Category 5
Category 5
Posts: 6108
Joined: Mon Feb 13, 2006 12:29 am
Location: Dallas, TX

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2041 Postby cheezyWXguy » Fri Sep 04, 2020 7:23 pm

It seems to me that these tracks are dependent on how close the two systems get to each other. They are both heading directly west until the first one slows down enough for the second to catch it. Once they get close enough they start some kind of Fujiwara action where they alternate swinging each other north. If the first one doesn’t slow down as much, it would seem to throw a wrench in some of this, and would likely explain why the storms get further west in the para runs.
2 likes   

User avatar
chaser1
S2K Supporter
S2K Supporter
Posts: 5476
Age: 64
Joined: Sat Oct 15, 2005 5:59 pm
Location: Longwood, Fl

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2042 Postby chaser1 » Fri Sep 04, 2020 7:58 pm

Emmett_Brown wrote:Just my opinion, but the model solutions for the eastern Atlantic are still very inconsistent. Models are struggling to figure out the complex interactions of 91, 92, and the 2 waves over Africa. They try to develop several at once, which could mean no development until one of these waves breaks free from the rest, or until they consolidate... a process that usually takes some time. I'm starting to think far east development may be delayed a week or so. Maybe conditions are "too favorable". Don't get me wrong, we'll have development out there before the middle of the month, but maybe not until that mess gets mopped up.


Totally aligns with how I see the bigger picture. This year has been a virtual sperm bank of Tropical disturbances with an over abundance becoming named storms. In spite of that, tropical storms have struggled to develop in the E. Atlantic. Disturbances have sputtered until tracking further west where conditions have proven to be more conducive. And here we are on the cusp of "peak season". Maybe conditions have changed. The models suggest they have and are lighting up the E. Atlantic with genesis. We have an active ITCZ and at least two distinct Tropical disturbances in the Eastern MDR. Done deal...... right?

In any typical year at peak season one or more of these disturbances would wind up pretty quick. Given 92L's latitude I'd normally bet the farm on a near term recurve. Thereafter, one or two would likely follow and we'd then turn our attention further west. I think Emmett makes a fantastic point when questioning
which disturbance develops and more importantly "when" it occurs. The entire picture changes if 92L somehow remains broad and disorganized long enough to eventually getting pushed further west. I have had little faith in the models' ability to forecast genesis this year. I'm not sure I buy into them having been quietly fixed suddenly. We should know soon enough right?
6 likes   
Andy D

(For official information, please refer to the NHC and NWS products.)

User avatar
Blown Away
S2K Supporter
S2K Supporter
Posts: 10148
Joined: Wed May 26, 2004 6:17 am

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2043 Postby Blown Away » Fri Sep 04, 2020 8:17 pm

12z EURO shows a wave moving of Africa @Tuesday and beelining west across Atlantic as a Cape Verde major hurricane. Bermuda may be only threat from all the current activity, this potential wave could be a problem.


https://www.tropicaltidbits.com/analysi ... 0412&fh=12
0 likes   
Hurricane Eye Experience: David 79, Irene 99, Frances 04, Jeanne 04, Wilma 05... EYE COMING MY WAY IN 2024…
Hurricane Brush Experience: Andrew 92, Erin 95, Floyd 99, Matthew 16, Irma 17, Ian 22, Nicole 22…

User avatar
toad strangler
S2K Supporter
S2K Supporter
Posts: 4543
Joined: Sun Jul 28, 2013 3:09 pm
Location: Earth
Contact:

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2044 Postby toad strangler » Fri Sep 04, 2020 8:25 pm

Blown Away wrote:12z EURO shows a wave moving of Africa @Tuesday and beelining west across Atlantic as a Cape Verde major hurricane. Bermuda may be only threat from all the current activity, this potential wave could be a problem.


https://www.tropicaltidbits.com/analysi ... 0412&fh=12


Well documented earlier today here but the GFS has said no since then, as have the climo watchers.
1 likes   

User avatar
Blown Away
S2K Supporter
S2K Supporter
Posts: 10148
Joined: Wed May 26, 2004 6:17 am

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2045 Postby Blown Away » Fri Sep 04, 2020 8:45 pm

toad strangler wrote:
Blown Away wrote:12z EURO shows a wave moving of Africa @Tuesday and beelining west across Atlantic as a Cape Verde major hurricane. Bermuda may be only threat from all the current activity, this potential wave could be a problem.


https://www.tropicaltidbits.com/analysi ... 0412&fh=12


Well documented earlier today here but the GFS has laughed it off since then, as have the climo watchers.


