Ramblings of a Graduate Student

A quick look at tonight’s Oklahoma Mesonet (specifically, a dewpoint plot) conveys the message that there is a lot of low-level moisture located across much of Oklahoma. Dewpoint temperatures in the mid 70s indicates a very moist atmosphere, no matter what time of year.

Contrast this to the dewpoint temperatures of the upper 30s in far western portions of the Oklahoma panhandle. Glossing over all the details of why it is so dry in the western panhandle, the boundary depicting the dry air to the west and moist air to the east is known as a dryline. Tomorrow, this dryline should mix (move) eastward to around the Oklahoma-Texas panhandle border. During the spring, dryline setups such as this often result in isolated severe thunderstorms that occasionally produce tornadoes. Tomorrow the mid-to-upper level winds are currently forecast to be fairly weak (at least across Oklahoma), limiting the threat of widespread severe weather. The more concentrated threat will occur across Kansas. With this said, it doesn’t matter if the severe weather was widespread or localized, if it affects you it’s still bad.

The National Oceanic and Atmospheric Administration’s National Weather Service Storm Prediction Center has issued a slight risk of severe thunderstorms tomorrow evening and overnight into Friday for a portion of the central United States (click here to view the outlook).  Conditions tomorrow afternoon are such that if, and that’s a big if, thunderstorms can develop in the central plains, they would have a good chance at producing severe weather.  So, how does a meteorologist go about trying to create a forecast for such a conditional threat?

To do so, we need to think about what ingredients are necessary for thunderstorms to develop.  Simply put, the necessary ingredients are moisture, lift, and instability.  In an effort to keep this blog post from being extremely long, I’ll go ahead and stipulate that, although not as great as it could be, enough instability will be present to support thunderstorms.  So what about the other two?

The image above is Wednesday morning’s 12 UTC North American Model (NAM) forecast of 2-meter dewpoint temperatures (a measure of moisture) valid at 21 UTC Thursday afternoon (4 PM CDT).  The greater the dewpoint temperature, the more moisture available in the atmosphere.  A quick rule of thumb is values greater than 50F are often needed for severe thunderstorms and a value greater than 60F are often needed for tornadoes.  (Remember, this is just a rule of thumb.  Severe thunderstorms and tornadoes can, and do, occur with dewpoint temperatures less than these rule of thumb values.) It is readily apparent by looking at the image that dewpoint values greater than 50F are forecast to be in place across a good portion of the southern and central plains.

However, on thing should stand out about this plot.  Notice the sharp edge on the western side of the higher moisture, whereas the eastern side of the moisture area is very diffuse.  This sharp gradient from moist air (east) to dry air (west) is known as a dryline.  A simple explanation as to why drylines develop is that on the dry side of a dryline, the wind advects dry air toward the moist air.  At the same time, on the moist side of a dryline, the wind advects moist air toward the dry air.  The more opposite the wind directions and the faster the wind speed, the sharper the moisture gradient.

Drylines are important in forecasting thunderstorm development because they can end up being a source of our second ingredient, lift.  This is a result of the wind on either side of the boundary converging at the same point.  When this happens, the air has two choices it can either go up or down.  Since the ground is below the converging air, it really can’t go down; this leaves “up” as the only option.  So, along a dryline, one can expect to find rising motion (lift), which is one of the main ingredients for thunderstorm development

The image above, valid at the same time as the previous dewpoint image, is of vertical velocity (also known as lift) at 850 milibars (about a kilometer above the ground over the central plains).  Areas shaded in warm colors indicate places with rising motion (ascent) and cool colors indicate areas of descent (sinking motion).  Notice how there is very little rising motion found along the area we identified as a dryline?  This is because the wind converging along the dryline isn’t particular strong at this time.  As such, only small pockets of rising motion would be expected.

The image above is the exact same as the previous image, except it is for 700 milibars (between two and three kilometers above the ground over the central United States) instead of 850 milibars.  Notice how the rising motion is a little bit stronger, but not by much.  Again, the wind convergence is not particularly strong.

So what does this mean for thunderstorm chances?  The image above is from an ensemble forecast (a bunch of models run together) run at around the same time as the NAM and is valid at 00 UTC Friday (7 PM CDT Thursday).  Areas contoured in red are places where at least one forecast model produced a thunderstorm during the last three horus, areas contoured in blue are places where every member produced a thunderstorm during the previous three hours, and areas in black are where the mean of all the forecasts produced a thunderstorm during the last three hours.  Notice how not a single model generated a thunderstorm over the dryline area between 21 UTC and 00 UTC (4PM CDT and 7 PM CDT).  This can almost certainly be attributed to the “lift” ingredient being insufficient to overcome a lack of really good moisture.

