It has been amazing to me to see everyone from the Old Farmer's Almanac to Accuweather saying this winter is going to be amazingly cold in the NE United States and very mild in the west. Even some local climate experts seem bent on a mild west this winter. I have been crunching a lot of numbers and come to a considerably different conclusion. It is fairly likely that the SW could have a warmer than normal winter, but the northern half of the western US should be a much different story! My evidence will be outlined in the following paragraphs and will be strongly supported.

For the most part the period of record used for the analogs in this report are from the period 1950 - 2000. I felt that period gave a good representation from both the cold (1950 - 1975) climate regime and the warm (1976 - 2000) regime. When I refer to a negative AO winter, the criteria I used was at least 3 months of minus AO in the Nov - Mar period, and a below normal AO average for the winter. A couple of key assumptions are...a negative AO this winter, and a very negative, but sharply uptrending QBO this fall and winter.

First of all some of the tropical indices are highly favorable for a cold NW winter. The Outgoing longwave radiation (OLR), in the tropics between 160W and 160E longitude, was stronger than normal in both August and September. In general strong OLR in the late summer and fall supports troughing, and cool temperatures, over the Pacific NW in the winter months. This map is a composite of years that had an OLR anomaly of at least +5 in August and at least +2 in September. map0001 How interesting this shows the NW is actually the most likely place to not have a warmer than normal winter! This next map further refines the analog years by adding strong 200mb zonal wind index, in the month of September to the mix. map0002

The 200mb zonal index, from 165W to 110W, has been really exciting over the past month. The records for this index go back to 1978 and this September had the third highest average of any in the period of record! It is interesting to note this index has not recorded a positive anomaly, in the month of September since 1990. The next map is a composite of winters following the top five September's for high 200mb zonal index values. map0003

The Quasi - biennial oscillation is yet another tropical index that points to a cold west. The following shows winters following a year where the QBO was either very negative in the summer and fall or negative in the fall and at least the early part of the winter. I use this map to illustrate that negative QBO in itself does not strongly favor cold or warm conditions in any section of the country. map0004 On the other hand, the QBO can be a useful prediction tool, when one takes trend and the severity of the anomalies and trends into account! This next map is really fun, and shows winters that had an extremely low QBO going into fall and then a sharp rise during the late fall and early winter (which is likely this winter). The criteria for the years used…1950 - 1975 lowest month -15.00 or below and 1976 - 2000 lowest month -22.00 or below. For both periods the lowest month had to be July - October with a strong up trend by the end of the year. map0005 This next one shows just the years in the 1976 - 2000 period. map0006Seasons that featured a negative QBO and also had negative globally averaged angular momentum (GLAAM) going into the winter also proved to be cold in the NW. map0007 Speaking of GLAAM, it is VERY LOW this year. This shows the winters following years that had GLAAM of -1.40 or lower for at least one month in the second half of the year. map0008

Years with positive SOI values in September often lead to cold winters for much of the nation with a strong signal for a warm SE. map0009 Now for a look at just the years during the most recent warm climate regime. map0010 The picture becomes even more clear with positive SOI in both September and October (which looks all but certain now).map0011

Note: The events talked about in this section failed to materialize so they cannot be used as evidence to support a cold winter in the NW. I am going to leave it so the map numbering will not be thrown off. It would have been nice to see the cold October in Seattle and warm in NYC, but this doesn't take anything away from the other evidence presented here. Just for the record Seattle and NYC both ended up above normal for October. As for the observed weather part of things, there are many signs a cold winter could be in the works. One the surest signs of a cold winter in the NW is a combination of a cool September and a cool October. There are surprisingly few years where Seattle has averaged below 60.0 in September and below 52.0 in October, and it looks like that may happen this year. September ended up at 59.5 and the October goal is still within reach, although not looking as likely now. map0012 One very telling sign of a warm winter in the east is if New York City and other eastern locations have a warm October. In years where NYC averaged 60.0 or above in October, an extremely clear picture is painted of what parts of the country are likely to be cold. map0013 If you go a step further and look at years where Seattle had a cool Sept and Oct and NYC a warm Oct, here is what you get. map0014

People have been saying that a high number of Atlantic tropical storms favors cold winters in the NE. That, of course, implies warm in the west. That just is not true. It is true if you only look at years with an extreme number of storms, but there are not enough of those years to be a sufficient sample. If one sets the bar at 18 storms the results clearly come out cold NE and warm west. If the bar is lowered to 16, things start to change. That is because the year 1936 had 16 storms, and that winter was tremendously cold in the west. That begins to cast some doubt on which part of the nation gets cold winters following big storm years. I found it most interesting to look at years with 12 or more Atlantic storms. That is still a high number, and the picture becomes crystal clear of who gets cold winters following such seasons. The winner is the NW, and in a big way! This first map shows years with a high number of storms from 1935 - 1975 map0015 and the second is for 1976 - 2000. map0016 It is very telling how the temperature profile is almost exactly the same for both periods. It is also interesting to look at what happens when you only look at such years that also had a negative AO the following winter. I added this to show people from the NE, they still could have a cold winter also. map0017 This of course assumes a negative AO winter, which most people agree is very likely.

