March 15-19 Potentially Significant Storm System

I'm eyeing the March 15-19 period for what could be a significant storm system.

The image above shows ECMWF 500mb height anomalies from a previous forecast, valid on March 9th, which was two days ago. While this forecast is not current, and the forecast image above is not current either, we did see a big storm move through Japan on March 9th. There is a rule, well explained by Joe Renken, that states a weather phenomenon in East Asia will be reciprocated in the United States 6-10 days later. This means that if there is a storm system in Japan on a certain day, we can expect a storm in the US 6-10 days after that. The same goes for high pressure and warm weather. Using the image above, we can expect a storm in the US 6-10 days after March 9th, which would bring us to a March 15-19 storm. Not only that, but this forecast indicates the storm in Japan will be/was strong when it passed through, and this means that the storm we experience March 15-19 could very well be strong, too.

Shown above is the latest ECMWF ensemble mean forecast for 500mb height anomalies on the evening of March 15th. Interestingly enough, we see the ECMWF ensembles already have this storm in the Plains, meaning the idea of a big winter storm for this timeframe is becoming popular. The ensembles have a positive PNA pattern in place, meaning we see high pressure shooting north along the West Coast, with consequential storminess in the Plains. We also see ever-so-slight ridging in the Southeast, but this is suppressed by the piece of the polar vortex displaced over southern Canada. It looks like the ensembles want to develop a split jet stream pattern, which could open the door for phasing of storm systems. Despite this potential, I feel that any storm in this timeframe would either move east and possibly up the coast, though I'm a bit hesitant about that potential due to the polar vortex pushing down on the jet stream as well as the storminess over Greenland (which typically does not favor East Coast snowstorms). The other idea would be the storm moving north and hitting the Midwest/Great Lakes, but I'm also skeptical on this idea due to the polar vortex piece trying to suppress the storm (which could very well happen), as well as the lack of high pressure in the Southeast to force the storm north. We'll get a much better handle on this in coming days, but be warned that the March 15-19 period could see a potentially significant storm hit the US.

It is also worth noting that we could see some severe weather, though to which degree is to be determined.

Andrew

Impending Snowstorm to Drop 12"+ of Snow Tuesday and Wednesday

Note: This post will concern the March 11-14 Significant Snowstorm, but the title has been changed. If you find this sort of title better than the date and storm classification, or if you prefer the date + classification, leave your input in the comments below.

I'm carefully watching the next few days for a snowstorm that looks to drop a foot of snow or more on the United States.

We'll begin with an analyzation of the big picture of this storm. The model I will be using here will be the NAM model, as its low-resolution GFS counterpart appears to be suffering from some errors we will address later on in this post. The image above shows mean sea level pressure (MSLP) values, as well as precipitation values and high and low-pressure demarcations for Tuesday afternoon, tomorrow. We see a strong storm system in the far northeast corner of Kansas, bottoming out at 994 millibars and creating a heck of a pressure gradient over Nebraska. In Nebraska, we also see a precipitation shield breaking out, which could be accompanied by some harsh winds if that pressure gradient happens to meet up with the precipitation. Luckily, this forecast keeps the strongest winds just south and east of the inclement weather in northwest Nebraska.

In the late evening hours of Tuesday, we see that our storm system has pushed off to the east, maintaining both its tight pressure gradient and precipitation shield as it progresses east. Our storm is now at about 995 millibars, roughly the same strength as what we saw Tuesday afternoon in the image above. The 1 millibar difference really isn't anything to write home about in this situation. Typically, we might see a severe weather situation arising in the Plains and Midwest due to that tight pressure gradient, but that swath of precipitation down in Mississippi, Alabama and the panhandle of Florida indicate there's a storm system there (a closed low, to be exact) that will prohibit such a severe weather event from forming. At this point in time, accumulating snow should begin to fall over the Midwest.

