I have performed a correlation of the temperatures measured between 1932 and
2000 to see if there is a linear trend of 'warming' at Churchill, Manitoba.
I took the data set from the site located at the Goddard Institute URL
www.giss.nasa.gov
Since I entered in the temperatures as absolute values the slope is negative
which means that the as the years progress towards the year 2000 the
temperatures warm up a bit. I have the correlations to use as required.
Okay for January since 1932
Temp_Jan= - 63.707 + 0.0187*Year
correlation: r= 0.1125
If the trend continues we can expect to see temperatures of -24.4 celsius in
January, 2100 a rise from -26.3 C expected in January 2001. The estimated
increase is not great for January. The predicted rise is 1.9 degree celsius.
This means that the temperature is becoming warmer each year during the
month of January. There is definately warming here. The coldest January
recorded temperature was 36 degrees celsius was in 1950, and the warmest was
1981 and in 1940
For February
Temp_Feb=-131.75 + 0.0540*Year
Correlation: r=0.3090
In 2100 the temperature will rise to an estimated value of -18.4 degrees
celsius from a predicted value of -23.8 celsius in 2000. That represents a
rise of 5.4 degrees celsius over the next one hundred years for Churchill,
Manitoba during February.
The warmest two Februaries were in 1998 and in 1999 [-18 degrees]. The
coldest two Februaries were in 1979 and in 1939 [-33 and -32].
The data for Churchill indicate that over the last 68 years the temperatures
during the months of January and February are getting warmer.
Test performed with Statistica (TM)
chao,
john foster
|