Ramblings of a Graduate Student

I’ve dreaded having to write this post for most of  the day.  When I woke up this morning, numerical guidance from overnight and this morning almost unanimously indicated the potential for a significant winter storm in central Oklahoma.  Deep down I knew that forecast and that kind of agreement wouldn’t last.  I was fairly sure that by the time I sat down to write tonight’s post, I would be even more uncertain about tomorrow’s weather events than when I woke up.

I was correct.

Instead of walking through various predictions, which are all over the place, I thought I’d attempt to explain what tonight’s 00 UTC (7 PM CDT) North American Model (NAM) appears to be doing.  The idea is that instead of giving pretending this model is correct, I’ll attempt to explain why it is producing the forecast it is producing.  Someone could then watch the evolution of these features overnight into tomorrow to have a better idea of where and when the heaviest snow will fall.  Here we go…

The next three images are 30, 36, and 42 hour forecasts, valid Sunday morning at 1 AM CDT, Sunday morning at 7 AM CDT, and Sunday afternoon at 1 PM CDT  respectively.  The color fill is the model’s predicted 6 hour accumulated precipitation for the previous six hours.  As a first guess, anywhere underneath blue dashed lines could expect to see snow or sleet, whereas places in the red dashed lines would expect to see rain.

Notice in the image above the light green color fill extending from northwest Texas, northeast to an area of heavier precipitation along the Arkansas, Oklahoma border.  This would be precipitation resulting from the mid-level low circulation centered around 700 mb.  It is often referred to as “wrap around” snow or “comma-head” snow.  The latter is more of a descriptive term since when this portion of a storm is at it’s most potent, it looks like the top portion of a comma.  At this time, the precipitation in the comma head is fairly light.

However, if we step forward in time 6 hours, we begin to see darker green shading develop in the comma head.  At the same time, the darkest green color begins to shift westward – opposite to the large scale flow.

Stepping forward yet another 6 hours, we see the trend continuing.  This would seem to indicate that the heaviest snow in the comma head would actually occur overnight Saturday night into early Sunday afternoon.  The key to being able to identify where and when the heaviest snow will occur with this storm is to identify what processes are occurring in the model that result in an increase in the comma head precipitation.  If we can identify these processes, we can then watch for them tomorrow, and adjust our forecasts accordingly.

Without going into why I knew to look here, let’s take a look at the 300 mb level, home to the jet stream for this storm.  The next three images are all of the 300 mb level and are 30, 36, and 42 hour forecasts just like the previous images.  The color fill in these set of images represents the forecast wind speed, with darker colors indicating stronger wind speeds.  The wind direction can be determined by the “barbs” (look like a stick and with more sticks and flags hanging off it).  The portion of the barb without anything hanging off of it is the direction the wind is blowing to. So, in southern California in the first image below, the wind is blowing from the northwest to the south east.

The key feature in the image above is the patch of darker shading located near the Big Bend of Texas.  This is known as an upper-level jet streak.  Notice how it is still to the west of the axis of symmetry in the upper-level trough.  Watch what happens to this jet streak as we step forward in time.

Notice how the jet streak has become longer and is now on the east side of the upper-level trough.  Again, we’ll step forward 6 more hours.

Now the jet streak is even larger and almost entirely on the east side of the upper-level trough.  This coincides with when the heaviest precipitation develops in the comma head (shown above).  So the question is, how are these two phenomena related?

The answer is somewhat subtle.  If you look closely at the middle image of the precipitation plots, you’ll notice that a “L” appears in central Arkansas.  The appearance of the “L” on that map indicates the development of a surface low pressure.  This surface low pressure rapidly strengthens as the left-exit region (northwest portion) of the jet streak spreads over the surface low.  As the surface low continues to develop and become stronger, the winds on the north side of the low also strengthen.  A consequence of all this strengthening is the rapid advection of moisture into the comma head.  An over simplification of what happens is that because there is more moisture available in the comma head, more precipitation can fall.