By the time we hit the morning of Wednesday, March 12, we find that phasing has occurred with our storm system. What this means is that our original system has phased, or merged, with a system coming south from Canada to create a single, stronger storm. This is clearly displayed by the abundant precipitation breaking out over Michigan, Indiana and Ohio, and also shown by the tighter pressure gradient expanded across much of the Central and East US. It is also possible the western flank of the gradient is a bit amplified thanks to the strong high pressure in Canada, but that's not a big piece of the puzzle right now. At this point in time, potentially significant accumulating snow looks to be impacting the Ohio Valley.

Lastly, by the evening hours of Wednesday, we see the system has now transferred offshore, and is now putting down significant accumulating snow across the interior Northeast, leaving coastal areas (and even some inland regions) to experience rain or a mixing event.  The extreme pressure gradient remains in place, maximized in this forecast graphic over Pennsylvania and New York, as well as the rest of the New England area. After this, the storm exits the region, leaving quite a snowstorm in its wake. Check out the latest snowfall forecasts for this event:

To add some model variety, we'll also show the GFS snowfall forecast.

If you look at these maps and think something's not the same, you're right. There are some differences here. The primary difference is that the GFS is more south with the storm system than the NAM model. This is displayed well when looking at the beginning of the accumulating snow swath all the way back out in Illinois. The NAM model gives a generous 6" or more to Chicagoland, but the GFS only lays down a handful of inches across Springfield, IL.

After looking at all the data, I think it might be best to lean towards a NAM solution with the placement of heavy snow further north in the Great Lakes, but not too far north, if you know what I mean. Let's go over why.

The first red flag is the NCEP Weather Prediction Center's preferred track superimposed on an ensemble suite for this storm system. We see the NCEP WPC track for this storm in black, but it's obvious that their track is on the southern envelope of guidance. The heavy majority of ensemble members prefer to take this storm further north, into central or northern Missouri rather than southwestern MO. This would result in snow totals displaced further to the north, like the NAM says. But we can dig deeper than this. Let's keep going in our search for why these northern amounts are more preferred over the GFS.

Let's take a close look at the image above. Here, we see the NAM model projecting 300mb wind speeds in color shades and wind barbs, valid early Wednesday morning. If you look closely, you can spot two distinct jet streaks, or two areas of stronger winds than the winds around them. These are shown best by the yellows in Oklahoma and Kansas, as well as in far northern New England. These two jet streams form the classic coupled jet stream, as defined by Uccellini and Kocin in their studies. This coupled jet stream idealizes that there is a space between two jet streaks, where heightened divergence is found. In other words, in the space between two jet streaks, the atmosphere wants to create more precipitation. So, if we look above for the space between these two jet streaks, we can draw a line roughly from Missouri to Lake Michigan of where this increased divergence, or increased precipitation would be expected to appear. This would seem to put the NAM's snowfall placement 'in the right' when compared to placement of the GFS snow totals, at least in the Great Lakes region.

Also, take a look at what the 18z GFS does with the storm as it moves along the Ohio Valley.


If you watch the low pressure denotation as it moves from Indiana, it goes to the southeast rather than to the east. You might not think this is so odd, but when you compare it to the agreed notion that this storm will be moving west-to-east, a southward movement is rather odd.

Additionally, a convective feedback issue was observed as the system crossed east through Indiana, likely hampering the forecast for this storm. In this case, you may want to defer to the 12z GFS, as it is more logical in comparison to this run.

To conclude, a significant winter storm is expected for a good chunk of the US. Model guidance differs on placement of the snow, mainly in the Great Lakes, but it is agreed on that the Northeast will see a substantial snowstorm.

Andrew

March 11-14 Significant Snowstorm

I'm still looking at the idea of a significant snowstorm on the March 11-14 period.

An upper air analysis at the 500mb level of the atmosphere shows our storm system as the depression in the contour lines over the northeast Pacific, which tells us the system isn't onshore yet. The storm not being onshore means that model guidance will continue to change until the storm actually gets onshore. The reason being that National Weather Service offices release weather balloons twice a day, and the data those balloons gets goes into the models to enhance their forecasts. Thus, when the balloons are sent into the storm when it comes onshore, the models get a better handle on the storm, and that's why we tend to see a model consensus come about when the storm system in question comes onshore. We also see suppressed ridging over the Western US, which will impact the eventual track of this event, as we'll go over later.

