Medium Range-Long Range thread

[Rainshadow]

Gotta love those 1994 dates.  Maybe we can get some epic ice......gulp

The ice storms in 1994 already occurred by this calendar date. February is a snow month and even in 1994 the events were snow and sleet in the local area.   The snowpack was so much greater in 94 than it is this year, cuts down on the ice potential.

[snowlurker]

the gefs 6-14 day outlooks should make adam pretty happy

[/img]

I'm kinda OCD and like to cross-reference the analog dates with KU storms. While it's reallllyyy tenuous, I noticed three of four of the analog dates are within 3-6 days of KU storms. Maybe it's just a factor of KU storms occurring more frequently in February historically or maybe it's a nice sign. In any case, it's something to "track" - ha ha.

[Quakertown needs snow]

Feb is our local snow month, lock n load.

Cold will be here!

Get the snow blowers gassed up!

Even yours Tony .

End.

[hazwoper]

The ice storms in 1994 already occurred by this calendar date. February is a snow month and even in 1994 the events were snow and sleet in the local area.   The snowpack was so much greater in 94 than it is this year, cuts down on the ice potential.

 

ah, well that is good to know.  forgot exactly when they occurred, just that they were terrible.

[Rainshadow]

Feb is our local snow month, lock n load.

Cold will be here!

Get the snow blowers gassed up!

Even yours Tony .

End.

That 2011 gas is locked and loaded for action.

[Rainshadow]

ah, well that is good to know.  forgot exactly when they occurred, just that they were terrible.

Yeah they were.  Thankfully we finally got cold enough at this point of that season for "not ice".

[NJwinter23]

Maybe someone more experienced can chime in on this, but ive never seen the ensembles do such a drastic change as early as day 4/5/6 like today's european ensembles did. I mean the op run started the drastic shift for the middle of next week obviously, but for the ensembles to jump on board hook line and sinker, it was pretty drastic for so early in the forecast period. 

 

anyhow, its a more muted warm up in the middle of next week and an over all colder look for the east.

[Harbourton]

Maybe someone more experienced can chime in on this, but ive never seen the ensembles do such a drastic change as early as day 4/5/6 like today's european ensembles did. I mean the op run started the drastic shift for the middle of next week obviously, but for the ensembles to jump on board hook line and sinker, it was pretty drastic for so early in the forecast period. 

 

anyhow, its a more muted warm up in the middle of next week and an over all colder look for the east.

Just a novice, but I think the models are reacting to  long range forecasting per HM and Adam that

 

1. the stratospheric warming caused

 

2, The MJO wave to propagate and increase in amplitude to Phase 8+1

 

Sure there are other variables, 

[HM]

Maybe someone more experienced can chime in on this, but ive never seen the ensembles do such a drastic change as early as day 4/5/6 like today's european ensembles did. I mean the op run started the drastic shift for the middle of next week obviously, but for the ensembles to jump on board hook line and sinker, it was pretty drastic for so early in the forecast period. 

 

anyhow, its a more muted warm up in the middle of next week and an over all colder look for the east.

Thanks for bringing this to the forefront because I gave a pretty stern warning about poor modeling within the day 3-10 range (I think beyond 10 speaks for itself, haha). Times of poor NWP have to do with battling regimes, sudden changes in global AAM and stratospheric/solar disturbances.

 

There is no denying the amazing shift in GLAAM with values approaching winters with strong El Nino conditions in place. The MJO signal is clearly getting muted on the RMMs by the h2 battle but it is clear that the MJO kicked its ass and is propagating east (just slower than what the consensus thought 2 weeks ago). The jet has extended and the remnant h2 westerly over the equatorial region got shifted toward the SH / S. America. 

 

There is also no denying a downwelling -u anomaly from the 10-30mb layer to the tropopause over the North Pole. This will reach the troposphere as we start February and it will likely lead to even more blocking than what's being modeled.

 

There is a stronger than normal chance for a significant snowfall Feb 5-10 (yes I know climo says snow is most likely then). The storm being modeled beforehand is certainly possible; but with the PV displacement forecasted then, it is quite possible the "shunt" scenario is the best one. However, with the arrival of the STJ Feb 2-5, we certainly could get rolling in the snow department sooner than I have been saying.

