The vertical bars on the chart below indicate daily temperature deviations. If the actual average temperature was warmer than the historic average for that date, the bar projects upward indicating a positive temperature difference, and vice versa for colder. You can count the bars on the chart, or you can trust me that there were 167 days warmer than average, 178 days colder than average, and 20 days in which the deviation from the historic average was zero degrees. By that measure, there were more days colder than average than days that were warmer than average. But since a day that was 1 degree colder than average counts as much as a day that was 30 degrees warmer than average, does that really tell the whole story?
Another way to look at the data is to look at the cumulative degrees above or below average. That data is reflected in the purple curve that wanders above and below the 0°F line in the chart above. As you can tell, the curve ends the year on the right side of the chart in positive territory -- 42 degrees above average to be exact. Dividing that number by 365 days tells us that each day in 2009 was, on average, 0.12 degrees above the historic average. That may be a small enough number to be statistically insignificant, or at the very least, indiscernible for most people.So, was 2009 colder than the historic average? That, like most conclusions about weather trends these days, depends on which criteria you use to make the decision. Thanks for visiting, and thanks to Weather Underground for the historic data.
As one who most usually is "satisfied with subjective feelings in matters like this," I am amused... it honestly would never occur to me to analyze all of this information like this! And I suspect it would never occur to you not to analyze the data;-)
ReplyDeletehm.... a very interesting ending to a good series of posts!
ReplyDeleteSo, on average - the temperature was warmer - but there were more days in which the temperature was colder than average.
The last paragraph has me a little confused though.... Yes, your conclusion will be determined by the criteria you choose for the data set/method. But at the same time, your choice of a set of criteria should be determined by what factual results you desire.
I'm not quite sure if I explained what I meant sufficiently, but what I'm trying to say is that there is objective truth - even in the area of whether or not it was colder on average in 2009. :-)
Diane -- I guess it takes all kinds...
ReplyDeleteEmil -- "ending?"
I also agree that there is objective truth, even in interpreting weather data, but the data I've shown could be used for opposite purposes. For instance, if you wanted to prove that the weather was colder than normal in 2009 in Peoria, you simply mention that there were more days colder than average than there were days warmer than average. Conversely, if you wished to prove that 2009 was warmer than average, you would simply look at the cumulative degrees above/below average. Both approaches would be presenting the data factually, but arguments could be made about which data better represents what you're trying to measure.
If one were trying to prove that Peoria's temperature is increasing, one would need many years of data, would want to measure temperature at several locations, and would probably want to evaluate continuous temperature data.
Ok... you got me... I thought the 'look back' in the title meant it was a sort of epilogue to the previous posts, and would therefore be an ending. :D I guess not!
ReplyDeleteI think I completely agree with all that you wrote there. (least ways - I think so... or it might be a little too early in the day for me to think critically)