Shown here is the GFS 500mb vorticity projection for the evening of March 11th. On this image, it seems our storm system is separated into two storms, located over Nevada and Kansas. In this case, the GFS model wants to take the westernmost portion of the storm and retrograde it into the ridge located along the West Coast. This piece of energy then closes off and separates from the other piece of vorticity that continues progressing eastward. This move is a bit suspect to me. It is well known that model guidance has the tendency to hold energy in the Southwest for too long, and this could very well be one of those situations where the GFS is too eager to bring a piece of the storm into the Southwest US.

By the time we get to the evening of March 12th, about 24 hours later from the image we just discussed above, we see a lot has changed. We now see that the storm has indeed separated into a closed system along southern Nevada and California, which is linked to its parent storm by an elongated lobe of positive vorticity draped across the Plains. We'll get to that parent storm in a second, but first let's go over the storm in the Southwest. The system has retrograded directly into the West Coast ridge, and has closed off in the process, leading to a Rex Block-style pattern, where we have a ridge directly north of a storm system. While the typical Rex Block produces a zonal flow (west-to-east flow) synoptic set-up across the areas downstream of the block (to the east of the block), the GFS prefers to initiate a northwest flow regime, with the West US ridging leading to deep troughing across the East US. This leads us into the parent storm, which is in its own category here. The aforementioned northwest flow has led to a phasing (merge) of the easternmost storm we saw in the second image of this post (the first GFS image we discussed) with a weak system originating from northern Canada. This comes as no surprise, with the northwest flow regime being very supportive of phasing should the opportunity arise. I'm a little skeptical on this idea, as models are notorious for phasing storm systems too eagerly. This could mean we actually see two weak, unphased systems when the March 11-14 timeframe comes around, but with guidance supporting this phasing more and more, I find that the positives for phasing outweigh the things going against it.

This is the ECMWF 500mb vorticity forecast for the evening of March 11, the same timeframe as the first GFS image we analyzed. In this model's forecast we see the two systems are still trying to split up, but are nowhere near as elongated as the GFS model portrays them to be. Rather, we see the two systems fairly well defined, with one over Utah and the other over Kansas.

The above graphic shows the ECMWF 500mb vorticity forecast for the morning of March 13th, about 6 hours after the second GFS image we analyzed. Here, we see a significantly differing view as to what happens. The ECMWF takes the westernmost system and does retrograde it into the Southwest, but it does so as an incredibly weak system, so the system does not become a closed low and the Rex Block cannot form. We see a strong lobe of positive vorticity extending across the southern Plains and Gulf Coast, before we arrive on the parent storm, which has phased with the Canadian storm. This solution also results in a snowy solution for much of the Northeast, and it is a viable idea. However, because this is a northwest flow regime, and the ECMWF model has performed poorly in northwest flow situations this winter, I'm not ready to buy in on this solution just yet. It will probably take another day or so before we can at least refine the solutions we have now to try and come across a more solid consensus.

Here's an overview of current model projections for snowfall.


Andrew

March 11-14 Potentially Significant Snowstorm

As I discussed back on February 25th, we're looking at the potential for a significant winter storm in the March 11-14 timeframe, narrowed down from the original March 12-17 timeframe.

The ECMWF model's 500mb vorticity forecast for March 12th is shown above, and we see the two ingredients for our storm on this map. We see the main piece of energy as an elongated swath of positive vorticity stretching from Nevada to Missouri, with our second player dropping south on the lee side (east side) of the ridge stretching across the West Coast. We also see a piece of energy in the Gulf of Mexico which will do its part to interrupt the severe weather aspect of this storm, which is why we'll be focusing on the snowy side.

The ECMWF develops a 997 millibar low centered over southern Illinois on the morning of March 12th, with overpowering high pressure to the north suggesting we aren't likely to see any big northward shifts with this storm in the future. Model verification confirms this idea. As far as the projected storm, this 997mb low is just the elongated system, not accounting for the system dropping from Canada which will eventually phase with the aforementioned system in Missouri.