[Harbourton]

Thanks for bringing this to the forefront because I gave a pretty stern warning about poor modeling within the day 3-10 range (I think beyond 10 speaks for itself, haha). Times of poor NWP have to do with battling regimes, sudden changes in global AAM and stratospheric/solar disturbances.

 

There is no denying the amazing shift in GLAAM with values approaching winters with strong El Nino conditions in place. The MJO signal is clearly getting muted on the RMMs by the h2 battle but it is clear that the MJO kicked its ass and is propagating east (just slower than what the consensus thought 2 weeks ago). The jet has extended and the remnant h2 westerly over the equatorial region got shifted toward the SH / S. America. 

 

There is also no denying a downwelling -u anomaly from the 10-30mb layer to the tropopause over the North Pole. This will reach the troposphere as we start February and it will likely lead to even more blocking than what's being modeled.

 

There is a stronger than normal chance for a significant snowfall Feb 5-10 (yes I know climo says snow is most likely then). The storm being modeled beforehand is certainly possible; but with the PV displacement forecasted then, it is quite possible the "shunt" scenario is the best one. However, with the arrival of the STJ Feb 2-5, we certainly could get rolling in the snow department sooner than I have been saying.

Wow! I'm not going to even ask for further details except that we must see return of the southern jet to get a plow able snow fall.

[famartin]

ah, well that is good to know.  forgot exactly when they occurred, just that they were terrible.

 

The worst ones were the 7th (worst in PECO history as I recall from Mt. Holly's narrative) and the 17th.  There were others on the 4th and 28th, I think, but they definitely weren't as severe.

[SnowGoose69]

Thanks for bringing this to the forefront because I gave a pretty stern warning about poor modeling within the day 3-10 range (I think beyond 10 speaks for itself, haha). Times of poor NWP have to do with battling regimes, sudden changes in global AAM and stratospheric/solar disturbances.

 

There is no denying the amazing shift in GLAAM with values approaching winters with strong El Nino conditions in place. The MJO signal is clearly getting muted on the RMMs by the h2 battle but it is clear that the MJO kicked its ass and is propagating east (just slower than what the consensus thought 2 weeks ago). The jet has extended and the remnant h2 westerly over the equatorial region got shifted toward the SH / S. America. 

 

There is also no denying a downwelling -u anomaly from the 10-30mb layer to the tropopause over the North Pole. This will reach the troposphere as we start February and it will likely lead to even more blocking than what's being modeled.

 

There is a stronger than normal chance for a significant snowfall Feb 5-10 (yes I know climo says snow is most likely then). The storm being modeled beforehand is certainly possible; but with the PV displacement forecasted then, it is quite possible the "shunt" scenario is the best one. However, with the arrival of the STJ Feb 2-5, we certainly could get rolling in the snow department sooner than I have been saying.

 

There is a pretty strong storm signal from 180-300 hours somewhere, but right now if that PV is over ERN Canada I cannot see anyway its an event for anyone but the TN Valley, SE, and Mid-Atlantic.

[NJwinter23]

Thanks for bringing this to the forefront because I gave a pretty stern warning about poor modeling within the day 3-10 range (I think beyond 10 speaks for itself, haha). Times of poor NWP have to do with battling regimes, sudden changes in global AAM and stratospheric/solar disturbances.

 

There is no denying the amazing shift in GLAAM with values approaching winters with strong El Nino conditions in place. The MJO signal is clearly getting muted on the RMMs by the h2 battle but it is clear that the MJO kicked its ass and is propagating east (just slower than what the consensus thought 2 weeks ago). The jet has extended and the remnant h2 westerly over the equatorial region got shifted toward the SH / S. America. 

 

There is also no denying a downwelling -u anomaly from the 10-30mb layer to the tropopause over the North Pole. This will reach the troposphere as we start February and it will likely lead to even more blocking than what's being modeled.