The next several images are valid 3 hours later than the images above (oo UTC / 7 PM CDT).

Notice how the dryline remains in roughly the same place.

However, notice how the rising motion at 850 milibars has increased along the dryline.  This implies that convergence along the dryline is increasing, resulting in greater lift.

We find the same thing at 700 milibars.

And consequently, some of the forecast models generate convection between 00 UTC and 03 UTC (7 PM CDT and 10 PM CDT).

Like last night’s post, this post is designed to demonstrate how a good meteorologist will interrogate model output, and won’t simply take it at face value.  A good meteorologist will attempt to understand why a model is doing what it is doing in an effort to have a better conceptual model about the event in question.

So, does this mean there won’t be storms tomorrow afternoon?  No.  The resolution of these models is such that it cannot resolve all the details of thunderstorm development.  What this forecast does indicate is that in order for thunderstorms to develop, some local processes are going to have to play a role.  This gives a good first guess as to what a meteorologist will need to watch during the day.

Note: The description of lift resulting from drylines given in this post is an oversimplification of what is actually occurring.  Meteorologists still have a lot to learn about the circulations and air-flow patterns along drylines.

Yesterday was a travel day for the armada as they headed north from Yukon, OK to Wichita, KS.  The thinking was that since severe thunderstorms would be possible anywhere from the Texas panhandle northeast through northern Missouri, staying in Wichita would provide the best opportunity to cover the entire risk area.  Unfortunately, Mother Nature didn’t show her hand this morning and provided us with a difficult forecast.  The armada’s choices were

1. Head northeast toward Emporia, KS / Kansas City, MO where there were better winds in the middle portion of the atmosphere and the possiblity of an outflow boundary from morning convection (which VORTEX I found can aid tornado development).
2. Stay in Wichita until late in the day and hope something develops nearby.
3. Head west toward SW KS / NW OK / northern OK & TX panhandles.  This area might have a brief opportunity for decent mid-level winds, east or southeast surface winds, and discrete storms.

The initial consensus was to head northeast toward Kansas City, however a few people (myself included) thought it would be better to go west toward the Texas panhandle.  Becuase of this, the armada decided to wait until 11AM to make a decisions, which is a delay of about an hour.

At 11AM the decision was made to head to Enid, OK, which is southwest of Wichita, but in the same general vicinity.  The armada convened in Enid around 2PM and then waited for thunderstorms to develop – which happened around 4PM in Kansas.  Additionally, an isolated supercell thunderstorm developed in the NE TX panhandle, where the cold front and dryline intersected.  (This thunderstorm went on to produce several brief, weak tornadoes.)

The thunderstorms in Kansas were developing southwest along the cold front and so the armada spent most of the afternoon waiting for the thunderstorms to come to them.  While they were waiting, the storm in the TX panhandle occurred and additional tornado warnings were issued for the Wichita, KS vicinity.  (Ironic, I know!)  By the time the thunderstorms reached the armada they were “gusting out” meaning that the rain cooled air was pushing out ahead of the storm, robbing it of the warm, moist air needed to sustain itself.  Along the leading edge of this gust front, things that resemble tornadoes often develop.  These are not true tornadoes (which occur with updrafts), but rather “gustnadoes”  (because they occur on the leading edge of the downdraft, which is known as the gust front).  One of these gustnado-like structures developed very near one of the mobile mesonets, which reported a brief funnel cloud.  However it never came in contact with the ground and was never really a serious threat.

All in all, today was a good day in terms of learning.  The armada was able to deploy and collect coordinated data, even though it wasn’t exciting.  (The next step will be to collect coordinated data on an exciting storm!)  The armada called it quits after the storms passed them and headed to their hotels.  Tomorrow, through at least the first part of the week looks to be “down” days because of the strong high pressure building over the central portion of the US.  This will limit severe weather chances for the forseeable future.  However, the armada is not taking it easy.  Tomorrow they will head to a new portion of the central plains and do “practice” for the next several days, in hopes that they become a well oiled machine for what appears like operations in South Dakota or North Dakota around mid-week.