Today, weather in southern Manitoba is smashing historical norms. Environment Canada predicts seven degrees this afternoon for Winnipeg, which is unheard of in early January. But how unusual is it?
The previous record for this day is 4.3 degrees, while the average maximum temperature for this time of year is minus 13 degrees. It is 20 degrees above normal. Such an extreme anomaly is highly unlikely to arise by chance. According to Environment Canada data for Winnipeg weather 1970-2000, the standard deviation for January 5 (a measure of the distribution of probable temperatures) is 8 degrees. Today’s temperature is fully 2.5 standard deviations from normal.
Events of such as this have a probability of 6/1000 of occurring by chance.
NASA’s climate scientist James Hansen writes how with global warming, it’s like loading the dice for our typical weather. Winnipeg keeps rolling snake eyes.
*This blog post includes a correction of a statistical error spotted by a Green Action Centre member.
Thwnks for a marvelous posting! I genuinely
enjoyed reading it, you’re a great author. I will make certain to bookmark your
blog and may come back in the foreseeable future. I want to encourage continue yohr great work, have a
nice evening!
I can tell you that for the period 2007-11, the average January daily maximum temperature (WPG Richardson) was -11.95C and the standard deviation was 7.45.
Environment Canada supplies historical temperatures in csv (comma separated value) files. You can download those files for a standard WMO period, such as 1971 to 2000, and calculate the standard deviation of the daily maximum for January. However, the csv files also contain 10 other weather-related data, such as precipitation, ground snow cover, heat days and maximum wind gust, so it is a bit of a chore to extract just the data you need. Environment Canada may also compile and provide the specific information you require upon request.
Once the data is obtained you can calculate the standard daily maximum average and deviation. Easier still, copy the maximum values (either for each day or the entire months worth) into a spreadsheet and let it calculate the standard deviation for you. If you do it by day, you can derive both an average maximum and deviation for each day of the month (which will probably not vary to any significant degree, other than a relative cooling trend toward the latter part of the month).
It’s important to remember that statistical tools are great for showing anomalies and outliers, (Winnipeg’s mean December temperature in 2011was the 3rd highest, behind 1997 and 1959, and Edmonton’s was third behind 1999 and 1997) but in climate (and weather) it is the long-term trend that is of more value.
Thanks b_nichol, that helps clarify things. Do you have any idea what the daily standard deviation is for January?
Someone has made a grievous statistical error:
1) The standard deviation of 3.9 that Environment Canada shows is for mean temperature, not maximum
2) The deviation is for monthly mean, not daily
3) If the forecast high of 7C is one in 1.7 million, where does that place the all-time January high recorded in 1942 at 7.8C?
)look for Wpg Richardson at http://www.climate.weatheroffice.gc.ca/climate_normals/index_e.html)
Warm indeed, but hardly the 11.7 we recorded in Edmonton on Wednesday afternoon (average January daily max -7.3C, 1971-2000)
Thanks, please note the correction.