Winter
2018-2019 was a winter of stark contrast. But going back and reviewing what My outlooks
said. I didn't get everything right, but the overall pattern was close to what
I said it would be like. I was closer to what actually happened than
what many of y'all think.
This post
will try to show and explain the things that caused winter 2018-2019 to act the
way it did.
This winter
I've had to deal with the reality of the situation VS the precipitation of the
situation. There is a wide gulf between the two. Why was winter 2018-2019 so close to my
outlook, yet so far? As I said, my chosen analog years did come close to the
overall pattern. But as I said in my three part pre outlooks released in
September and the final ones released in October, an outlook is based on an
average of all those analogies; so different parts of those analogies will show
up at some point. But more
So what
happened?
The
Madden-Julian Oscillation (MJO):
The MJO and
the Southern Oscillation Index (SOI) didn't work together very well during
winter 2018-2019. As a rule of thumb,
during the winter Phases 1&2 are cold phases, while phases 5 &6 are
warm phases. The other phases tend to be more transitional, going from warmer
to colder and vice versa. The farther the MJO is from the center Null circle in
the center of the diagram, the stronger the MJO is considered . The MJO went through the warm phases twice
with a very high amplitude. The first
one was in December of 2018. But then we went through the Sudden Stratospheric
Warming; which lead to the Polar Vortex split in early January. During with
2017-2018 the split in the Polar Vortex happened in Mid February. That was the
major reason April 2018 saw near record cold. With the PV split in 2019
occurring in January, we saw the cold for February and extend through March
2019. For much of winter 2018-2019 the
temperature and precipitation anomalies didn't match up that well to the
pattern. The MJO and SOI not really
phasing this winter, is undoubtedly the major reason the models had so much
volatility during winter 2018-2019.
Figures 13
and 14 show how the SOI behaved.
Figure 13
Figure 14
Subtle
Nuances:
When I went
back and placed the winter of 2018-2019 under my microscope; at first things
looked like they should. But under higher magnification, subtle differences
came into focus. there were several
little things that stood out as strange. Taken by themselves none of these
variables were enough to have any real impact; at least to the degree and scale
of what happened. The power of these
nuances was in their cumulative effect on the overall pattern.
The El Nino
Southern Oscillation (ENSO):
For much of
the Fall of 2018 the Pacific equatorial region was warmer than average. It was
clear we were going into an El Nino. By late fall the Pacific looked like we
were in a Modoki (central based) El Nino.
When we deal
with the ENSO we're dealing with how
warm sea surface temperatures are, and where those warm SST are
sitting. Depending on this affects wind
speed, and direction, due to a temperature gradient that develops. This in turn
determines where convection and precipitation will set up.
Figure 1
The
temperature gradient is called the Walker circulation. During El Nino neutral conditions, we have an
east-west temperature gradient; where the warm SST in the Western Pacific form
a huge contrast to the cooler SST in the Eastern Pacific. The direction and
strength of the Pacific trade winds, depends on the strength of the Walker
circulation. So, when we have neutral EL
Nino conditions, the trade winds blow from East to West. During La Nina, the
Walker Circulation is very enhanced. So we have very strong trade winds blowing
from East to West. Now, during an El
Nino the Eastern Pacific warms and the western Pacific cools; this causes the
walker circulation to normally reverse. So, the trade winds blow from West to
East. This typically causes the main area of convection and rainfall to setup
in the Central Pacific. But this
reversal in the gradient doesn't always
happen the same way.
It wasn't
until February 14th, 2019, that a weak
El Nino was finely able to be declared by NOAA. During the Fall and into Winter
2018-2019, we had a weak El Nino Modoki signal. In fact at times we exceeded the El Nino 0.5 ⁰ C threshold. But the atmosphere didn't respond as it was expected to. The reason for this
lack of and very delayed coupling with the atmosphere and ocean, very likely
has to do with the behavior of the walker circulation.