 

There is a stronger than normal chance for a significant snowfall Feb 5-10 (yes I know climo says snow is most likely then). The storm being modeled beforehand is certainly possible; but with the PV displacement forecasted then, it is quite possible the "shunt" scenario is the best one. However, with the arrival of the STJ Feb 2-5, we certainly could get rolling in the snow department sooner than I have been saying.

 

What you had mentioned with poor modeling a few days back was definitely in the back of my mind when I saw these runs come in this afternoon. I forgot the exact reasoning behind your statement, but it certainly seems to have panned out. I think you have correctly diagnosed the source of the drastic change as well possibly being the glaam spike. It all started in the pacific very early on in the run and progressed into western NA by day 4.

 

Of course I think cooler risk to any modelled warm ups were the way to go in the over all pattern anyway given so many signals pointing to the cold pattern.

 

PS, 30mb temps in the tropics appear to be baking again, what say you?

 

10mb is much colder though.

[hazwoper]

The worst ones were the 7th (worst in PECO history as I recall from Mt. Holly's narrative) and the 17th. There were others on the 4th and 28th, I think, but they definitely weren't as severe.

Yeah, I was actually in college at the time at Shippensburg. Was the week long state of emergency during those ice storms or later in Feb? What a crazy year it was!

[HM]

What you had mentioned with poor modeling a few days back was definitely in the back of my mind when I saw these runs come in this afternoon. I forgot the exact reasoning behind your statement, but it certainly seems to have panned out. I think you have correctly diagnosed the source of the drastic change as well possibly being the glaam spike. It all started in the pacific very early on in the run and progressed into western NA by day 4.

 

Of course I think cooler risk to any modelled warm ups were the way to go in the over all pattern anyway given so many signals pointing to the cold pattern.

 

PS, 30mb temps in the tropics appear to be baking again, what say you?

The 30mb warm anomaly has strengthened due to the subsidence wave of the MJO. This area of warming has grown more exceptional than before but this time it is confined to the region. It could mean we are in for a long duration of forcing that stays in the IO. It is quite possible we just loop around 8-1-2-3-4-2-3-4 etc. right through March.

[tombo82685]

Ok Harbourton, finally got around to the explanation of the roundy MJO. I know Adam, HM, and Mike may butcher this explanation and I'm probably wrong in some of this but this is what i have tried to grasp so far. The chart is dissected into the western hemisphere and eastern hemisphere. The western hemisphere is not all that important and has very little mjo waves that traverse it. The solid green line running parallel that i have a green arrow pointing to is the current day, (this sometimes doesn't update daily). The black numbers below are the phases, you know which phases are the good ones and the bad ones. The orange arrow pointing to the red dashed line is +OLR (outgoing longwave radiation). +OLR is associated with decreased in thunderstorm activity or decrease in clouds. solid red line which the red arrow is pointing to is the mjo wave ( -OLR) is increase in thunderstorm activity/clouds. The more the solid red circles inside the big circle the stornger the wave. Same goes for the dashed red line the more circles the stronger the subsidence is. Now the blue arrow pointing to the blue shading is an increase in rising air which is aided for thunderstorm development. The gray arrow pointing to the yellow shading is the downward propagation of air, aka subsidence, which leads to weakening of thunderstorms. Just like when you have a mega band in major snowstorm. You notice the mega band, then on the other side of the band is an area that very light in activity do to the sinking air behind it. The darker the blues, the more upward rising of air (more thunderstorm development). The darker the yellows (less/weakening of thunderstorms). Their are a lot of other things going on in this map like the kelvin waves which i haven't learned yet, but I'm sure Adam, HM, and Mike would explain.

 

 

[NJwinter23]

The 30mb warm anomaly has strengthened due to the subsidence wave of the MJO. This area of warming has grown more exceptional than before but this time it is confined to the region. It could mean we are in for a long duration of forcing that stays in the IO. It is quite possible we just loop around 8-1-2-3-4-2-3-4 etc. right through March.