Yeah, but it’s the only TW the Euro consistently blows up into a major hurricane run after run.
0 likes   
Hurricane Eye Experience: David 79, Irene 99, Frances 04, Jeanne 04, Wilma 05... EYE COMING MY WAY IN 2024…
Hurricane Brush Experience: Andrew 92, Erin 95, Floyd 99, Matthew 16, Irma 17, Ian 22, Nicole 22…

User avatar
cheezyWXguy
Category 5
Category 5
Posts: 6108
Joined: Mon Feb 13, 2006 12:29 am
Location: Dallas, TX

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2046 Postby cheezyWXguy » Fri Sep 04, 2020 8:50 pm

toad strangler wrote:
Blown Away wrote:12z EURO shows a wave moving of Africa @Tuesday and beelining west across Atlantic as a Cape Verde major hurricane. Bermuda may be only threat from all the current activity, this potential wave could be a problem.


https://www.tropicaltidbits.com/analysi ... 0412&fh=12


Well documented earlier today here but the GFS has laughed it off since then, as have the climo watchers.

I think you’re saying that in jest, but as a general statement, I don’t understand the fixation on climatology I sometimes see. Hurricanes don’t follow climatology, climatology follows hurricanes. It is not only a historical average of past storm tracks, but also a historical average of the environmental factors that steered the storms. If climatology had a significant impact on individual events, all Cape Verde storms would recurve and the CLP5 would be the most accurate tool we have. If someone can explain climatology’s relevance to forecasting individual events, I am open to hearing it, but I am of the opinion that it is much more valuable to use historical analogs of similar setups than a single historical average.
4 likes   

sma10
Category 5
Category 5
Posts: 1710
Joined: Tue Aug 02, 2005 1:13 pm

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2047 Postby sma10 » Fri Sep 04, 2020 8:53 pm

The 5+ day forecasts are always wait and see affairs in the best of times, but I think even more so now:

- models have been, let's be charitable and say, "uneven" this season
- 6-10 day patterns can turn out to be wrong
- the genesis of this wave train is highly uncertain, as well as interaction between the systems

Given the above, I will be gobsmacked if the current 6 to 10 day forecast points turn out to be correct.

In a way, this is actually more fun to follow. It seems that as modeling has improved, we don't get truly shocked anymore - just the occasional mild surprise. So, in the next week, will these systems interact in the MDR? Will that ULL completely disrupt the ridge? I don't know ... but my sense is that the solution may end up being something we aren't even looking at yet
4 likes   

User avatar
Blown Away
S2K Supporter
S2K Supporter
Posts: 10148
Joined: Wed May 26, 2004 6:17 am

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2048 Postby Blown Away » Fri Sep 04, 2020 9:26 pm

cheezyWXguy wrote:
toad strangler wrote:
Blown Away wrote:12z EURO shows a wave moving of Africa @Tuesday and beelining west across Atlantic as a Cape Verde major hurricane. Bermuda may be only threat from all the current activity, this potential wave could be a problem.


https://www.tropicaltidbits.com/analysi ... 0412&fh=12


Well documented earlier today here but the GFS has laughed it off since then, as have the climo watchers.

I think you’re saying that in jest, but as a general statement, I don’t understand the fixation on climatology I sometimes see. Hurricanes don’t follow climatology, climatology follows hurricanes. It is not only a historical average of past storm tracks, but also a historical average of the environmental factors that steered the storms. If climatology had a significant impact on individual events, all Cape Verde storms would recurve and the CLP5 would be the most accurate tool we have. If someone can explain climatology’s relevance to forecasting individual events, I am open to hearing it, but I am of the opinion that it is much more valuable to use historical analogs of similar setups than a single historical average.


It’s usually a better bet to go with climatology because the averages favor a Cape Verde hurricane forming in the East to Central Atlantic recurving before the Caribbean. I agree that patterns tend to repeat themselves with hurricanes and if analog patterns can be identified it is more valuable than just climatology.

Euro for many runs showing a strong hurricane and when it’s that persistent with a strong system it’s deserves attention. JMHO
0 likes   
Hurricane Eye Experience: David 79, Irene 99, Frances 04, Jeanne 04, Wilma 05... EYE COMING MY WAY IN 2024…
Hurricane Brush Experience: Andrew 92, Erin 95, Floyd 99, Matthew 16, Irma 17, Ian 22, Nicole 22…

User avatar
toad strangler
S2K Supporter
S2K Supporter
Posts: 4543
Joined: Sun Jul 28, 2013 3:09 pm
Location: Earth
Contact:

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2049 Postby toad strangler » Fri Sep 04, 2020 9:30 pm

Blown Away wrote:
cheezyWXguy wrote:
toad strangler wrote:
Well documented earlier today here but the GFS has laughed it off since then, as have the climo watchers.

