Wednesday 30, 2015

Usually, I keep my weather thoughts local, but hurricanes seem to catch people’s eyes no matter where they are.  So, some  things I find interesting about what I’ve seen concerning Hurricane Joaquim today.

First, the National Hurricane Center’s current forecast path chart (below) hugs the east coast from the Carolinas all the way to eastern Canada.  However, no model I’ve looked at shows anything like that.  I know the projection is a composite, but it’s not really telling us anything.  (Actually, it’s probably intended that way.  CYA as the saying goes.)

Also, the chart’s timing doesn’t seem to follow much of the guidance, either.

Now, let’s try to really do some forecasting of Joaquim.  First we have two “outlyers” in the models.  The European takes the storm far off the east coast with the only problem being a pass near Bermuda, but probably only as a fading tropical storm.  The NAM basically keeps it stationary for a couple of days, then rips it apart as it begins to move (upper shear increases).  But this still leaves the possible of same very heavy rains along the east coast over the weekend into early next week.

Now, the GFS models, joined by the Canadian and Navy models) point toward a far different solution – one that makes a lot of sense to me.  But it does lie outside of the “official” guidance envelope.

Nevertheless, here’s what the models say to me at this time.  The weak upper air disturbance which has brought the very much needed rains to us the past two days has basically “cut off” for the main upper level flow and will drift slowly southward for next couple of days – basically a small pool of cold air aloft to keep weather unstable over the areas south and east of the Ohio River.  By late week, another pocket of cold air aloft with arrive to strengthen the “cutoff” and push it a little eastward – just far enough to absorb the circulation of Joaquim and drag it westward into the southeast coast.  When the cutoff low absorbs we start developing the major transition of Joaquim from a hurricane (or tropical storm) into what’s now called an extratropical low  (Years ago the NHC started calling these rare hybred systems “neutercanes” but the name never caught on.)

In  laymen’s terms what that means is mid latitude low pressure dynamics coupled with tropical moisture.  The usual result is not much of a wind problem, but PLENTY of rain and flooding problems.

Assuming the above process happens, here’s what I expect – Tropical Storm (or small hurricane) Joaquim will hit Saturday night along the Carolina Coast – probably between Myrtle Beach and the Outer Banks.  The decaying storm will then drift around the Carolinas for up to 48 hours causing massive rains and major flooding.  Heavy rains will also move up the east coast Sunday, Monday and Tuesday.

Update tomorrow!

Note:

So far  this year eight people have been killed by sharks.  And, 12 people have been killed while taking “selfies.”    Message:  Beware the selfie!

Friday afternoon (Mar 13)

It’s been a week with two very nice early spring days, but also with some very wet days as well.  The second major rain system of the week is over us now and looks like it’ll hang around until at least midday tomorrow.

One reason for the warm wet weather has been the absence of the northern (Polar) jet stream.  It has receded far north of us this week allowing weak subtropical weather disturbances to drift slowly into the Ohio Valley.  This one is more widespread and slower moving than the one earlier this week, so another large dose of rain is likely.  Our area could easily see 2″+ plus of rain before it fades away tomorrow afternoon.  So, once again, we’ll see renewed urban flooding and the larger rivers and streams will (perhaps) see a second crest this weekend and an extension of time remaining above flood stage.

The northern jet is about to reenter the weather situation this weekend.  It’s influence on us will start slowly tomorrow when it sends a batch of cooler, drier air to end the current rainy weather pattern.  The jet will ease a bit early in the week (pleasant Sun, Mon and Tue), but will send us a colder air mass by mid week and an even colder one by next weekend.  Yes, winter’s not over.

NWS Radar rain estimates.

I’ve long been amazed at how well the “rainfall total” estimates from the National Weather Service’s radar match up with the so-called “ground truth” –  the totals measured by surface rain gauges.  I was amazed again Monday, but for a completely different reason.  This time the radar estimates over the whole range of the radar were all BELOW one inch (.6″ to .99″ range).  However, the rain gauges at the airport and Bowman Field both recorded about 1.5″.  So, actual measured values were more than 50% higher than radar estimates!  Something is waaaay off.  I don’t know why, but I do have a theory.  The algorithm  currently being used by the radar is probably designed for winter weather.  In the winter, in rain/snow situations, radar will often display an area of high reflectivity.  That looks like an area of heavy precipitation.  In reality, what it is usually “seeing” is a region where snowflakes are melting as they fall toward the ground.  Thus, you have snowflakes surrounded by liquid water.  The radar interprets that as huge raindrops – thus assuming heavy rain which creates an artificially intense radar return.  The common term for this is bright banding.  So, the winter radar algorithm makes allowances for this bright banding and reduces the “radar detected precipitation” accordingly.

However,  in a warmer situation when the freezing level is far above us (like the two rain situations this week), radar echoes displaying the same levels of intensity as winter “bright banding” are, in fact, measuring just what they seem to be – heavy rain.  Thus, using winter schemes during a spring rain will greatly reduce the “radar detected rainfall.”

Now, this seems like a problem with an easy fix – change to the correct algorithm.  Evidently, however, my theory must not be correct.  Because it’s happening again today!  As of 5 P.M., radar rain fall estimates for an area roughly 40 miles around Louisville showed about 75% of the area rain totals below .30″ while the other 25% have had between .30″ and .59″.  Meanwhile, SDF and Bowman Field (LOU) rain gauges were reporting between .70″ and .80″ of rain so far.  But then, what’s that old line – Fool me once, your fault.  Fool me twice, my fault.?

Very unusual Pi Day tomorrow.

Within the past two decades or so, math and science oriented people have thought that the very important math symbol “pi” should get some credit.  pi’s value is 3.14…, so March 14 or 3/14 is celebrated annually as “Pi Day.”  But, tomorrow is an extra special version of the day – it will not be repeated for a 100 years!  Going out a few more numbers in the value of pi we get 3.1415…  3/14/15 is tomorrow’s date.  Then the next three numbers of pi are 926.  So, tomorrow at 9:26 A.M. and 9:26 P.M. we’ll have the following number string – 3/14/15 9:26  The most digits of pi you’ll ever see on Pi Day (unless you live a very long time).  And, if you want to go to seconds, the next two numbers are 53.

    Now, that seems to be a problem with an easy fix – change to the correct algorithm.  Evidently, however, my theory may not be correct.  Because it’s happening again today!  As of 5 P.M., Radar estimates within 40 miles (or so) of Louisville showed about 75% of the area with rain less than .30″  and the rest of the area in the .3″ to .60″.  However, both airport rain gauges in Jefferson Co.  were reporting totals between .7″ and .8″.