 

Had a feeling that had a lot to do with it, just wasnt sure with the reds reaching to the date line as well. Good stuff as always

[RitualOfTheTrout]

Thanks for bringing this to the forefront because I gave a pretty stern warning about poor modeling within the day 3-10 range (I think beyond 10 speaks for itself, haha). Times of poor NWP have to do with battling regimes, sudden changes in global AAM and stratospheric/solar disturbances.

 

There is no denying the amazing shift in GLAAM with values approaching winters with strong El Nino conditions in place. The MJO signal is clearly getting muted on the RMMs by the h2 battle but it is clear that the MJO kicked its ass and is propagating east (just slower than what the consensus thought 2 weeks ago). The jet has extended and the remnant h2 westerly over the equatorial region got shifted toward the SH / S. America. 

 

There is also no denying a downwelling -u anomaly from the 10-30mb layer to the tropopause over the North Pole. This will reach the troposphere as we start February and it will likely lead to even more blocking than what's being modeled.

 

There is a stronger than normal chance for a significant snowfall Feb 5-10 (yes I know climo says snow is most likely then). The storm being modeled beforehand is certainly possible; but with the PV displacement forecasted then, it is quite possible the "shunt" scenario is the best one. However, with the arrival of the STJ Feb 2-5, we certainly could get rolling in the snow department sooner than I have been saying.

Now that the MJO is moving into the favorable phase 8, does a slower progression mean we will see favorable forcing for a longer period? Or is it better for it to steadily move along? I am assuming of course in either scenario the convection continues and we see a healthy wave.

[Mike.Ventrice]

Ok Harbourton, finally got around to the explanation of the roundy MJO. I know Adam, HM, and Mike may butcher this explanation and I'm probably wrong in some of this but this is what i have tried to grasp so far. The chart is dissected into the western hemisphere and eastern hemisphere. The western hemisphere is not all that important and has very little mjo waves that traverse it. The solid green line running parallel that i have a green arrow pointing to is the current day, (this sometimes doesn't update daily). The black numbers below are the phases, you know which phases are the good ones and the bad ones. The orange arrow pointing to the red dashed line is +OLR (outgoing longwave radiation). +OLR is associated with decreased in thunderstorm activity or decrease in clouds. solid red line which the red arrow is pointing to is the mjo wave ( -OLR) is increase in thunderstorm activity/clouds. The more the solid red circles inside the big circle the stornger the wave. Same goes for the dashed red line the more circles the stronger the subsidence is. Now the blue arrow pointing to the blue shading is an increase in rising air which is aided for thunderstorm development. The gray arrow pointing to the yellow shading is the downward propagation of air, aka subsidence, which leads to weakening of thunderstorms. Just like when you have a mega band in major snowstorm. You notice the mega band, then on the other side of the band is an area that very light in activity do to the sinking air behind it. The darker the blues, the more upward rising of air (more thunderstorm development). The darker the yellows (less/weakening of thunderstorms). Their are a lot of other things going on in this map like the kelvin waves which i haven't learned yet, but I'm sure Adam, HM, and Mike would explain.

 

 

 

You could look at Paul's time-lon plots all day and still not alwasy get the jist of it! This plot takes a lot of time of understanding, but it can be very helpful once you do so!

So this is a time-longitude plot of unfiltered OLR anomalies with MJO filtered, ERW filtered, CCKW filtered, and TD type filtered anomalies in contours (dashed if positive anomalies).  The forecast is based off of a time-extended EOF, where it looks at past cases similar to the present scenario and builds a projection into the future without the addition of zeros in future times and thus is better to space-time filter (this isn't important for the comments you made but it's something that others may find interesting).

 

So in convection (OLR), the MJO has the most variance over the Eastern Hemisphere. You have noted this, as MJO filtered OLR anomalies are not as strong over the Western Hemisphere as they are over the Eastern Hemisphere. However what can get confusing is that the MJO still propagates across the Western Hemisphere, even though MJO-filtered OLR anomalies reduce in magnitude. Think about it this way, in order for convection to get going over an ocean, SSTs need to be warm. So where are SSTs the warmest near the equator? The Indian Ocean and the West Pacific Warm Pools. During El Nino years, it will be over the East Pacific as well. On most occasions over the East Pac and equatorial Atlantic, there are cold SSTs that suppress convection. Thus, MJO-convection will not be as pronounced over these regions  as they are over the Eastern Hemisphere. So if one were to filter convection (OLR) alone like this analysis, it will show weak MJO activity about the equator over the Western Hemisphere. So the phases that you penciled in should be drawn equally distant apart ranging the entire globe. An example (even though this is boreal summer) is given below.

