How to Create a Temperature Graph

In another post I discussed how Environment Canada temperature records show climate change happening right here in Ottawa. This post will describe how you can reproduce the graphs I showed there for any weather station across Canada; plus, what that means.

We’ll start by going to Environment Canada’s Weather Office website. Once there navigate by map or by pull-down list to the weather station you’re interested in, in this case Ottawa (Kanata – Orleans).

In the lower right corner next to “Historical Data” click on “More Info +” then select “Historical Weather.”

In the lower right corner of the page that appears, in small type, is the link “Customization Chart.” Click that.

What will be displayed is more than likely to be the high and low temperature records that have happened over the days of whatever month you’re in right now. That’s nice, but it isn’t enough. We need more.

Look down at the bottom of that page and you’ll see the headings “Display Items” and “Display Options.” (these are circled in the image at left, but you have to click to enlarge to see them)

To reproduce something similar to what I’ve done for 2011 in Ottawa tick all the tick-boxes under “Display Items” and choose “Display Size: Large” and “Display Period: Twelve Months.”

But let’s examine what all those tick boxes give you.

The graph at the right here shows the smooth temperature curves of what historical “normal” temperatures have been between December 1 of one year to December 1 of the following year. (The weirdness of this not being January 1 is due to Ottawa’s weather station data collection breaking in December 2011. Other cities will be able to plot more logical time periods.) This smooth graph is achieved by ticking only the two tick-boxes “Normal Maximum Temperature” and “Normal Minimum Temperature” under “Display Items.”

This “normal” is based on the 30 year average of temperatures for any given day from the past records for this weather station between 1971 and 2000. So that fact that more recent temperatures are a little scary is made even more scary when you realize that what’s shown here as “normal” has already been adjusted because it no longer represents what was normal in decades before 1971. Environment Canada follows the advice of the World  Meteorological Organization and updates its normal every decade to refresh the 30 year normal. The period ending in 2000 was an update from the earlier normal ending in 1990, which in turn replaced the one ending in 1960. You can read all about that here.

The next layer we’ll add to the graph are the “Extreme Maximum Temperature” and “Extreme Minimum Temperature” records, also by ticking their appropriate tick-boxes under “Display Items.”  These are how hot it has ever gotten on any given day and how cold it has ever gotten on any given night according to records collected between 1938 and 2011 (and maybe more recently if the weather station you’re looking at is still working properly).

This is a more jagged line because any given day isn’t an average, but a one shot record temperature.

Finally we can lay the actual temperatures from the last year on the graph. This messes things up considerably. Each day of the year now has the actual daily high and low to compare with normal and extremes.

The chart is too crowded to be able to easily count how many times the daily high met or exceeded the record which is what I did here. The way around this is to choose shorter time periods under “Display Options” > “Display Period.” By choosing quarterly or monthly you can see more clearly and count how many times the temperature pushed the record high even higher.

How do you recognize this? When there is no data point for the extreme high, but instead the data line joins the daily high that’s when the high that day was equal to or greater than any temperature before recorded for that day. The flipside logic for extreme lows. That’s how I got the 10 to 0 ratio for Ottawa and 14 to 0 ratio for Montreal.

The fact that there are fewer extreme lows is in itself another indicator of global warming. Temperatures fall at night mainly because no heat is being added by the sun and so the world is staring out at the cold of space and radiating heat energy into that cold. Carbon dioxide acts as an insulating blanket against the cold of space and so it is particularly at night, since we now have more blankets on, we see less cooling.

If you are really really keen you can even count the daily highs above and below what should be “normal.” Do it again for lows and minimums and the combination will give you your local equivalent to Al Gore’s slide which says that across the US instead of having 1:1 variation from normal, the decade up to 2001 had 2:1 hot to cold, and in 2012 up to July had 10:1 hot to cold.

This entry was posted in Updates. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *