Wednesday, July 24, 2013

Skew-T, Part 3 -- Historical Skew-T Diagrams

(Note: This article is Part 3 of a series on interpreting Skew-T/Log-P Diagrams. If you have not already done so, you can read Part 1 here and Part 2 here.)

When I left you Monday, I promised you we’d examine some real soundings plotted on Skew-T diagrams. And while many of these cases will be from the Philadelphia area, we’re going to start down in Birmingham, Ala. so that we can examine what a classic severe weather day looks like in sounding form.

The University of Wyoming’s Meteorology Department has soundings archived from 1973, so we will use their images for this discussion. Notice that on their Skew-T, DALR lines are green, Mixing Ratio lines are purple, and every other base line is blue. Temperature and Dew Point are both in solid black, with Dew Point obviously being the line to the left (we’d certainly hope it’s always to the left of the temperature curve!).

This sounding is from 00z April 28, 2011 (which, remember, is 7 p.m. CDT on the 27th).

You probably remember April 27, 2011 as the day that Mississippi, Alabama, and much of the South got raked by multiple long-track violent tornadoes. And given this sounding, it’s easy to see why.

There is a robust Elevated Mixed Layer in the sounding, as depicted by a large area of very dry air aloft.

The Temperature Curve nearly parallels a DALR line for some period, which means that Mid-Level Lapse Rates are very steep.

There is a very low LCL. You can estimate it, but this particular site also computes it. At right, the LCL is indicated to be 917.5 mb, which was 293.7 meters.

CAPE is also computed. 2944 J/Kg is impressively high. (Summertime CAPEs in the Plains can reach 5000 J/Kg, but those atmospheres are almost always capped. This sounding has a CIN of 6.19 J/Kg, which is minimal. For reference, about 70-80 J/Kg of CIN would signify a difficult-to-break cap, and greater than 150 J/Kg would be unbreakable.)

And lastly, wind vectors (plotted to the right of the Skew-T) show wind veering from southerly at the surface to westerly in the mid-levels, and increasing in speed with height, which is classic for supercells and tornadoes. (On those vectors, each small notch represents 5 knots, each long notch 10 knots, and each flag 50 knots.)

We can ascertain all of this information from one chart, which is why the Skew-T is so useful.

But I promised that we’d look at some Philadelphia-centric cases, so let’s look at October 1, 2010 – a heavy rain event for much of Pennsylvania.

Now unfortunately the nearest upper-air sites to Philadelphia are in Virginia and on Long Island, so we’ll have to work with those. The image below is from 00z on October 1, 2010 at Wallops Island, Va., which is on the extreme northeast coast of the Virginia peninsula.

What’s remarkable about this sounding is that the entire profile is nearly saturated, as seen by the Temperature Curve and Dew Point Curve being very close to each other, and that the wind at every level is southerly, helping transport moisture from the Gulf and Atlantic at all levels of the atmosphere.

The Wyoming site also plots Precipitable Water, or PWAT, which in this case is 65.15 mm, or 2.56 inches. Any PWAT values above 1.50 inches indicate the potential for heavy rain, and PWAT values over 2.00 inches are extreme.

As for what happened on October 1, 2010…

I also promised before that Skew-Ts were useful in winter as well.

Snow was supposed to fall on the evening of December 5, 2005. Most media outlets were predicting the snow to begin by 4 p.m. But the evening news opened with…

So what happened? Well, take a look at the sounding from 12z that morning (at nearby Upton, N.Y.):

Notice how incredibly dry the atmosphere was just nine hours before snow was supposed to begin.

Unfortunately for us snow-lovers, dry air was still a problem at 00z:

Yes, the lower levels of the atmosphere moistened significantly over the course of the day, but that dry layer around 900 mb was still enough to evaporate snowflakes before they reached the ground.

Snow did eventually fall, but it only accumulated to 4.0 inches at Philadelphia International (where the NWS was forecasting 5 to 7 inches), and areas northwest of the city saw significantly less.

But dry air isn’t the only problem snow weenies face in winter. Often sleet is in the forecast as well. And what more classic day is there to talk about sleet than March 16, 2007?

This sounding from 18z on the 16th at Dulles Airport depicts a classic set-up for sleet: 

I’ve highlighted the 0 C isotherm in teal since otherwise it might be hard to see.

Notice that as precipitation is falling through the atmosphere, it is melting at about 850 mb, but then refreezing just below that where the Temperature Curve takes a sharp bend to the left. When snow melts as it falls, but then refreezes before it hits the ground, we get sleet.

In this particular sounding, the surface is then once again above freezing, so sleet may not have been accumulating there, but in Philadelphia’s north and west suburbs, most places saw 4 to 6 inches of solid sleet.

Now contrast that sounding with the one below, from 12z on February 2, 2011 at Upton, N.Y.: 

In this sounding, as in the last one, the Temperature Curve warms above freezing and then cools below freezing before the surface. Only this time, the area where the temperature is above freezing is much larger than before, and the area with below-freezing temperatures much smaller, which means that freezing rain, rather than sleet, is falling.

If the warm layer is significant enough to completely melt the snowflake, and the cold layer below it is too shallow to allow enough time for it to refreeze into an ice pellet, then you get liquid rain hitting a below-freezing surface and freezing on contact, or as we know it, freezing rain.

Obviously there is still a lot more we could talk about, and we could pull soundings from historical weather events and analyze them until we die, but this little lesson was hopefully enough to convey an understanding and when, how, and why to use Skew-T diagrams.

Now try it for yourself! Access archived soundings from any date you remember here and test what you’ve learned. (And look up Station Numbers here. They are the five-digit numbers starting with 7 in parentheses after each site name.)