 

 

Something I found to track the MJO easier over the entire globe is its circulation signature. Velocity Potential in low-levels or upper-levels is by far the easiest to track the MJO.  I have a similar time-longtiude plots to Pauls but following a different methodology and uses Velocity Potential instead of OLR and isolates MJOs and CCKWs. 850 hPa and 200 hPa.  Now my anomalies are standardized, so all filtered VP anomalies begin at 1 STD and are in 0.5 STD increments. Further, my time goes down on the y-axis and the entire plot is centered about Africa. You can see that these fields show the MJO-type variability much easier than OLR. You actually don't even need the filtered fields on top to see where the MJO is.  In this type of analysis, the current MJO event has sped up and is centered over Africa. It has a stronger low-level circulation signature than upper-level, where the 850 hPa VP fields suggest a 2 sigma event over Africa, where 200 hPa VP suggests 1 sigma event.

 

Using the filtered VP200 data, I built MJO-phase space diagrams. You can see similar results using those as well:

 

 

Since I use a padding with zero in future times techinque, this analysis will struggle during amplifying and decaying MJO signals. So it definitely has its drawbacks as any MJO-identifier. However, it definately helps with interpreting RMM signals and is a reason why I was skeptical that the MJO would just magically stop as it traversed the Pacific and not make it into the Western Hemisphere. These plots remove all noise outside of the MJO and RMM will often follow it but making erratic loops and jolts here and there.

 

Something very useful that can be observed in Paul's or my time-longitude plots is that it is a means of depicting whether RMM is identifying a CCKW or a true MJO event. If these time-lon plots do not show any MJO signal, yet RMM does, you know that RMM could be just depicting noise. Anyway, there's my rant!

 

Something UNRELATED to MJO/CCKWs (sort of) is the range of model uncertanity for am upcoming extra-tropical wave break over the North Pacific. A d(prog)/dt analysis of dynamic tropopause heights for Monday the 28z at 12Z shows how much uncertanity there in subsequent model runs for the same time. GFS tending to more of an amplifying flow pattern over the North Pacific, suggesting a large cyclonic wave breaking-event on the horizon over the Pacific. This will be causing headaches for medium-range prediction in the upcoming week or two!

[kgriffin]

Mike - any chance you have your MJO diagrams for things like the CFS forecasts? I'm curious to see how an actual 30-day (or even 180-day) model can do vs. Paul's projection techniques. I'm not sure how valuable they would be, but could potentially highlight regions or regimes where either the model or statistical technique outdoes the other. Which would be cool, of course

[NaoPos]

Great stuff in here guys. Its truly been a pleasure to really read this thread this year. Keep it up.

Now, for the important stuff (haha jk fellas)

decent trends on the GFS runs today with next weeks system:

1st up, is the 0z GFS (left side) and the 18z GFS (0z GFS hour 138).

2nd up is the 0z gfs (left and hour 138 ) compared to the 12z GFS.

While its trending stronger and digging more with the northern stream, gotta be careful. its how we got burnt with tomorrows system. At one point, models were showing that shortwave dropping in MT, not the Upper Peninsula of Mich.

Just thought i'd point it out.

[Rainshadow]

You could look at Paul's time-lon plots all day and still not alwasy get the jist of it! This plot takes a lot of time of understanding, but it can be very helpful once you do so!

So this is a time-longitude plot of unfiltered OLR anomalies with MJO filtered, ERW filtered, CCKW filtered, and TD type filtered anomalies in contours (dashed if positive anomalies).  The forecast is based off of a time-extended EOF, where it looks at past cases similar to the present scenario and builds a projection into the future without the addition of zeros in future times and thus is better to space-time filter (this isn't important for the comments you made but it's something that others may find interesting).

