Poor Internet connectivity continues, so once again another short — yet informative — post.
The Storm Prediction Center (SPC) is forecasting a moderate risk of severe thunderstorms across a portion of Minnesota tomorrow. People across the northern plains should review their severe weather preparedness plans and be prepared to enact them tomorrow.
First, I just wanted to say that I’ve been having Internet connectivity problems of late which has kept me from doing indepth posts. I hope to have this fixed by early next week. *crosses fingers*
Second, the heat that was promised has continued to develop north and east into parts of Illinois, Indiana, western Kentucky, western Tennessee, and northern Mississippi. In fact, much of Oklahoma and south-central Kansas are under Excessive Heat Warnings. Just to the west of a large area of heat advisories, severe thunderstorm watches have been posted from southwest Kansas northeast to the United States-Canada border north of Wisconsin! Thunderstorms have been trying to develop along the leading edge of a cold front that will hopefully bring slightly cooler, and drier, conditions to portions of the area experiencing this heat wave.
Author’s Note: It was only after writing tonight’s post that I learned about Tropical Depression #2. I’m sure I’ll be blogging about that tomorrow.
As I was preparing to head home this afternoon, I received an email from the Storm Prediction Center’s (SPC) Warning Coordination Meteorologist (WCM), Greg Carbin. He had created a cursory overview of June 2010′s severe thunderstorm and tornado warnings. In my conversations with Greg this week, he informed me that Monday (5 July 2010) was the first day the SPC had not issued a watch since 15 May 2010. That means the entire month of June had at least 1 severe thunderstorm or tornado watch issued per day. That’s very impressive!
All severe thunderstorm warnings issued during the month of June are plotted above. The darker the shading of blue, the more times a given area was under a severe thunderstorm warning. It is quite apparent that the central plains east into Ohio had a very active month, with numerous severe thunderstorm warnings issued for most areas. There are several states in this swath that were almost entirely covered by severe thunderstorm warnings.
All tornado warnings issued during June are plotted below. Once again, the darker shading indicates the more times a given area was under a tornado warning. The same area from central plains east toward Ohio stands out once again. However, in this plot the northern Minnesota and eastern North Dakota areas also stand out. Most of these warnings were issued on 17 June 2010, when a very large tornado outbreak unfolded across the northern United States.
As a plug, I should point out that these images were created using the Python programming language – a favorite of mine! If there is the slightest of chances that you might need to learn to program at some point in your career (note: this includes ALL METEOROLOGY STUDENTS, undergraduate and graduate alike) this is one language that you need to have under your belt. It’s very easy to learn and very powerful!
In storm chasing circles “The Day” is known as the day in which all the numerical models forecast a severe weather outbreak. Often “The Day” has been forecast several days in advance and storm chaser excitement builds as the the event draws closer. ”The Day Before” is also known in storm chasing circles. This is the day before “The Day” and is often characterized by high risk, high reward for viewing severe thunderstorms and tornadoes. In other words, you aren’t guaranteed to see something (high risk), but if a thunderstorm can develop, there is a chance you could be the only “chaser” there to observe it (high reward). Tonight, I’m going to highlight a potential “Day Before the Day Before” situation.
As I’ve alluded to in recent posts, numerical models are hinting at the possibility of severe thunderstorms on Friday across portions of the central United States. This would be “The Day”, making Thursday the “Day Before”. However, there is an extremely high risk, high reward situation setting up on Wednesday.
Above (annotated below) is a forecast from this evening’s 00 UTC (7 PM CDT) run North American Model (NAM) valid 18 UTC (1 PM CDT) on Wednesday. It shows accumulating precipitation falling across portions of northeast Oklahoma. This precipitation is forecast to develop late Tuesday in northwest Kansas and northeast Colorado, in the higher elevations. (I would just like to point out that this is typical of late summer patterns, not late April!!) This complex of showers and thunderstorms is then forecast to track southeast overnight and dissipate during the afternoon over western Arkansas.
This is an important feature because the precipitation and associated cloud cover act to keep the temperatures cooler in the wake of the precipitation than the surrounding areas. At the same time, warm, moist air is forecast to began moving northward in response to the developing low over the southwest United States. As the warm, moist air advances far enough northward it will impinge upon the rain-cooled airmass forecast to be in place across northern Oklahoma and southern Kansas. This increasing gradient in temperature is, by definition, a front. In this case it is a warm, front. Warm fronts are important in severe thunderstorm and tornado forecasting because they act as a source of low-level convergence which can help aid thunderstorm development. Also, just north of the warm front winds at the surface tend to be out of the southeast or east and quickly switch to being from the southwest aloft. This rapid change in wind direction with increasing height is a crucial component for thunderstorms to be able to sustain themselves. This allows them to become stronger than than otherwise might.
The previous images (first four) were all valid at 18 UTC (1 PM CDT) on Wednesday. This last image is valid at 00 UTC Thursday (7 PM CDT Wednesday). It displays the forecast Convective Available Potential Energy (CAPE) late Wednesday. This is a measure of how much energy a thunderstorm would have to work with should one develop. Forecast values in the image above in far southwest Oklahoma and northwest Texas are more than sufficient for large updrafts to develop – a key ingredient of severe thunderstorms this time of year. Couple this large amount of instability with a potential warm front in the area, Wednesday afternoon just might turn out to be “The Day Before The Day Before”.
Lastly, I want to say that this signal has been in the last several model forecasts of both the NAM and the Global Forecast System (GFS) model. By no means is anything a guarantee, but it is certainly something to watch. Thursday and Friday, almost certainly have a greater potential to produce severe thunderstorms. Thus, don’t let your guard down if nothing happens on Wednesday. There is a reason why it is known as a high risk, high reward kind of day!
Tonight I don’t have much time, so the post won’t be as long or as detailed as it has been the last few nights.
Last night’s post mentioned convergence along a dryline as being the most likely location for thunderstorms to develop this evening. Well, that’s exactly what happened. In the image above, notice how the thunderstorms are located in a thin, narrow line. This squall line, as it is called, is developing along along a convergence zone very similar to what the model’s in last night’s post predicted. Note, even though last night’s models did not predict thunderstorms, they did predict convergence. This is where a skilled meteorologist can make improvements to a model forecast by having an up-to-date understanding of the situation.
Update: Please read the first comment from the Storm Prediction Center’s (SPC) Warning Coordination Meteorologist, Greg Carbin. He points out that I was making extreme oversimplifications regarding the evolution last night, and is absolutely correct. My response to his comment is below.
Warning: This post does contain somewhat technical information.
In preparing tonight’s blog post, which was originally going to be a forecast discussion regarding late week severe weather potential, I came across something that I thought would end up making a better blog post. Especially when you consider that the severe weather prospects for Friday didn’t look good based on this morning’s 12 UTC model forecasts.
From time to time, meteorologists have to play detective. What I mean is that numerical model forecasts often do things that at first glance don’t seem all that extraordinary. However, thorough meteorologists will investigate why a model did what it did, and relying on theory, attempt to glean useful information that isn’t exactly obvious. A good example of this can be found in examining the 12 UTC Global Forecast System (GFS) forecast for 12 UTC Thursday (7 AM CDT Thursday) through 00 UTC Friday (7 PM CDT Thursday).
The image above is valid 12 UTC Thursday (7 AM CDT), and displays the forecast 500 milibar heights and vorticity (top half) and 500 milibar heights and wind speeds (bottom half).
Through the course of these images, pay attention to the area of southern Colorado and northern New Mexico. Watch for changes in wind speed and vorticity (both color fills).
The image above is valid 18 UTC Thursday (1 PM CDT).
The image above is valid 00 UTC Friday (7 PM CDT Thursday)
Did you see it?
Notice the model’s forecast wind speeds for northern New Mexico increase dramatically between 12 UTC and 18 UTC and then decrease dramatically between 18 UTC and 00 UTC. In fact, the maximum seen in wind speeds at 18 UTC are greater than the surrounding magnitudes at 12 UTC and a lot greater than the surrounding magnitudes at 00 UTC. Did you also notice the same thing happen with the vorticity? Why might this be?
First, let me explain why the two images are linked together in a single frame. Typically, meteorologists look for kinks in the 500 milibar height field to identify potential shortwave troughs and/or ridges. In fact, the dip in the 500 milibar heights throughout the western United States are associated with a (larger) trough. Troughs are associated with increases in (relative) vorticity and ridges are associated with decreases in (relative) vorticity. Because increases in (relative) vorticity are associated with troughs, whenever the colors on the top-half of the image increase, one could infer the presence of a trough. Without going through all the explanation as to why, an increase in wind speed can also be attributed to the presence of a trough, espcially downstream (in this case on the east side) of a trough. (It has to do with strengthening lower-level temperature gradient(s)).
In the vorticity images, one can track an area of higher vorticity values from southern Utah (image 1) to northwest Colorado (image 2) to southeast Wyoming (image 3). This is associated with a shortwave trough breaking off from the southern portion of the trough and rapidly ejecting northeast. This is known as a lead-shortwave trough as it “leads” the main shortwave trough. As the lead shortwave trough races through the flow of the longer wavelength trough, it helps to increase 500 milibar wind speeds in northern New Mexico.
