2021 Indicators: SST's / SAL / MSLP / Shear / Steering / Instability / Sat Images

[Teban54]

I remember there were extensive discussions last year regarding CFS long-range forecasts. The consensus was that CFS as a climate model is designed to forecast large-scale patterns, not individual storms. People can look at the storm tracks predicted on CFS, but keep in mind they probably have low skill.

Edit: Here's the start of the chatters on CFS last year: viewtopic.php?f=31&t=120763&start=1900

Latest CFS shows little to nothing from here to the third week of August, followed by moderate activity for most of September and essentially shutting down in the MDR once we hit October, though there are quite a few systems in the subtropics in October that are hard to tell on that model if they're tropical or not.

Not trying to put the blame on Hammy - there's nothing wrong with simply posting model outputs - but just to show one should take them with a grain of salt.

Last year's CFS did correctly predict waves coming off too far north during peak season. However, it showed little to no Gulf activity, and we all know how that turned out.

[drezee]

 https://twitter.com/andretrotter/status/1421336938766872576

[Hammy]

I remember there were extensive discussions last year regarding CFS long-range forecasts. The consensus was that CFS as a climate model is designed to forecast large-scale patterns, not individual storms. People can look at the storm tracks predicted on CFS, but keep in mind they probably have low skill.

Edit: Here's the start of the chatters on CFS last year: viewtopic.php?f=31&t=120763&start=1900

Latest CFS shows little to nothing from here to the third week of August, followed by moderate activity for most of September and essentially shutting down in the MDR once we hit October, though there are quite a few systems in the subtropics in October that are hard to tell on that model if they're tropical or not.

Not trying to put the blame on Hammy - there's nothing wrong with simply posting model outputs - but just to show one should take them with a grain of salt.

Last year's CFS did correctly predict waves coming off too far north during peak season. However, it showed little to no Gulf activity, and we all know how that turned out.

Something I learned from last year is it seems to have difficulty resolving storms in the Gulf given the resolution--particularly the western Gulf. Smaller storms will just sort of blend into the generally lower pressure in the area. I've seen it show major-ish looking hurricanes go into the gulf and then just turn into large, diffuse areas that didn't really go anywhere. It did, however, show a very active Caribbean in October and November which unfortunately played out.

It did correctly show two instances of double Gulf storms months out--one of those was Laura and Marco, and the other turned out to be Beta's incipient low while Sally was active.

CFS (after seemingly having a limit on how far out it would go) is back up and running, the model shows the Atlantic remaining largely quiet for the first half of August and then picking up after that. Looks like a burst of MDR activity in the last week of August into early September, a short lull, and then more continuous activity with the MDR remaining in play through mid-October.

Most MDR storms seem to recurve, including 4-5 long trackers, but a few make it to the Bahamas. Gulf/Caribbean maps are not loading beyond about 400h but there were a few storms that crossed over Florida and appeared to strengthen after so that could be a hot spot.

I thought there was a ridge predicted to be over Eastern Canada, making recurves less likely (if I recall Eric Webb was the one who showed that first)? Also how exactly is the CFS able to forecast storm tracks that far out (I did not know CFS even had that ability until now to be honest)

I believe the ridging over eastern Canada was the anomalies--that said the recent CFS runs did show quite a few storms making it to the Bahamas and Florida which means there will still be ridging from time to time where there isn't normally. Part of the reason for more recurves is that the MDR itself appears to be more active than last year, and stronger storms farther east are generally more likely to curve out to sea.

Map is here, it takes some getting used to know how to read it as it's low resolution, and if you take it at face value then everything's at tropical storm intensity. https://www.weatheronline.co.uk/cgi-bin/expertcharts?LANG=en&MENU=0000000000&CONT=noat&MODELL=cfs&MODELLTYP=1&BASE=-&VAR=pslv&HH=12&ZOOM=0&WMO=&ARCHIV=0&LOOP=1&RES=0

This is unfortunately the only source that actually has the 12h pressure maps--everywhere else just shows the weekly or monthly totals/averages. It is also subject to change from day to day so anyone who wants to use it to get an idea, it's best to watch it for about a week or so and essentially average it.

[Shell Mound]

...