 

So in convection (OLR), the MJO has the most variance over the Eastern Hemisphere. You have noted this, as MJO filtered OLR anomalies are not as strong over the Western Hemisphere as they are over the Eastern Hemisphere. However what can get confusing is that the MJO still propagates across the Western Hemisphere, even though MJO-filtered OLR anomalies reduce in magnitude. Think about it this way, in order for convection to get going over an ocean, SSTs need to be warm. So where are SSTs the warmest near the equator? The Indian Ocean and the West Pacific Warm Pools. During El Nino years, it will be over the East Pacific as well. On most occasions over the East Pac and equatorial Atlantic, there are cold SSTs that suppress convection. Thus, MJO-convection will not be as pronounced over these regions  as they are over the Eastern Hemisphere. So if one were to filter convection (OLR) alone like this analysis, it will show weak MJO activity about the equator over the Western Hemisphere. So the phases that you penciled in should be drawn equally distant apart ranging the entire globe. An example (even though this is boreal summer) is given below.

 

 

Something I found to track the MJO easier over the entire globe is its circulation signature. Velocity Potential in low-levels or upper-levels is by far the easiest to track the MJO.  I have a similar time-longtiude plots to Pauls but following a different methodology and uses Velocity Potential instead of OLR and isolates MJOs and CCKWs. 850 hPa and 200 hPa.  Now my anomalies are standardized, so all filtered VP anomalies begin at 1 STD and are in 0.5 STD increments. Further, my time goes down on the y-axis and the entire plot is centered about Africa. You can see that these fields show the MJO-type variability much easier than OLR. You actually don't even need the filtered fields on top to see where the MJO is.  In this type of analysis, the current MJO event has sped up and is centered over Africa. It has a stronger low-level circulation signature than upper-level, where the 850 hPa VP fields suggest a 2 sigma event over Africa, where 200 hPa VP suggests 1 sigma event.

 

Using the filtered VP200 data, I built MJO-phase space diagrams. You can see similar results using those as well:

 

 

Since I use a padding with zero in future times techinque, this analysis will struggle during amplifying and decaying MJO signals. So it definitely has its drawbacks as any MJO-identifier. However, it definately helps with interpreting RMM signals and is a reason why I was skeptical that the MJO would just magically stop as it traversed the Pacific and not make it into the Western Hemisphere. These plots remove all noise outside of the MJO and RMM will often follow it but making erratic loops and jolts here and there.

 

Something very useful that can be observed in Paul's or my time-longitude plots is that it is a means of depicting whether RMM is identifying a CCKW or a true MJO event. If these time-lon plots do not show any MJO signal, yet RMM does, you know that RMM could be just depicting noise. Anyway, there's my rant!

 

Something UNRELATED to MJO/CCKWs (sort of) is the range of model uncertanity for am upcoming extra-tropical wave break over the North Pacific. A d(prog)/dt analysis of dynamic tropopause heights for Monday the 28z at 12Z shows how much uncertanity there in subsequent model runs for the same time. GFS tending to more of an amplifying flow pattern over the North Pacific, suggesting a large cyclonic wave breaking-event on the horizon over the Pacific. This will be causing headaches for medium-range prediction in the upcoming week or two!

 

Mike,

 

Thank-you for the great explanation. It also explains why sometimes in the tropical season you can have 3 or 4 named storms at once and then two weeks later cant buy a thunderstorm. Just looking at the model runs overnight and how little consistency there was, while we are entering a window of opportunity phase of this winter, the run to run continuity and skill into the medium range sure looks like its going to suffer.

[Mike.Ventrice]

Mike - any chance you have your MJO diagrams for things like the CFS forecasts? I'm curious to see how an actual 30-day (or even 180-day) model can do vs. Paul's projection techniques. I'm not sure how valuable they would be, but could potentially highlight regions or regimes where either the model or statistical technique outdoes the other. Which would be cool, of course

Kyle! Actually, Kyle MacRichitie has CFS v2 MJO phase space diagrams following Paul's time extended EOF analysis. The BAD thing about the CFS is it is always too slow with propagating the MJO eastward. Either there's an error in the code, or the model has a slow bias. I presume it's the latter.