I think you’re saying that in jest, but as a general statement, I don’t understand the fixation on climatology I sometimes see. Hurricanes don’t follow climatology, climatology follows hurricanes. It is not only a historical average of past storm tracks, but also a historical average of the environmental factors that steered the storms. If climatology had a significant impact on individual events, all Cape Verde storms would recurve and the CLP5 would be the most accurate tool we have. If someone can explain climatology’s relevance to forecasting individual events, I am open to hearing it, but I am of the opinion that it is much more valuable to use historical analogs of similar setups than a single historical average.


It’s usually a better bet to go with climatology because the averages favor a Cape Verde hurricane forming in the East to Central Atlantic recurving before the Caribbean. I agree that patterns tend to repeat themselves with hurricanes and if analog patterns can be identified it is more valuable than just climatology.

Euro for many runs showing a strong hurricane and when it’s that persistent with a strong system it’s deserves attention. JMHO


It's always a better bet to go with statistical basin climo. That I will ALWAYS agree with. But in the approach to peak season you always lend a wary eye to what a model like the Euro has been showing.
1 likes   

User avatar
chaser1
S2K Supporter
S2K Supporter
Posts: 5476
Age: 64
Joined: Sat Oct 15, 2005 5:59 pm
Location: Longwood, Fl

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2050 Postby chaser1 » Fri Sep 04, 2020 9:48 pm

Blown Away wrote:
cheezyWXguy wrote:
toad strangler wrote:
Well documented earlier today here but the GFS has laughed it off since then, as have the climo watchers.

I think you’re saying that in jest, but as a general statement, I don’t understand the fixation on climatology I sometimes see. Hurricanes don’t follow climatology, climatology follows hurricanes. It is not only a historical average of past storm tracks, but also a historical average of the environmental factors that steered the storms. If climatology had a significant impact on individual events, all Cape Verde storms would recurve and the CLP5 would be the most accurate tool we have. If someone can explain climatology’s relevance to forecasting individual events, I am open to hearing it, but I am of the opinion that it is much more valuable to use historical analogs of similar setups than a single historical average.


It’s usually a better bet to go with climatology because the averages favor a Cape Verde hurricane forming in the East to Central Atlantic recurving before the Caribbean. I agree that patterns tend to repeat themselves with hurricanes and if analog patterns can be identified it is more valuable than just climatology.

Euro for many runs showing a strong hurricane and when it’s that persistent with a strong system it’s deserves attention. JMHO


Good points all around. Take the best of one or two really good matched analog seasons - great. Or take 100 years of rock solid climo - wonderful. Either could well steer us toward a likely solution going forward. However, in a year such as this where rapid development has been primarily limited to baroclinic systems, I don't buy into that 90% rule that says "if it fits climo AND that's what the models say.... then that's what'll probably happen. THIS year...... I might give a 50/50 nod to that combination of outcome, and I'm being generous there.

It's all pretty easy really. If we have one or more T.S.'s by Monday morning, they'll probably head out to sea. On the other hand, if these globs of disorganization are still just percolating while waiting for a new easterly surge to shove 'em west a bit...... ya might just wanna reach out to family & friends in P.R. or the Bahamas and suggest they start picking up the yard a little bit.
2 likes   
Andy D

(For official information, please refer to the NHC and NWS products.)

User avatar
Blown Away
S2K Supporter
S2K Supporter
Posts: 10148
Joined: Wed May 26, 2004 6:17 am

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2051 Postby Blown Away » Fri Sep 04, 2020 9:55 pm

That being said, Climo heavily favors a W moving Cape Verde Hurricane won’t landfall in the CONUS after Sept 21... So in a few more days any wave moving off Africa and forming into a hurricane before the Caribbean will likely not be a issue for the CONUS...
0 likes   
Hurricane Eye Experience: David 79, Irene 99, Frances 04, Jeanne 04, Wilma 05... EYE COMING MY WAY IN 2024…
Hurricane Brush Experience: Andrew 92, Erin 95, Floyd 99, Matthew 16, Irma 17, Ian 22, Nicole 22…

LarryWx
S2K Supporter
S2K Supporter
Posts: 6325
Joined: Sun Sep 07, 2003 2:04 pm
Location: GA

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2052 Postby LarryWx » Fri Sep 04, 2020 9:57 pm

toad strangler wrote:
Blown Away wrote:12z EURO shows a wave moving of Africa @Tuesday and beelining west across Atlantic as a Cape Verde major hurricane. Bermuda may be only threat from all the current activity, this potential wave could be a problem.


https://www.tropicaltidbits.com/analysi ... 0412&fh=12


Well documented earlier today here but the GFS has said no since then, as have the climo watchers.


