There are a number of factors used in determining in seasonal forecasts. For the winter season I use the following variables:
The trend the past two months has been for very warm and above normal temperatures. September and especially October were very warm. October 2007 was the warmest ever for both Washington, DC and Philadelphia, PA. October has shown to be a very good premonition of what to expect in the winter months and much research has been done to back up this claim. To summarize, October saw a dominant Ridge over the Northwest Atlantic, leading to record warmth along the east coast. The north-central US saw plenty of storminess and severe weather outbreaks. The Northwest, Artic, Canada and Western Canada saw very cold conditions and much below temperatures in certain areas. September also saw above average temperatures for most of the US and very dry conditions in the Northeast and Southeast states.
The big player this winter will be the state of ENSO. The data is overwhelming that this will be a La Nina winter and one that will be in the moderate to strong phase. In fact, it is very possible that this could be one of the top three strongest La Nina's since 1950. The La Nina has really developed since July and temperatures have cooled considerably in the equatorial Pacific. Current Nino3.4 values are below -1 and closer to -1.5, already in the moderate phase. Of all the La Nina events since 1965, only the 1973/1974 event had a Nino3.4 value more negative than at this current date. That analog winter saw a snowless winter.
Sea Surface Temperatures can say a lot in determining the short and long term weather patterns. In the Atlantic there is a large area of above average sea-surface temperature anomalies between Nova-Scocia and Bermuda and stretching north and east through the central Atlantic, north of the Azores, and continuing eastward all the way to the coast of Portugal. This has persisted for the past month and has grown in area. The Pacific has an area of below average anomalies off the West Coast of the US and Canada from Oregon to British Columbia. This area streches north and west to the coast of Alaska, but the anomalies are weaker further north. Out in the Central Pacific between 30-45 N latitude and 150-180 W longitude, there is a large much above anomaly. This has grown in size and anomaly had gotten warmer during the first half of October. The past two weeks it has cooled off a bit, but is still very prevalent.
Discussing the climate patterns, let me first add that the latest PDO value for October was just put out and it is one of the most negative values of all time. Cool phases are often associated with warm winters in the East. The SSTA pattern in the Atlantic shows a strong signal for a positive NAO pattern. La Nina winters have a predominantly positive phase of the NAO and research has strongly supported it. They also have highest number of Alaksa Blocking episodes. This would support a trough over the western US and a negative PNA pattern. All the above would lead to a positive phase of the AO. This pattern set up does not bode well for a cold and snowy winter in the Mid-Atlantic.
Finally, the pattern trends for the first half of fall will have a bearing on the winter setup. This is not the case 100% of the time, however. Dry conditions have persisted over the Southeast with a large area under a severe and historic drought. Storm systems tend to stay away from dry areas and this will help to enhance ridging inland and off the Southeast coast. The storm track has taken a course from the central US to the Great Lakes and into Canada west of 70W longitude. This type of storm track sets up a scenario where ridging will persist off the East Coast, leading to warm weather ahead of storm systems and cold blasts behind them. The Pacific Northwest and Western Canada have seen numerous storms slamming into the coast, indicative of an active Pacific Jet Stream.