 

[Mike.Ventrice]

Things aligning for February? At least climate is on our side.  I'm curious as to how long it takes the MJO to force the pattern? In other words, once it enters Phase 8, how long does it take for cold to enter North America? Or perhaps I'm confused as to how that works haha.

 

This isn't something that is really understood. I'm not 100% sure, but I believe when the MJO is active over the Western Pacific, it builds ridges in both Hemisphere via extra-tropical Rossby wave dispersion. From there, downstream development occurs, where in the northern hemisphere, a trough anchors over the Western US and Ridge over the Eastern US (WARMTH in Phases 5+6!) This sets the stage for a low-frequency type Rossby wave train. So as the active MJO phase moves into RMM phase 8, 1.. convection is usually enhanced over the n. East Pac and rest of Western Hemisphere. The ridge that was over the N. Pac slides eastward, and the trough that was over the western US slides eastward over the East US, and the ridge that was over the East US moves out over the western Atlantic.  But you have to consider short waves will facilitate the exact timing of cold or warm. The MJO just primes the large-scale flow regime over the US to promote warmth or cold. It doesn't necessarily always line up right though! Understanding times where it is will cooperate or not cooperate is the key

[am19psu]

Awesome, awesome work in here last night guys. Thanks for contributing to this thread.

 

Tombo, of course Mike is right in terms of the science of the MJO, but I actually think your approach works for the layman who is just trying to see things at a high level.

[am19psu]

The MJO just primes the large-scale flow regime over the US to promote warmth or cold. It doesn't necessarily always line up right though! Understanding times where it is will cooperate or not cooperate is the key

This is really a key statement here and why I keep saying HM is better at this stuff than me. I can pick out favorable periods based on making skillful MJO forecasts, HM has other things he looks at that lets him zero in on cyclogenesis.

[am19psu]

There is a stronger than normal chance for a significant snowfall Feb 5-10 (yes I know climo says snow is most likely then). The storm being modeled beforehand is certainly possible; but with the PV displacement forecasted then, it is quite possible the "shunt" scenario is the best one. However, with the arrival of the STJ Feb 2-5, we certainly could get rolling in the snow department sooner than I have been saying.

Oh, and: #climo

[tombo82685]

This isn't something that is really understood. I'm not 100% sure, but I believe when the MJO is active over the Western Pacific, it builds ridges in both Hemisphere via extra-tropical Rossby wave dispersion. From there, downstream development occurs, where in the northern hemisphere, a trough anchors over the Western US and Ridge over the Eastern US (WARMTH in Phases 5+6!) This sets the stage for a low-frequency type Rossby wave train. So as the active MJO phase moves into RMM phase 8, 1.. convection is usually enhanced over the n. East Pac and rest of Western Hemisphere. The ridge that was over the N. Pac slides eastward, and the trough that was over the western US slides eastward over the East US, and the ridge that was over the East US moves out over the western Atlantic.  But you have to consider short waves will facilitate the exact timing of cold or warm. The MJO just primes the large-scale flow regime over the US to promote warmth or cold. It doesn't necessarily always line up right though! Understanding times where it is will cooperate or not cooperate is the key

Mike or adam, what is the RMM? Is their an example this year of noise that mike alluded where the kelvin wave was mistaken for the mjo?

[am19psu]

Mike or adam, what is the RMM? Is their an example this year of noise that mike alluded where the kelvin wave was mistaken for the mjo?

RMM = Real-time Multivariate MJO = the first and second non-ENSO principal components of the spatial variance of the OLR/200mb u-wind/850 u-wind fields

[tombo82685]

RMM = Real-time Multivariate MJO = the first and second non-ENSO principal components of the spatial variance of the OLR/200mb u-wind/850 u-wind fields

which the better (relatively speaking) RMM or Roundy? I would imagine 2 charts would be modeled with the 200mb and 850mb since storms can be low level like mike stated?