So, the key to tomorrow is knowing where and when the upper-level jet streak will overspread the surface low pressure.  When this occurs, the entire storm will strengthen and the precipitation rates (should be mainly snow and sleet) will increase in the comma head.  However, one thing to remember is that a slower upper-level jet streak would indicate that the mid-level cold air (necessary for snow) will be slower to arrive than previously thought.   This would indicate a slower transition from rain to snow, and result in a wetter snow that is easier to melt on a warm ground.  It will be interesting to see how this evolves…

Note:  The “left-exit” region of a jet streak is not always on the northwest side of the jet streak. Ultimately, the “left-exit” region is dependent on the orientation of the jet streak.

Note:  For completeness, the surface low pressure system isn’t the only part of this storm that strengthens as the upper-level jet streak begins to overspread the surface low.  The entire storm strengthens.  Examine the plots below. Notice that the number of black contours increase in the middle of the “low”. Also notice how the numbers associated with those contours (black lines) also become smaller. These are signs of a strengthening storm.

The next three images are the evolution of the 500 mb low at the same time steps as above.

The next three images are the evolution of the 700 mb low at the same time steps as above.

The next three images are the evolution of the 850 mb low at the same time steps as above.

The image above was taken approximately the same time a tornado was hitting Hammon, OK.  The tornado occurred with the convection associated with the brighter colors in far western Oklahoma.  It was a classic cold core setup, meaning a lot of instability was generated along the eastern edge of the cold-pool aloft associated with the mid-level low.  It so happened that the mid-to-upper level jet streak (cyan color fill below) associated with the mid-level low was positioned such that an area of large scale lift was juxtaposed over the mid-level cold pool.  These two factors, combined with surface convergence along subtle surface boundaries evident on KFDR radar (not attached) enabled thunderstorms to develop in an environment with a lot of environmental vorticity.  At least one thunderstorm this afternoon was able to tap into this environmental vorticity and produce several tornadoes.

Storm season is officially “here”.

A very active upper-air pattern persists across the United States with several speed maximums inferred in tonight’s water vapor imagery.  A powerful (fast) upper-level jet stream, with embedded jet streaks, is evident across the Gulf of Mexico into the far western Atlantic Ocean.  This jet streak is aiding the development of the northeast “snowicane”, as I’ve heard it referred to by some in the media.  A mid-level jet streak is propagating southeast out desert southwest into central Texas.  This mid-level jet streak is aiding the development of a mid-level low over the northern Texas Panhandle.  This mid-level low will slowly move southeast tonight, bringing with it the chance of a cold rain and even some snow (although, I’m not nearly as optimistic tonight about seeing much in the way of accumulations in central Oklahoma).

Below, I’ve annotated the mid-level lows (yellow X) and jet streams/jet streaks in thick cyan arrows.  The thing cyan arrow is not necessarily a jet stream, but more of an indication of the flow pattern over the northeastern United States.

Lastly, I thought I’d show a graphic that helps explain why I’m not so gung-ho about central Oklahoma’s snow chances tonight/Friday morning.  Below shows the northern hemispheric 500 hPa heights (color filled with lower heights in cool colors and higher heights in warmer colors) and surface pressure (solid black contours).  The Day 50 post discussed how it appears a surge of arctic air was preparing to plunge southward out of Canada into the southern plains.  Well, tonight’s northern hemispheric chart (below) clearly shows that the cold air did not plunge south into the southern plains.  Thus, without a true source region of cold air needed for heavy snow, central Oklahoma should stay a tad too warm for much in the way of accumulating snow.

Now, central Oklahoma might see snow overnight, and here is how it would happen.  As the upper-low and corresponding surface reflection move southeast across far eastern Texas panhandle and southwest Oklahoma into north central Texas, a lot of dry air from the east will become entrained into the storm.  This dry air will allow evaporational cooling to occur, allowing surface temperatures to cool sufficiently for snow to reach the ground.  However, whenever dry air is being continually advected into an area, this tends to decrease precipitation intensity…which would cut-back on any snow accumulations that might otherwise occur.