Possible Trend Towards Later Winters?

CO-OPERATOR CONTRIBUTOR

The No. 1 question or comment I hear at this time of year is either when are we going to get snow that will stay? or I hear the normal time to get snow is November? you fill in the blank, as everyone seems to have a different date! Each year, I re-examine this topic trying to dig a little deeper. This year, I m going to take a look at the dates for the start of winter in Winnipeg to see if there is any noticeable trend. I ve overheard a number of comments the past few years and it seems that the general perception out there is that winter is starting later than it once did.

Read Also

The lowdown on winter storms on the Prairies

It takes more than just a trough of low pressure to develop an Alberta Clipper or Colorado Low, which are the biggest winter storms in Manitoba. It also takes humidity, temperature changes and a host of other variables coming into play.

As with most things associated with the weather, it looks to be a fairly simple topic. But as you start to look at it a little deeper, it becomes more and more complicated.

As we ve been learning over the years, certain weather-related questions sound simple enough, but when you actually start to look at the question, it becomes tougher to figure out. The tough part about trying to figure out when winter starts is how to define just what constitutes the start of winter. Should it be the first significant snowfall? How about when the high temperature consistently stays below 0 C, or should we simply use the astronomical date of Dec. 21 or maybe just stick to the meteorological date of Dec. 1?

I think most people would agree that winter doesn t really arrive until you have snow on the ground, so for us, I use this as our measure of when winter arrives. Even narrowing it down to this still has some problems. Take this year for example. A good portion of agricultural Manitoba received snow on Nov. 6. Areas that received more than about five cm still have snow on the ground, while areas that received less than five cm have seen pretty much all of it melt away. So now the question is, did winter arrive on Nov. 6 or should we call this a false start to winter? For those areas that received snow and it continues to stick around, I would say that winter started. If the snow melted and is gone for more than a couple of days, I would say that it was a false start to winter and that the true start date won t happen until it snows again and then sticks around.

With this in mind, I re-went through the snowfall records for Winnipeg, Brandon, and Dauphin, and came up with the following results for when we should expect winter to start.

From this table we can see that all three regions of agricultural Manitoba have seen winter start in October and as late as mid-December. Winnipeg and Brandon both have an average date for snow to stick around of November 14, with Dauphin being four days earlier at November 10. The usual range is a measure of the standard deviation around the average. It indicates the range of days that we should expect winter to begin. If winter begins before or after these dates, it is a very unusual year.

These numbers help us to know when we should usually expect winter to arrive. How about the question as to whether winter is trending towards arriving later? Now, before people start getting mad at me about just using Winnipeg s data, let me just point out that due to the complex nature of determining the start date of winter, I pretty much have to go through the weather records manually which takes a lot of time. So for now I will stick to Winnipeg. But later, when I have more free time, I ll crunch the numbers for Brandon and see what comes up.

I ve included a graph of the data for you to look at. The graph shows the five-year moving average of temperature for November from 1980 to 2010 on the left axis and the date of the start of the permanent snow cover on the right-side axis.

Using a five-year moving average helps to remove some of the year-to-year variation that we would naturally see in the data. The first thing that we can pull from the graph is the fact that years with colder Novembers tend to see an earlier development of permanent snow cover. Since the two really go hand in hand this is not surprising. What is a little surprising is what appears to be a switch towards warmer Novembers, and a later start to permanent snow cover starting in the mid-to late 1990s. It is a little early to say that this is an actual trend in the data, but interestingly, this coincides with the start in the fairly rapid decrease we ve witnessed in the amount of Arctic sea ice; we ll have more on that in upcoming issues. Until then, let s hope that we get just the right amount of snow at just the right time!

———

Earliest StartLatest

Average Usual Range

Winnipeg Oct. 27, 1972

Dec. 16, 1939 Nov. 14

Nov. 3-26 Brandon

Oct. 6, 1959

Dec. 15, 1974 Nov. 14

Nov. 2-27 Dauphin

Oct. 6, 1959

Dec. 14, 1981 Nov. 10

Oct. 29-Nov. 22