If we think about the main shortwave trough, we would expect to find stronger winds located downstream (east) of the trough axis. The same can be said of the lead-shortwave trough. We’d expect to find an increase in wind speed downstream from the trough axis. When the expected area of increased wind speed from both troughs overlap, we get a cumulative effect where the winds increase to more than what they would without interaction of the two troughs.
If a meteorologist had simply examined the noisy vorticity plot above, the presence of the lead shortwave could have been missed. The same if a forecaster had only examined the forecast wind speeds. However, by examining the two together, and understanding that it is possible to indirectly link increases in vorticity to increases in wind speed, we can determine the presence of a lead-shortwave trough. Why is this important? Downstream from a shortwave trough, rising motion is often found. When a “cap” (discussed more in upcoming blog posts) is present, the rising motion associated with a shortwave trough will often help thunderstorms initiate, when otherwise thunderstorms wouldn’t. This often results in what chasers refer to as “the day before the day” events. That is, thunderstorms and tornadoes developed on the day before the main trough moves through, which most people consider to be “the day”.
Now, I’m not saying that Thursday will be “the day before the day”, however, if moisture return is a little better than forecast, and a lead shortwave trough moves through during the afternoon hours, I wouldn’t be surprised if an isolated severe thunderstorm or two developed out in the Texas panhandle. I also wouldn’t be surprised if nothing happened, either.
It has been a very hectic day, and the numerical weather prediction models for this weekend have helped make it even more hectic than it would have been otherwise. Last night, I mentioned how almost all the numerical guidance (at least those available to me when I blogged) indicated that the upcoming southern plains storm would be relatively minor, with most of the precipitation being generated in the quickly passing frontal zone. This was mainly the result of the numerical guidance keeping the mid-level (short) wave (trough) “open” and preventing the development of what is known as a closed low. Beginning with today’s 12 UTC (7 AM CDT) numerical models, the forecast of an “open wave” during the weekend has dramatically changed. However, before I jump into all of this, let’s first take a look at what I mentioned a couple of days ago - the chance of severe thunderstorms on Friday.
There are several things working against the chance of a widespread severe weather outbreak tomorrow. However, the biggest one is the lack of deep, rich atmospheric moisture. (Moisture is an essential ingredient in the development of any thunderstorm.) The image above depicts the amount of moisture currently throughout the troposphere. Notice how the Gulf of Mexico has relatively low moisture values. This is because a storm that moved through the Gulf of Mexico during the last 36 to 48 hours has “cleared out” all the available moisture and shoved it south and east out of the Gulf. This means that it will take some time for the moisture to come back into the southern plains – most likely longer than we have.
Another ingredient necessary for severe thunderstorm development is a source of lift (or rising motion). The image above is from this evening’s 00 UTC (7 PM CDT) North American Model (NAM) and depicts a forecast of rising motion (lift) at 00 UTC (7 PM CDT) tomorrow evening. What is important about this forecast is that the rising motion is found in a skinny band located along the cold front and not in a widespread area ahead of the cold front. This leads me to believe that the warm sector will most likely be void of enough lift to develop thunderstorms, so we’ll be forced to rely on the cold front to generate enough lift for severe thunderstorms. This means there will really only be one shot for severe thunderstorms – not repeated chances – however thunderstorms will certainly be possible.
Lastly, severe thunderstorms need to have a lot of instability. One measure of instability is known as Convective Available Potential Energy (CAPE). The image above depicts the amount of CAPE at the same time as the previous image’s vertical velocity (lift) depiction. Notice how there is hardly anything there! This means that the entire line of thunderstorms that might develop along the cold front would have to share what little instability is depicted above. This will also act to limit the severe threat.
And, lastly, the snow potential. It’s a bit late, so I won’t go into all the details that I had previously planned to cover. The image above from the National Weather Service Forecast Office in Norman, OK, pretty much says it all. A good portion of Oklahoma will see the potential for a significant snowfall over the weekend. Keep in mind that these are “accumulation” forecasts, or in other words, the amount of snow you’ll see on the ground. This is impressive when you consider that a lot of the area will see temperatures in the upper 60s to lower 70s tomorrow. Ground temperatures will be incredibly warm at the start of the event, meaning that a lot of snow will melt initially. A lot of snow would have to fall in order to get accumulations forecast above. Actually, if we took some of the models at face value, portions of Oklahoma could see 1-2 FEET of snow…
With all this said, referring back to the first image, I have concerns about moisture into this storm. This could be one potential failure mode. This certainly bears watching…