Please clarify why you said the “climo watchers” said “no”. To whom are you referring and what exactly did they say?
0 likes   
Personal Forecast Disclaimer:
The posts in this forum are NOT official forecasts and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or storm2k.org. For official information, please refer to the NHC and NWS products.

LarryWx
S2K Supporter
S2K Supporter
Posts: 6325
Joined: Sun Sep 07, 2003 2:04 pm
Location: GA

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2053 Postby LarryWx » Fri Sep 04, 2020 10:12 pm

cheezyWXguy wrote:
toad strangler wrote:
Blown Away wrote:12z EURO shows a wave moving of Africa @Tuesday and beelining west across Atlantic as a Cape Verde major hurricane. Bermuda may be only threat from all the current activity, this potential wave could be a problem.


https://www.tropicaltidbits.com/analysi ... 0412&fh=12


Well documented earlier today here but the GFS has laughed it off since then, as have the climo watchers.

I think you’re saying that in jest, but as a general statement, I don’t understand the fixation on climatology I sometimes see. Hurricanes don’t follow climatology, climatology follows hurricanes. It is not only a historical average of past storm tracks, but also a historical average of the environmental factors that steered the storms. If climatology had a significant impact on individual events, all Cape Verde storms would recurve and the CLP5 would be the most accurate tool we have. If someone can explain climatology’s relevance to forecasting individual events, I am open to hearing it, but I am of the opinion that it is much more valuable to use historical analogs of similar setups than a single historical average.


Climo/historical stats should be used as a tool to assess probabilities as opposed to using it alone as a crystal ball. It sounds like you think that folks are using it alone and as if it were a crystal ball. I use it as a tool, a very useful tool, but just a tool that I use along with model consensus, teleconnections, and indices. I don’t see using climo/stats as a tool hurts in any way.

Example: per history, not a single TC genesis after 9/25 E of 50W on record since 1851 has reached the CONUS. Therefore, if there were a TC at, say, 30W, on 10/1, I’m going to bet very heavily that it won’t reach the CONUS. I’d give it no more than a 1% chance of doing so and would forecast with high confidence that it won’t.

Edit: Example 2: If history shows that 99% of all TCs crossing 70W between 24N and 26N moving westward or WNW in August later hit FL, then I’m likely going to forecast a FL hit with high confidence absent model consensus vehemently showing a sharp recurve and miss.
Last edited by LarryWx on Fri Sep 04, 2020 10:30 pm, edited 2 times in total.
4 likes   
Personal Forecast Disclaimer:
The posts in this forum are NOT official forecasts and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or storm2k.org. For official information, please refer to the NHC and NWS products.

User avatar
toad strangler
S2K Supporter
S2K Supporter
Posts: 4543
Joined: Sun Jul 28, 2013 3:09 pm
Location: Earth
Contact:

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2054 Postby toad strangler » Fri Sep 04, 2020 10:26 pm

LarryWx wrote:
toad strangler wrote:
Blown Away wrote:12z EURO shows a wave moving of Africa @Tuesday and beelining west across Atlantic as a Cape Verde major hurricane. Bermuda may be only threat from all the current activity, this potential wave could be a problem.


https://www.tropicaltidbits.com/analysi ... 0412&fh=12


Well documented earlier today here but the GFS has said no since then, as have the climo watchers.


Please clarify why you said the “climo watchers” said “no”. To whom are you referring and what exactly did they say?


What are you, the FBI? :lol: Nothing I said was meant in a negative way as I'm a climo guy mostly. The answers are on the last page or so.
1 likes   

User avatar
cheezyWXguy
Category 5
Category 5
Posts: 6108
Joined: Mon Feb 13, 2006 12:29 am
Location: Dallas, TX

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2055 Postby cheezyWXguy » Fri Sep 04, 2020 10:28 pm

LarryWx wrote:
cheezyWXguy wrote:
toad strangler wrote:
Well documented earlier today here but the GFS has laughed it off since then, as have the climo watchers.

