What the Hell Is Going On?

A lot of people have requested a blog post explaining what the hell has been going on this winter.  Well here goes.

I don't know.

At least that's the tl;dr summary.

OK, I can do a little better than that, but explaining the whims of the atmosphere is very difficult and I don't want to suggest that I'm some sort of mystic who knows what is going on.  I'm not sure anyone is.  

Let's start in the fall.  As usual, a variety of seasonal outlooks were issued by various groups.  These always get a lot of attention, but their value is often limited.  

I will use the official forecast issued by the National Weather Service Climate Prediction Center as example.  Below is the three-month outlook issued in November for December through February.  Due to the presence of La Nina (which has now waned), the forecast suggested that the dice were loaded for a dry southwest and a wet northwest.  Utah could go either way.  

Mother Nature, however, had different ideas, and for the mountain west, basically flipped that forecast on its head, producing a wet southwest and a dry northwest.  

Mother Nature 1, NWS Forecasters 0.

To be fair, the forecasts issued by the Climate Prediction Center are probabilistic.  They don't say that the southwest will be dry and the northwest wet. Instead, they forecast a shift in the likelihood of possible outcomes. Pure chance for any given winter gives a 33.3% chance of above, near median, or below median precipitation (in this case, median is broadly defined as a range around median).  Equal chances, as was forecast for Utah, says there's no loading of the dice and they simply don't know what is going to happen.  Below median, as was forecast for the southwest, indicates a shift to something like a 25% chance of above median, 35% of near median, and 40% of below median (the exact numbers vary somewhat regionally).  Thus, the seasonal outlooks issued by the National Weather Service in November predicted a small shift from chance toward drier conditions in the southwest, although they are often interpreted or communicated in ways that make them seem more confident.  

Long-lead-time forecasts, such as seasonal outlooks like the one above, are based largely on forecasts of more slowly varying components of the Earth's climate system that exhibit some predictability (over chance) at long lead times.  For example, although the day-to-day weather cannot be reliably predicted after about two weeks, ocean temperatures in the tropical Pacific (e.g., La Niña or El Niño) exhibit some predictability at long lead times.  

The outlook above was based largely on the presence and anticipated persistence of La Niña for this winter.  Forecasts probabilities of above, near, or below median precipitation were based largely on what happened during past La Niña, with some consulting of computer models forecasts and expert judgment.  

This sounds straightforward, but the the Earth is a complicated place.  First, although we can anticipate La Niña with some skill, the forecasts aren't perfect or highly detailed.  During La Niña the central and/or eastern tropical Pacific cools to below average, but the strength and coverage of the cold water varies.  This in turn affects the distribution of thunderstorms and thunderstorm complexes in the tropical Pacific and in turn that  affects the atmospheric circulation in different ways.  

Then there is the inherent randomness of weather and weather systems.  Waves and other flow patterns in the atmosphere can be highly chaotic.  Think of a loose firehose.  You can see there are waves in the hose, and they tend to exhibit preferred wavelengths and amplitudes, but there's also a lot of randomness.  A firehose is relatively simple compared to the complex three-dimensional flows in the atmosphere where there are interactions between circulations near the surface and at upper levels and between the tropics, midlatitudes, and polar regions.  

The existence of La Niña and other slowly varying characteristics of the Earth's surface (e.g., Arctic ice coverage, etc.) can affect some characteristics of the large-scale flow pattern, but there's still a lot of random stuff happening.  How the atmosphere got the way it did during the 2022/23 winter is complicated, much like Harry Potter finding the sword of Godric Gryffindor.

With that in mind, let's look at the atmospheric circulation over the past few months.  To do this we will use analyses of 500-mb height anomalies.  The 500-mb level is often used to describe the upper-level flow pattern.  Areas with positive height anomalies are places where there were anomalously high heights and ridging at upper levels. Areas with negative height anomalies are places where there were anomalously low heights and troughing at upper levels.

The 90-day height anomaly for October 4 to January 1 shows an anomalous tough over northeast Siberia (cool shading), a monster anomalous ridge over the North Pacific centered near the Aleutian Islands (warm shading), and a weak anomalous trough over the interior western US.

Source: psl.noaa.gov

For November 4 to February 1, the location and amplitude of these shift some, but the general pattern remains the same.  Anomalous troughing remains over Siberia, ridging near the Aleutians, and troughing over the western US.  

Source: psl.noaa.gov

And for December 2 to March 1, the story remains largely the same. 

Source: psl.noaa.gov

And suppose we looked at the 30 day period ending in mid March (our current pattern still approximates this).  Apologies for the color pattern change, but lo and behold, anomalous troughing over northeast Siberia (perhaps shifted a bit northward), ridging near the Aleutians, troughing over the western United States (and Canada).

Source: psl.noaa.gov

So, this general pattern of a trough over northeast Siberia, a ridge over the north Pacific, and a trough over the western US has been locked in for the better part of the winter.  Overall, this pattern resembles something known as the negative phase of the Pacific-North American Pattern, or negative PNA, which is one of the leading patterns of atmospheric variability in the north Pacific and North America.  

Scientists have created an index to measure the strength of the PNA pattern.  Below is a graph illustrating 3-month averages of this index from 1950 through February 2023.  The PNA index tends to flop fro positive to negative.  This winter we have indeed been in a negative phase of the PNA pattern.  

All of this is well and good, but it is where I am going to get off this train and here's why.  First, the negative PNA pattern is associated with La Niña, so that makes some sense.  However, La Niña weights the dice toward a drier southwest, which didn't happen this season.  Something else is going on with the large-scale circulation and one index, like PNA, doesn't tell us enough.  There's a lot happening in the Earth's climate system globally that might shift the PNA pattern characteristics or perhaps randomness played an important roll this season.  People who are smarter than me and specialize in the large-scale circulation can perhaps answer these questions, but I can't.  

As Tom Petty sang, "baby even the losers get lucky sometimes" and boy did we get lucky this winter.  

Damn the Torpedoes.