The same model also suggests that most systems remain fairly weak until reaching the subtropics, perhaps indicative of a relatively stable MDR.

I believe the ridging over eastern Canada was the anomalies--that said the recent CFS runs did show quite a few storms making it to the Bahamas and Florida...

I do not see any well-defined lows reaching the Bahamas or Florida via the MDR. May you please highlight some of them? I don’t see them on the CFS.

[Shell Mound]

 https://twitter.com/AndyHazelton/status/1421117726123298818



 https://twitter.com/JoeKotulak/status/1421133302983974914



 https://twitter.com/AndyHazelton/status/1421147007054401537



 https://twitter.com/dmorris9661/status/1421148746411352071



 https://twitter.com/AndyHazelton/status/1421153543533338626

[Shell Mound]

The relatively high amplitude of the MJO suggests that most of this year’s activity could come in relatively short but intense “bursts.” This also implies that we could well see multiple systems simultaneously. Therefore, the window for potential impacts will be narrower than it was last year, which featured a lower-amplitude MJO and thus more “constant” activity. If activity manages to coincide with a period of weak ridging, then it could end up defining much of the season. Note that the CFS suggests the upcoming “burst” of activity in late August and early September will coincide with a suppressed WAR, so every system that develops in the MDR and strengthens in the subtropics ends up curving OTS or toward Bermuda and/or the Canadian Maritimes. Afterward, the run suggests that most of September and the first half of October will be relatively inactive. So the potential for the CONUS to escape serious impacts is much higher this year than it was last year. Overall, the CFS has handled the MJO relatively well this year, particularly in July, so in my view it holds credibility.

[Shell Mound]

Link: https://youtu.be/jARp26WeLuU

[SteveM]

The relatively high amplitude of the MJO suggests that most of this year’s activity could come in relatively short but intense “bursts.” This also implies that we could well see multiple systems simultaneously. Therefore, the window for potential impacts will be narrower than it was last year, which featured a lower-amplitude MJO and thus more “constant” activity. If activity manages to coincide with a period of weak ridging, then it could end up defining much of the season. Note that the CFS suggests the upcoming “burst” of activity in late August and early September will coincide with a suppressed WAR, so every system that develops in the MDR and strengthens in the subtropics ends up curving OTS or toward Bermuda and/or the Canadian Maritimes. Afterward, the run suggests that most of September and the first half of October will be relatively inactive. So the potential for the CONUS to escape serious impacts is much higher this year than it was last year. Overall, the CFS has handled the MJO relatively well this year, particularly in July, so in my view it holds credibility.

Is this a season cancel?

[AlphaToOmega]

I split into two main MJO classes: MJSN (May-July-September-November) and JAOD (June-August-October-December). If more storms form in June, August, October, and December, it is a JAOD season; if not, it is an MJSN season. From the looks of it, it seems like 2021 is going to be a JAOD season. Some other JAOD seasons include 2017, 2014, 2012, 2010, 2009, 2004, 2003, 2001, 2000, and 1996. What is interesting to not is that in nearly every JAOD season since 1967, with the exceptions of 1968 and 1982, August is significantly more active than June. Enjoy the quiet period while it lasts

[crownweather]

Not sure if you knew that this site also has the CFS model out to 768 hrs. I use it more as it tends to be easier to load up and also it has both the 00Z and the 12Z CFS model.

https://www.pivotalweather.com/model.ph ... &dpdt=&mc=

I remember there were extensive discussions last year regarding CFS long-range forecasts. The consensus was that CFS as a climate model is designed to forecast large-scale patterns, not individual storms. People can look at the storm tracks predicted on CFS, but keep in mind they probably have low skill.

Edit: Here's the start of the chatters on CFS last year: viewtopic.php?f=31&t=120763&start=1900

Latest CFS shows little to nothing from here to the third week of August, followed by moderate activity for most of September and essentially shutting down in the MDR once we hit October, though there are quite a few systems in the subtropics in October that are hard to tell on that model if they're tropical or not.

Not trying to put the blame on Hammy - there's nothing wrong with simply posting model outputs - but just to show one should take them with a grain of salt.

Last year's CFS did correctly predict waves coming off too far north during peak season. However, it showed little to no Gulf activity, and we all know how that turned out.