By the morning of March 13, the system is rapidly deepening in the Mid-Atlantic as the two pieces of energy have begun to phase. This means that they are combining into a single storm system, which only intensifies the snow potential for this event. From March 12th's jet stream forecast, I would expect we see the system try to push east due to a rather zonally-oriented jet stream across the United States, which isn't that favorable for the system to shoot up the coast. However, by the time it hits the coast, the jet stream is essentially laying out the red carpet for the system to go northeast and affect the Northeast.

The snow map for the ECMWF would dump amounts of upwards of 12" across much of the Northeast, including coastal regions, while laying down over 6" in Ohio and a portion of Indiana. This would all depend on phasing of these two systems, which the folks at the National Centers for Environmental Prediction (NCEP) indicate may not be so likely. They indicate that models tend to phase systems too often, when in reality, they don't end up phased. This could be one of those situations, but I guess we'll just have to wait and see.

The GFS model, on the other hand, is much less enthusiastic when compared to its European counterpart. We see both systems shown above, valid for the same timeframe as the 500mb vorticity map we looked over at the top of this post, but in the GFS' case, the layout is different. The GFS model keeps the system back in the Southwest, on a positively-tilted axis, meaning the strongest vorticity values are pointing in a southwest direction. I have a feeling the GFS might not be grasping this system correctly. The NCEP agency also states that models in general have a bias to keep storms in the Southwestern US for too long, when in reality they eject from the Southwest quicker than forecasted. This would work in favor of the ECMWF's solution. However, the GFS does not phase the two systems, which results in noticeably less snow and precipitation in general. We'll need to watch both models for this system in the next few days, as both appear to be hanging on to one bias while letting go of another one. In this case, the ECMWF may be phasing the systems too eagerly but not holding the energy back in the Southwest, while the GFS looks to be keeping the system in the Southwest but not phasing the storms.

Andrew

Major Severe Weather Outbreak Expected in Mid-March

I'm watching out for a potentially significant severe weather outbreak that would occur in the middle of March. Please note that due to the long range nature of this post, it is possible that the event does not happen. What I show below gives you my current thoughts on the situation that are very much subject to potentially drastic change.

(The following is copied from the March 12-16 Potential Storm post) Shown above are ECMWF model projected 500mb height anomalies over the North Pacific, valid on March 6th. We want to take a look at Japan in this image, which is on the left-hand side of this graphic. If we look at Japan, we see deep purples over the island nation, indicating the presence of anomalously low heights. This tells us there is a storm system afoot; the purples indicate the storm is of the stronger variety when compared to other systems. How can weather in Japan relate to weather here in the US? If we use the 6-10 day correlation, explained by Joe Renken, which states that a storm system in Japan can affect the United States 6-10 days later, we find that this storm system would be expected to create a storm in the United States during a March 12-16 time period. 

There is something I'm watching in this time period and on model forecasts. If you were to animate the maps over the North Pacific, you would find that it almost seems like two separate systems impacting Japan over the March 6th time period, but the proximity to one another makes it seem like one big storm. It might take a little while to determine if this really is one strong system or two moderate systems, but until then, it's better to err on the side of caution and discuss how a single, strong storm system could impact us.

We can actually figure out the projected track for this storm as well, by using a tool titled the Lezak Recurring Cycle. The Lezak Recurring Cycle, or LRC, is a tool developed by meteorologist Gary Lezak that, in essence, can enable forecasters to predict the overall weather pattern months in advance. The gist of the LRC involves a cycling weather pattern that develops in October and November of each year; no pattern is the same from year to year. Around mid November, the LRC begins to repeat, meaning we start to see a similar weather pattern in mid November that we saw in early October. This means that the cycling pattern has begun, and it will continue to cycle on a regular, unchanging 40-60 day interval for the next ~10 months before it dissipates over the following summer. Since this season's cycle has been holding at around 57 days, we can go back from the March 12-17 period and arrive around the  January 16-20 time period. The image above shows observed 500mb height anomalies from the January 16-20 timeframe, where North America is located on the upper right portion of the image. If we look at what the atmosphere was like in that mid-January period, we find a defined northwest flow scenario, with strong ridging over the West Coast leading to deep negative height anomalies across the East Coast. The jet stream below for the same January 16-20 period reflects this northwest flow.