In the Fall
of 2018 the expected enhanced convection and rainfall in the Central and
Eastern Pacific never materialized. Here
are some images that show how Fall 2018 compared to what we would have expected
under weak El Nino conditions.
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 2
shows how the equatorial Pacific rainfall pattern looked like for Fall of 2018.
Figure 3 shows how 2018 stacked up in a scatter plot. When we look at the figure 5, We compare the
SST pattern in 2018 to what would be expected from a typical El Nino of the
same strength; we see a pattern that looks very La Nina like. This is roughly
the opposite of what was expected. When I looked back at pass years that
developed El Nino conditions by winter, the only match I could find was in
2014. The winter of 2014-2015 had a similar pattern. NOAA didn't issue an El Nino
advisory until March of 2015 for that season.
The Fall of 2014 had the same lack luster convection and lack of
rainfall in the Central and Eastern Pacific as Fall of 2018. Figure 4 shows what the SST pattern would look
like during a neutral El Nino would look like. Figure 6 shows how the SST pattern and the convective
rainfall pattern from the temperature gradient combined. The result of which
was quite similar to what we actually ended up with.
The 500 mb
during February 2019 showed above average heights over the Gulf of Alaska and
the Eastern U.S. At the same time we had
below average heights over the Central and Western US. These anomalies are the exact opposite to
those typically seen during an El Nino.
The height anomalies show the westward retraction of the Pacific Jet
stream and the downstream amplification in the circulation.
Clearly the
investigation has found a major culprit for what happened. But it's not the
only one.
The
Tropical/Northern Hemisphere pattern (TNH):
I've been
reading some post by Thomas E. Downs and Joe D'Aleo at WeatherBELL Analytics LLC. They raised the idea that the TNH
teleconnection was one of the reasons for the trough/ridge temperature pattern
this past winter.
The TNH is
only calculated during the Months of December, January, and February. Normally
during an El Nino the TNH is negative. But during the Fall 2018 into Winter the
TNH index was positive. Not just positive but very positive. Given the nature of the weak Modoki El Nino
signal this is quite puzzling.
Figure 7
Figure 8
Looking at
figures 7 and 8 we can see the positive THN footprint and the resulting
temperature pattern that was the overall thyme for this past winter. Like the
PNA, the THN plays a big role in the placement and eastward extent of the
Pacific Jet. For most of the time
during December through February we had
a split Pacific Jet pattern. The southern split brought storms into the West
Coast with a southern storm track. While the northern split blocked the storms
away from the East Coast. We also had a weaker trough centered over the Hudson
Bay than was expected.
Looking back
at past winters that had a weak El Nino. We find 1958-1959, 1976-1977, and
2006-2007 also saw very positive TNH index readings. Figures 8,9,and 10 show a
composite of how those Februarys turned out.
February 1959 matches up very well with how February 2019 turned out.
Figure 9
Figure 10
Figure 11
The rest of these figures show how the temperature, precipitation, and teleconnections stacked up during February.
Figure 12
Figure 13
Figure 16
Figure 17
Figure 18
Figure 19
Figure 20
The overall
storm track for winter 2018-2019 can be laid at the feet of the rare positive
phase of the THN.
The North
American Oscillation (NAO) and the Arctic Oscillation (AO):
The NAO and
the AO were at odds for most of the winter. But ended up being basically zero
to positive for much of Winter 2018-2019.
The Atlantic
Multi-decadal Oscillation (AMO) and the Pacific North American :
Just like
winter 2017-2018 the AMO was negative for winter 2018-2019. As for the PNA is
was positive for the first two thirds of January, then for the last 10 days of
January the PNA was weakly positive to negative. We also, had a negative PNA
during February 2019.
All four of
these teleconnections had a part to play in the lack of Upstream Greenland
blocking and lack of any major Nor'easter storms along the Middle Atlantic and
Coast of New England.
Well that's
about it. This shows my analysis of the two primary reasons and a few negligible
reasons, for winter 2018-2019 temperament issues.