The recent dramatic drop in the PDO also needs to be mentioned. Few people thought the PDO had any real chance of going negative and yet the September reading came in at -0.46. Not too shabby at all! If you look at years that had both a negative PDO (somewhere between 0.00 and -0.95) and positive SOI in September, you see yet another indicator of a cold NW this winter. Perhaps even a cold SW, but I still consider that unlikely. map0018 One forecaster who I really respect thought to look at years that had a PDO near -0.50 in September after the index had dropped for four consecutive months (just like this year). Now hold onto your seats for this one! map0019.

Now for the AO. This is one index that most people agree will be negative this winter, and if it is, that indeed increases the chances the east will see a cold winter like many are saying. I am also pretty convinced the AO will be negative, but not certain. For one thing, the AO being positive in September is actually a bad sign for a negative AO winter, and this September had one of the most positive AO averages on record for that month. Another thing that is really odd is the fact we have only had one negative AO winter after positive SOI in September and October since 1970! That one winter was 2000 - 01. That fact could mean the chances of a negative AO winter are very low or it could be a sign we have seen a regime shift this decade. I say that because such seasons were quite common from 1950 - 1970. Anyway, we will assume a negative AO winter. Many will be surprised to learn that a negative AO can also be good for cold winters in the NW. This first map shows all negative AO winters from 1950 - 1975. map0020 Notice how there is no clear trend except for a cold signal in the extreme northern plains. This next one shows the negative AO winters from 1976 - 2000. map0021 While this map shows a clear bias for cold in the east, I would like to point out it merely shows normal for the NW. Also keep in mind three of those years feautred El Nino's, which we will not see this winter. Notice how the national picture changes when only neutral enso winters are considered. map0022 The picture really changes when you look at years that had a very minus but uptrending QBO during the fall and winter. map0023 These next two maps illustrate what a difference it can make when one looks at the SOI before a negative AO winter. This year will have a positive SOI in both September and October. This shows years prior to 1975 that had positive SOI in September and October and then negative AO during the winter. map0024 Since there were no years after 1975 to achieve that feat (with the exception of 2000) this next one shows the years that just had positive SOI in September. map0025 As you can see it is very important to use qualifiers when basing a forecast on what the AO will be.

I haven't talked much about the ENSO conditions, because they are close to zero and will not be the deciding factor for this winter's weather. In light of the recent high SOI, I think we will likely see a drop and may achieve very weak La Nina conditions, but for now we will assume neutral conditions somewhere between -0.5 and +0.5. Contrary to popular belief La Nada conditions do not favor a cold winter in any part of the country. There is a VERY WEAK signal for a cold winter across the northern half of the country, with SST's of 0 to -0.5, but we are basically left with a level playing field for all. There is just nothing to support the wild persistent ridging that so many are talking about over the west! Such winters are normally extremely variable across the nation. This composite shows winters that feautred 0 to -0.5 SST's. map0026 This one is for 0 to +0.5. map0027 You will notice I used 1978 - 79 and 1980 - 81 in both maps. This is due to the fact both of those were right at zero for the SST's. Besides, doing that actually favors a slightly cooler map for the NE quarter of the country, and has no effect on the NW.

One final thing I would like to touch on is a relatively rare ocurrence, at least during the warm climate regime. This was the first September to average below 60 degrees with a positive PNA since 1978! I later found out this was only true with the new PNA formula the CPC is using, but since they updated their historic tables with the new formula, it still holds true. Perhaps this new formula will have some redeeming quality yet. Anyway, here is a composite map of winters that followed cool Septembers with a positive PNA. map0028

I feel this report certainly casts some serious doubt on the contention that the Pacific NW is certain to have a warm winter like many are saying. The tropical indices in particular say, cold all the way! There is little if any doubt this winter will have some amazing twists and turns along the way. Many people, including myself have noticed there is a sort of disconnect between the atmospheric indices and the observed weather this year. Could this be a sign of a regime shift? There is also little doubt the atmosphere "thinks" we are in a La Nina right now. The GLAAM, OLR, 200mb zonal winds, and SOI are all in a La Nina state. We either have to assume a weak La Nina may still happen, or the atmosphere is some odd state that is above and beyond what the actual SST's would dictate. This could be a sign of a wonderful and exciting winter for much of the nation. Perhaps we will see some long standing records broken. We can all hope the records will be for cold and snow instead of warm and dull. This goes for the entire nation, not just the NW. Good luck to all!