(End copied portion) Take a look at the jet stream pattern above. For this period (January 16-20) and the period prior to this (roughly January 11-15), we saw a phased jet stream over the United States. The phased streams are more pronounced in this image, as we see the subtropical jet stream and Pacific jet feeding into a band of stronger winds located over the Gulf of Mexico. Using the LRC, we can predict that not only will this sort of weather pattern come back, but the jet stream pattern should return as well. Additionally, with the jet stream naturally lifting north as we enter the spring season (though it sure doesn't feel like it), the severe weather risks begin to shift more on-land as the stronger jet streams gather more moisture from the Gulf of Mexico. We're already seeing the Pacific jet stream pretty amped up as of today, which looks to play into this potential mid-March severe weather outbreak.

Observed jet stream

Based on the LRC and the rather ominous jet stream pattern observed in Mid-January, my concern is raised for a severe weather event in the middle of March, particularly around March 13-17.

Paul Roundy

Here's where we get into the really concerning stuff. Shown above is a chart of six different types of outgoing long wave radiation (OLR) anomalies. We want to focus on the middle panel of the left-hand side, where it says MJO OLR anomaly on February 16, 2014. On February 16, 2014, we saw negative OLR anomalies just south and east of the subcontinent of India. This location of the -OLR anomalies (which indicate enhanced tropical convection) tells us the Madden-Julian Oscillation was in Phase 2, out of its 8 possible phases. Check out the severe weather event that occurred just four days later, on February 20, 2014.

On February 20, 2014, we saw a large-scale severe weather outbreak, as the chart of storm reports from the Storm Prediction Center shows above. The outbreak was primarily a damaging wind event, as you can see by the large swath of blues across Tennessee and Kentucky, just to name a few of the many affected states. However, we also saw a handful of tornado reports. The fact of the matter is, this severe weather event and the MJO entering Phase 2 are not coincidence. They are correlated with one another. See the AMS (American Meteorological Society) article below:

The Madden–Julian oscillation (MJO) has been linked to weather variability in the midlatitudes via its associated overturning circulations and Rossby wave trains that redistribute the thermal and mass fields at higher latitudes. This work examines the relationship between the MJO and violent tornado outbreaks in the United States. A census of events shows that violent tornado outbreaks during March–April–May (MAM) are more than twice as frequent during phase 2 of the Real-time Multivariate MJO (RMM) index as during other phases or when the MJO was deemed inactive.

If you didn't get what this was saying, don't worry. Basically, it tells us that when the MJO is in Phase 2 during March, April and May, or when we see enhanced convection south and southeast of India, we tend to see the chances of a violent tornado outbreak skyrocket in comparison to the nonactive MJO, or the MJO in other phases. While the February 20th event wasn't exactly a tornado outbreak, it was a severe weather event, and this has me concerned for mid-March.

Let's take a look at the forecasted OLR anomalies for March 12, 2014. Once again, we'll look towards the panel labeled MJO OLR Anomaly. If we look at the forecast by Paul Roundy above, we can see very deep negative OLR anomalies placed just south and southeast of India. These blues are far deeper than what we saw on February 16, meaning that this Phase 2 MJO event could be even stronger, and thus the severe weather event would be even stronger as well. Going by the LRC we discussed above, as well as the AMS document analyzed here, I am concerned that we will see a potentially significant severe weather outbreak for the mid-March period. Because there could be a lag period between the MJO Phase 2 event and the severe weather event of about 4 days, like we saw with the February event, I'll make a tentative timeframe of March 13-18 for a potential severe weather outbreak. This one has me worried, and while you should not be worried yet, it's something to watch out for in the future.

Andrew