I think you’re saying that in jest, but as a general statement, I don’t understand the fixation on climatology I sometimes see. Hurricanes don’t follow climatology, climatology follows hurricanes. It is not only a historical average of past storm tracks, but also a historical average of the environmental factors that steered the storms. If climatology had a significant impact on individual events, all Cape Verde storms would recurve and the CLP5 would be the most accurate tool we have. If someone can explain climatology’s relevance to forecasting individual events, I am open to hearing it, but I am of the opinion that it is much more valuable to use historical analogs of similar setups than a single historical average.


Climo/historical stats should be used as a tool to assess probabilities as opposed to using it alone as a crystal ball. It sounds like you think that folks are using it alone and as if it were a crystal ball. I use it as a tool, a very useful tool, but just a tool that I use along with model consensus, teleconnections, and indices. I don’t see using climo/stats as a tool hurts in any way.

Example: per history, not a single TC genesis after 9/25 E of 50W on record since 1851 has reached the CONUS. Therefore, if there were a TC at, say, 30W, on 10/1, I’m going to bet very heavily that it won’t reach the CONUS. I’d give it no more than a 1% chance of doing so and would forecast with high confidence that it won’t.

I don’t think it hurts to use climo as a tool, like you say. But I see a tendency, sometimes, to subvert the need for analysis of current conditions and how they may deviate from the statistical norm, or narrow the historical scope to similar setups. Nothing has ever happened before, until it happened. No season had more storms than 1933 until 2005 came around, and no season had more storms than that (so far) until 2020. No cat5 had formed in July before emily, and no cat5 existed east of Hugo until Lorenzo. Climatology is a historical average of individual anomalies. It should be used as a rule of thumb, no less, no more.
5 likes   

LarryWx
S2K Supporter
S2K Supporter
Posts: 6325
Joined: Sun Sep 07, 2003 2:04 pm
Location: GA

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2056 Postby LarryWx » Fri Sep 04, 2020 10:33 pm

toad strangler wrote:
LarryWx wrote:
toad strangler wrote:
Well documented earlier today here but the GFS has said no since then, as have the climo watchers.


Please clarify why you said the “climo watchers” said “no”. To whom are you referring and what exactly did they say?


What are you, the FBI? :lol: Nothing I said was meant in a negative way as I'm a climo guy mostly. The answers are on the last page or so.


Lol. I asked because I mentioned some historical stats/climo and didn’t know if you were referring to me among others since I didn’t say “no”. Apparently you weren’t.
0 likes   
Personal Forecast Disclaimer:
The posts in this forum are NOT official forecasts and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or storm2k.org. For official information, please refer to the NHC and NWS products.

LarryWx
S2K Supporter
S2K Supporter
Posts: 6325
Joined: Sun Sep 07, 2003 2:04 pm
Location: GA

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2057 Postby LarryWx » Fri Sep 04, 2020 10:42 pm

cheezyWXguy wrote:
LarryWx wrote:
cheezyWXguy wrote:I think you’re saying that in jest, but as a general statement, I don’t understand the fixation on climatology I sometimes see. Hurricanes don’t follow climatology, climatology follows hurricanes. It is not only a historical average of past storm tracks, but also a historical average of the environmental factors that steered the storms. If climatology had a significant impact on individual events, all Cape Verde storms would recurve and the CLP5 would be the most accurate tool we have. If someone can explain climatology’s relevance to forecasting individual events, I am open to hearing it, but I am of the opinion that it is much more valuable to use historical analogs of similar setups than a single historical average.


Climo/historical stats should be used as a tool to assess probabilities as opposed to using it alone as a crystal ball. It sounds like you think that folks are using it alone and as if it were a crystal ball. I use it as a tool, a very useful tool, but just a tool that I use along with model consensus, teleconnections, and indices. I don’t see using climo/stats as a tool hurts in any way.

Example: per history, not a single TC genesis after 9/25 E of 50W on record since 1851 has reached the CONUS. Therefore, if there were a TC at, say, 30W, on 10/1, I’m going to bet very heavily that it won’t reach the CONUS. I’d give it no more than a 1% chance of doing so and would forecast with high confidence that it won’t.