Something I learned from last year is it seems to have difficulty resolving storms in the Gulf given the resolution--particularly the western Gulf. Smaller storms will just sort of blend into the generally lower pressure in the area. I've seen it show major-ish looking hurricanes go into the gulf and then just turn into large, diffuse areas that didn't really go anywhere. It did, however, show a very active Caribbean in October and November which unfortunately played out.

It did correctly show two instances of double Gulf storms months out--one of those was Laura and Marco, and the other turned out to be Beta's incipient low while Sally was active.

CFS (after seemingly having a limit on how far out it would go) is back up and running, the model shows the Atlantic remaining largely quiet for the first half of August and then picking up after that. Looks like a burst of MDR activity in the last week of August into early September, a short lull, and then more continuous activity with the MDR remaining in play through mid-October.

Most MDR storms seem to recurve, including 4-5 long trackers, but a few make it to the Bahamas. Gulf/Caribbean maps are not loading beyond about 400h but there were a few storms that crossed over Florida and appeared to strengthen after so that could be a hot spot.

I thought there was a ridge predicted to be over Eastern Canada, making recurves less likely (if I recall Eric Webb was the one who showed that first)? Also how exactly is the CFS able to forecast storm tracks that far out (I did not know CFS even had that ability until now to be honest)

I believe the ridging over eastern Canada was the anomalies--that said the recent CFS runs did show quite a few storms making it to the Bahamas and Florida which means there will still be ridging from time to time where there isn't normally. Part of the reason for more recurves is that the MDR itself appears to be more active than last year, and stronger storms farther east are generally more likely to curve out to sea.

Map is here, it takes some getting used to know how to read it as it's low resolution, and if you take it at face value then everything's at tropical storm intensity. https://www.weatheronline.co.uk/cgi-bin/expertcharts?LANG=en&MENU=0000000000&CONT=noat&MODELL=cfs&MODELLTYP=1&BASE=-&VAR=pslv&HH=12&ZOOM=0&WMO=&ARCHIV=0&LOOP=1&RES=0

This is unfortunately the only source that actually has the 12h pressure maps--everywhere else just shows the weekly or monthly totals/averages. It is also subject to change from day to day so anyone who wants to use it to get an idea, it's best to watch it for about a week or so and essentially average it.

[aspen]

CFS (after seemingly having a limit on how far out it would go) is back up and running, the model shows the Atlantic remaining largely quiet for the first half of August and then picking up after that. Looks like a burst of MDR activity in the last week of August into early September, a short lull, and then more continuous activity with the MDR remaining in play through mid-October.

Most MDR storms seem to recurve, including 4-5 long trackers, but a few make it to the Bahamas. Gulf/Caribbean maps are not loading beyond about 400h but there were a few storms that crossed over Florida and appeared to strengthen after so that could be a hot spot.

I hope it’s right about multiple recurving long-trackers. They’re fun to track, will ramp up a bunch of ACE, and most importantly, will have very limited to no impacts. Hopefully none sink large ships like Joaquin or Lorenzo.

Very interesting that it’s predicting MDR activity in October. Compare that to last year, where it correctly showed the MDR shut down after September. It is quite far out so we’ll have to wait and see, but the August into September forecast seems very reasonable given many indicators (strong WAM, improving MDR SSTs) and climatology.

[Shell Mound]

The relatively high amplitude of the MJO suggests that most of this year’s activity could come in relatively short but intense “bursts.” This also implies that we could well see multiple systems simultaneously. Therefore, the window for potential impacts will be narrower than it was last year, which featured a lower-amplitude MJO and thus more “constant” activity. If activity manages to coincide with a period of weak ridging, then it could end up defining much of the season. Note that the CFS suggests the upcoming “burst” of activity in late August and early September will coincide with a suppressed WAR, so every system that develops in the MDR and strengthens in the subtropics ends up curving OTS or toward Bermuda and/or the Canadian Maritimes. Afterward, the run suggests that most of September and the first half of October will be relatively inactive. So the potential for the CONUS to escape serious impacts is much higher this year than it was last year. Overall, the CFS has handled the MJO relatively well this year, particularly in July, so in my view it holds credibility.

Is this a season cancel?

I am simply highlighting the possibility. Do you have evidence to the contrary? The CFS is suggesting weak ridging while storms form during August.

[Category5Kaiju]

How well did CFS predict storms and their general tracks out in time in years from 2017-2019?

[AlphaToOmega]

The CFSv2 was pretty bad with MJO forecasts. It suggested a burst of activity in July followed by a lull in August. It seems like the opposite is verifying.

[toad strangler]

The relatively high amplitude of the MJO suggests that most of this year’s activity could come in relatively short but intense “bursts.” This also implies that we could well see multiple systems simultaneously. Therefore, the window for potential impacts will be narrower than it was last year, which featured a lower-amplitude MJO and thus more “constant” activity. If activity manages to coincide with a period of weak ridging, then it could end up defining much of the season. Note that the CFS suggests the upcoming “burst” of activity in late August and early September will coincide with a suppressed WAR, so every system that develops in the MDR and strengthens in the subtropics ends up curving OTS or toward Bermuda and/or the Canadian Maritimes. Afterward, the run suggests that most of September and the first half of October will be relatively inactive. So the potential for the CONUS to escape serious impacts is much higher this year than it was last year. Overall, the CFS has handled the MJO relatively well this year, particularly in July, so in my view it holds credibility.

Is this a season cancel?

I am simply highlighting the possibility. Do you have evidence to the contrary? The CFS is suggesting weak ridging while storms form during August.

There isn’t any evidence of future tracks. Model runs are not evidence. I suggest watching the Sudduth blog from yesterday. Very informative

[SFLcane]

I see we’re now making assumptions there’s nothing to worry about this season based on 1,000 hr model runs.

Come on guys… seriously?

Maybe I need a small break til things actually pick up.

[Cat5James]

The relatively high amplitude of the MJO suggests that most of this year’s activity could come in relatively short but intense “bursts.” This also implies that we could well see multiple systems simultaneously. Therefore, the window for potential impacts will be narrower than it was last year, which featured a lower-amplitude MJO and thus more “constant” activity. If activity manages to coincide with a period of weak ridging, then it could end up defining much of the season. Note that the CFS suggests the upcoming “burst” of activity in late August and early September will coincide with a suppressed WAR, so every system that develops in the MDR and strengthens in the subtropics ends up curving OTS or toward Bermuda and/or the Canadian Maritimes. Afterward, the run suggests that most of September and the first half of October will be relatively inactive. So the potential for the CONUS to escape serious impacts is much higher this year than it was last year. Overall, the CFS has handled the MJO relatively well this year, particularly in July, so in my view it holds credibility.

Is this a season cancel?

I am simply highlighting the possibility. Do you have evidence to the contrary? The CFS is suggesting weak ridging while storms form during August.

We can not accurately predict ridging/steering a week out, let alone a month + out. If we could 240+ hours out on models wouldn’t be considered “fantasyland”

[aspen]

I see we’re now making assumptions there’s nothing to worry about this season based on 1,000 hr model runs.

Come on guys… seriously?

Maybe I need a small break til things actually pick up.

Quite the opposite for me. I don’t like how close some of those CFS storms get to the East Coast. Yeah it’s far out and things can change drastically, but there’s potential that I’ll need to keep an eye out for come the last two weeks of August. Also, the super long range CFS has an October East Coast Sweeper with the track of Irene and the time frame of Sandy…please don’t verify.

[Category5Kaiju]

I see we’re now making assumptions there’s nothing to worry about this season based on 1,000 hr model runs.

Come on guys… seriously?

Maybe I need a small break til things actually pick up.

Quite the opposite for me. I don’t like how close some of those CFS storms get to the East Coast. Yeah it’s far out and things can change drastically, but there’s potential that I’ll need to keep an eye out for come the last two weeks of August. Also, the super long range CFS has an October East Coast Sweeper with the track of Irene and the time frame of Sandy…please don’t verify.

I'm keeping an eagle eye out in Eastern Virginia; it's been nearly 18 years since Isabel and my friends and I are hoping nothing like that happens this year

[cycloneye]

I am simple with this. I dont like the peak of season for the NE Caribbean because of the Atlantic Niño.