Several months ago I received a question about winter temperatures and humidity levels along with another question about overnight low temperatures. In particular, about whether or not there has been a trend towards warmer overnight lows. While these two questions are not totally connected I began looking into the numbers and here is what I have come up with so far.
To start off, looking into humidity levels has turned out to be tougher than I originally thought. The problem is that Environment Canada’s weather stations, or rather the data collected from these weather stations and made available to the public, does not contain humidity data. This makes it particularly difficult to dig into this part of the question. There are sources of data that I am working on, so I hope to be able to continue to look into this question both for the winter and summer months.
The second question as to whether or not we have seen any trends in overnight temperatures across the Prairies was a little easier to work with. According to the climate models, we should be seeing a trend towards warmer overnight lows, so the question is just how to try and figure out whether or not we are seeing any significant trends in the data. This question does, in my opinion, also tie into the question about humidity levels.
As most of us already know, the drier the air the quicker it will typically cool down at night. We can often see this in the spring when dry air moves in and there is no active plant growth to add humidity to the air. Daytime highs will often get very warm as all the energy from the sun goes into heating the soil and the air, instead of warming the water in the air. Conversely, at night, the air will then quickly cool down as there is little heat stored in the air due to the lack of water. Therefore, if there is a trend towards warmer overnight lows it could be partly due to an increase in moisture.
With this in mind, I looked at the data for the main weather stations I use to compare weather across the Prairies. I then had to pick a time frame to use and with Peace River having the shortest dataset (1944-2017) I chose to use this 74-year period for this study. I then looked for any missing data, and while there were a few days missing here and there across all the stations, overall, the data quality was very good. I then calculated the average monthly minimum temperature for each month in the dataset for each station. This data was then plotted and a linear trend line was added. Finally, the trend line was analyzed to determine how many degrees of change has occurred over the 74 years of data. Trying to analyze all the months of data was becoming time consuming, so I selected four months to analyze, one for each season (January, April, July, and October).
I began my analysis in Alberta and was quickly surprised by what I found. Starting in January I found that at all three locations in Alberta, there was a significant warming trend in January overnight lows. Peace River had the largest change, with a trend line increase of about 9 C over this period. Both Edmonton and Calgary also had fairly large increases, with each station reporting a trend line increase of 6 C. When I saw this I had to check to see the other stations across the Prairies to see if they had the same large increases. Looking at Regina and Saskatoon in Saskatchewan, along with Dauphin and Winnipeg in Manitoba, I found that they also had increases in their January overnight lows, but to a little less extreme, with changes ranging from 3 C in Winnipeg to 5 C in Dauphin and Saskatoon.
With these fairly dramatic results I then looked at spring temperatures using April’s data. While there was a general warming of overnight lows across the Prairies the amounts were not as dramatic as January. Across the three locations in Alberta I found a warming trend of about 2 C over the 74-year period. In Saskatchewan, both Regina and Saskatoon saw about 1 C of warming and in Manitoba this fell to only about 0.5 C of warming.
This trend in overnight lows continued into the summer months. When I analyzed July’s numbers I found that across Alberta the warming trend was a little less than spring, with a trend line value of about 1.5 C. In Saskatchewan and Manitoba during July, there has been no overall change in overnight temperatures. Continuing on into the fall using October as the representative month, I found that across Alberta there was no overall change in overnight lows according to the trend line, with the exception of Edmonton, which saw a 1 C warming. In Saskatchewan and Manitoba there appears to be an actual cooling, with the Saskatchewan stations seeing a trend of about -0.5 C across the 74 years of data and Manitoba seeing a trend of about -1 C.
While this might not be the most scientific study, I think it does show that we are definitely seeing a warming of overnight lows in the winter and to a lesser extent in the spring, across the Prairies. This trend does not appear to continue into the summer or fall with even a small amount of cooling occurring in the fall. I will continue to explore these and other possible trends in the upcoming months.
Daniel Bezte is the weather columnist for the Manitoba Co-operator.