I don’t think it hurts to use climo as a tool, like you say. But I see a tendency, sometimes, to subvert the need for analysis of current conditions and how they may deviate from the statistical norm, or narrow the historical scope to similar setups. Nothing has ever happened before, until it happened. No season had more storms than 1933 until 2005 came around, and no season had more storms than that (so far) until 2020. No cat5 had formed in July before emily, and no cat5 existed east of Hugo until Lorenzo. Climatology is a historical average of individual anomalies. It should be used as a rule of thumb, no less, no more.


If history shows, say, that 100% of, say, 50 TCs crossing 70W between 24N and 28N moving westward at 20+ mph in August later hit the SE US, then I’m likely going to forecast a SE hit with high confidence absent model consensus vehemently showing a sharp recurve and miss. And if they show there is a miss, I’m going to take that into consideration but examine those models extra closely to see if there’s something glaring telling me they’re off.

Do you see anything wrong with using stats the way I did here?

Edit: Similar to the last few runs, the 0Z ICON has a strong low form in the E ATL ~9/10.
2 likes   
Personal Forecast Disclaimer:
The posts in this forum are NOT official forecasts and should not be used as such. They are just the opinion of the poster and may or may not be backed by sound meteorological data. They are NOT endorsed by any professional institution or storm2k.org. For official information, please refer to the NHC and NWS products.

sma10
Category 5
Category 5
Posts: 1710
Joined: Tue Aug 02, 2005 1:13 pm

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2058 Postby sma10 » Fri Sep 04, 2020 11:00 pm

Ok.. so ICON says one week from tonight, we'll have 5 tropical cyclones: 2 doing a dance in the central atlantic; TS just south of CV islands; 1 just off the Carolina coast; and 1 just west of Naples.
1 likes   

User avatar
cheezyWXguy
Category 5
Category 5
Posts: 6108
Joined: Mon Feb 13, 2006 12:29 am
Location: Dallas, TX

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2059 Postby cheezyWXguy » Fri Sep 04, 2020 11:09 pm

LarryWx wrote:
cheezyWXguy wrote:
LarryWx wrote:
Climo/historical stats should be used as a tool to assess probabilities as opposed to using it alone as a crystal ball. It sounds like you think that folks are using it alone and as if it were a crystal ball. I use it as a tool, a very useful tool, but just a tool that I use along with model consensus, teleconnections, and indices. I don’t see using climo/stats as a tool hurts in any way.

Example: per history, not a single TC genesis after 9/25 E of 50W on record since 1851 has reached the CONUS. Therefore, if there were a TC at, say, 30W, on 10/1, I’m going to bet very heavily that it won’t reach the CONUS. I’d give it no more than a 1% chance of doing so and would forecast with high confidence that it won’t.

I don’t think it hurts to use climo as a tool, like you say. But I see a tendency, sometimes, to subvert the need for analysis of current conditions and how they may deviate from the statistical norm, or narrow the historical scope to similar setups. Nothing has ever happened before, until it happened. No season had more storms than 1933 until 2005 came around, and no season had more storms than that (so far) until 2020. No cat5 had formed in July before emily, and no cat5 existed east of Hugo until Lorenzo. Climatology is a historical average of individual anomalies. It should be used as a rule of thumb, no less, no more.


If history shows, say, that 100% of, say, 50 TCs crossing 70W between 24N and 28N moving westward at 20+ mph in August later hit the SE US, then I’m likely going to forecast a SE hit with high confidence absent model consensus vehemently showing a sharp recurve and miss. And if they show there is a miss, I’m going to take that into consideration but examine those models extra closely to see if there’s something glaring telling me they’re off.

Do you see anything wrong with using stats the way I did here?

Edit: Similar to the last few runs, the 0Z ICON has a strong low form in the E ATL ~9/10.


I don’t think so, that sounds pretty reasonable to me. I think a more unreasonable scenario would be using climo in that situation while refusing to acknowledge any model guidance that shows a deviation. My prior point was basically that just because 100% of storms in a sample behaved the same way in the past doesn’t guarantee that each one in the future will as well. Considering your viewpoint is to use climo as a tool and not as gospel, I would say we are largely in agreement
4 likes   

Fancy1001
Category 1
Category 1
Posts: 452
Joined: Thu Jul 16, 2020 10:16 pm

Re: 2020 Global Model Runs Discussion (Out thru day 16)

#2060 Postby Fancy1001 » Sat Sep 05, 2020 1:32 am

Am I the only one that thinks this whole discussion belongs in a separate thread.
2 likes   


Return to “Talkin' Tropics”

Who is online

Users browsing this forum: jconsor, jlauderdal, Ulf and 30 guests