Winter Forecast is unlocked: Here's how we determine it

Visit the Complete Guide to Winter 2016/17 for the Winter Forecast, tips to survive it and much more.

Monday, November 21, 2016, 9:05 PM -

Autumn has a reputation for wild weather, and this year has been no exception across Canada. 

So far this fall has featured summer-like heat in the Great Lakes, tropical downpours and devastating flooding in Atlantic Canada, early-season snowfall followed by record-breaking warmth in the Prairies and powerful Pacific storms in B.C. 

Though the story of this autumn is not yet over, The Weather Network’s meteorologists are already preparing for the next change of seasons. Here’s a look at some of the key questions that we have been asked as we have been developing our 2016-2017 Winter Forecast.

How will La Niña Affect Winter Across Canada?

La Niña has a reputation for focusing the coldest and most active winter weather across Western Canada with more variable and often milder weather from the Great Lakes to Atlantic Canada. 

However, one should not make assumptions about the upcoming winter simply based on the development of La Niña. No two La Niña events are alike and the current La Niña event is rather weak. 

So, while La Niña will be a factor in our winter pattern, it is just one of the numerous variables that we are looking at in developing our winter forecast. Water temperatures in the Gulf of Alaska and the potential for blocking patterns to develop in the atmosphere over the arctic are also key considerations in developing a winter forecast.

Will this Winter be a Repeat of Last Winter?

Global patterns are very different right now than they were this time last winter.  This makes it rather unlikely (though not impossible) that we will see a repeat of last winter’s weather patterns.

One of the key considerations in comparing last year to this year are ocean water temperatures, especially in Pacific Ocean near the Equator. The map on the left shows current ocean water temperature relative to normal (blue = cooler than normal and red = warmer than normal) while the map on the right shows ocean water temperatures this time last year.

The black rectangle highlights the region that we look at for La Niña vs. El Niño conditions. This time last year we were in the midst of one of the strongest El Niño events on record with considerably warmer than normal ocean water temperatures near the Equator and throughout the North Pacific Ocean (map above on the right). 

However, the reversal in the temperature pattern in parts of the Pacific Ocean should result in a different jet stream pattern across the Pacific Ocean and North America from what we saw last winter.

Is it Really Possible to Forecast the Weather Months in Advance?

Yes, but it’s important to understand that there are real differences between our day-to-day weather forecasts and these seasonal forecasts. 

Because of the nature of seasonal forecasting, we cannot provide the level of detail that you expect to see in a daily forecast. So if you want to know if it will snow on Christmas, or if you should make outdoor plans on Valentine’s Day, you’ll have to wait for those specifics. However, we can often get a very good idea of how the big picture of a season will look – the overall patterns, and how they relate to climatological normal – well in advance.

This type of seasonal forecasting can provide really useful information, and it’s a common approach in a lot of areas, even outside of weather. Take sports for example – at the beginning of a given season, it would be very difficult to predict the outcome of any single game far in advance. This would be like trying to make a daily forecast months ahead of time. But often we can predict with reasonable certainty which teams are likely to be better than average and to win more games than they lose, or vice versa.

You can think of a seasonal weather outlook in the same terms. It provides information on which areas are likely to be warmer more often than colder, or wetter more often than drier, in relation to a normal year. But of course for every rule there will be exceptions. An area expecting “Above Normal” temperatures for the season will still see cold days, just like a sports team having a great season will still take occasional losses.

How do you develop the Winter Forecast?

The Weather Network’s meteorologists take a three-pronged approach when developing seasonal forecasts. Each of these methods gives us information about how the upcoming season is likely to play out, and using all three in concert with each other gives us the best understanding of where the overall weather patterns are headed.

The first forecast method we use is the analysis of teleconnections – or global weather drivers. Earth’s atmosphere and oceans form one big interconnected system, and things that happen in one part of the system can have big influences elsewhere. The classic example of a teleconnection is the El Nino cycle. Water temperatures over the equatorial Pacific affect trade wind patterns, which influences the jet stream pattern over North America. If you enjoyed last year’s mild winter, you have a teleconnection to thank. 

El Nino is one example, but there are many others. Water temperatures in the Gulf of Alaska, or air pressure patterns over Greenland, or the tracks of eastern Pacific typhoons can all influence North America’s weather. Understanding these influences is a major part of developing a seasonal forecast. 

The second method we use is the analysis of analogues – or past years with similar characteristics to current conditions. This is based on the general principle that if something happened in the past, it is likely to happen again given similar circumstances.

Our forecasters look through over 60 years of past weather data, to identify years with similar global weather drivers to the current year. These are our analogues. Once we find a good set of analogues, we can run the tape forward to see how those winters played out, and see if there are any common patterns that we can identify. Essentially with the analogue method, we are using the past to predict the future.

The third method we use is the analysis of forecast models. Seasonal forecast models are similar to the ones that we use to make your daily forecasts, but they are specifically tuned to help identify long term patterns and trends. Weather agencies around the globe develop and run their own seasonal forecast models, and we take the results of all of these models into consideration when developing our forecasts. 

The models aren’t perfect, and they don’t always agree, but by using them together with the information we gain from the other two forecast methods, we can filter out the solutions that seem the most likely. When model forecasts show good consensus with the teleconnections and analogues, it gives us higher confidence that they are correct.