Ever since Nate Silver hired the Global Warming denialist Roger Pielke over at the new fivethirtyeight.com website, I have to admit that I've been a bit curious as to what kind of articles we might find there. Nate's got a very nice website going on over there, he's got areas like economics, sports, life... but I'm especially interested in the "science" section. Apparently Nate shares my interest in Weather/Climate since about half the articles under "science" deal with Weather/Climate.
Some of the most recent science in the study of weather and climate involves the link between weather events and Global Warming. Over the last few weeks we've seen some pretty strong warnings from science based organizations like the IPCC and the American Assn. for the Advancement of Science about the grave impacts that Global Warming is about to have on our civilization.
But Nate Silver takes a different approach, a kind of happy-go-lucky approach to weather and climate. While the rest of the world frets over rising oceans, super droughts, melting ice caps... Nate's website is kind of reassuring, calming, there's a silver lining around that dark mesoscale supercell coming right at us.
Here's my very brief take on those first articles, before we get to the latest.
Nate started out his "science" section with a piece from Roger Pielke called Disasters Cost More Than Ever — But Not Because of Climate Change. This was an interesting approach that says sure, all those super storms and hurricanes, and those massive California and Texas droughts cost more, but not because of Global Warming, but because we're a richer country now. I gotta admit I really didn't think of it like that. Isn't that nice now? We're all richer so we all should feel better.
Then there was that kind of Noah's Ark article titled Can Evolution Outrace Climate Change? Isn't that a nice ray of hope to know that some species will evolve around Climate Change? Now I might be a little cautious, but I think I would put a disclaimer in a story like that, in several places, and it might say something like;
FOR EVERY FREAKEN SPECIES THAT EVOLVES TO CLIMATE CHANGE, HUNDREDS MORE ARE GOING TO GO EXTINCT!!!
Then there was the nice story about El nino and how it's responsible for variations in our weather. And then the story about how the annual number of tornadoes has leveled off. But nowhere in either of these stories is the word "climate" even mentioned. To talk about weather patterns like tornado seasons or droughts and floods related to El Nino without mentioning the latest climate science, is like me giving you the street addresses of some fantastic museums and restaurants without mentioning what cities they were in.
Nate put up another calming, reassuring piece last week titled When April Snow Showers Blanket Spring Flowers.
It’s the middle of April — and it’s snowing as I write this in New York City. This isn’t quite as unusual as you might have gathered from the reaction on Twitter. Based on data collected in Central Park dating to the 1860s, 64 percent of New York Aprils have experienced at least some snow, although it usually comes earlier in the month.
So Mr. Silver averaged out about 140 years worth of data and came out with this very nice chart.
I'm not a statistician, but I do like to play around with numbers. Now normally, if you're trying to look at patterns over time, you want to go back as long as you can to get the biggest data set, but not always. Think about baseball for a second. If I averaged all the home runs hit in April since the late 1800's, would that give me a pretty good idea of how many home runs were hit during any period of time? Not Really. In the 1800's and early 1900's the game was played a lot differently, the rules were a little different, the equipment was different... And then there was that steroid thing in the past 20-30 years.
So it got me wondering, is there a steroid type thing that might have an effect on weather patterns?
Yep! Sure looks like we've got weather on steroids lately.
So I thought to myself, what if I just download that weather data and look at it from a steroid perspective.
This is the raw data graph of all the snowfalls in the month of April for New York. The line going horizontally in the graph is the trend. And if you look real close, and compare the number and depth of snowfalls in the early years (left half) with those in the later years (right half) you can see there's a difference.
One way to kind of smooth out the data a bit and see the trends a little better is to take a moving average of the data. So each of the following points is just the average of the previous 10 years.
Now you can see the trend much better, and you can see the number and depth of snowfalls really kind of falls off in the last 30 years.
Since the trend is rather obvious, I thought I'd divide the whole 144 year period in 3 parts and look at each part.
This first graph goes from 1870-1918. Take note of how many and how deep the snowfalls are.
This period is from 1919-1966. Again look at the number and depth.
And the final period is from 1967-2013. Notice the big difference in number and depth of April storms since Global Warming started to really impact our weather? In fact, notice there are only 2 Aprils in the last 45 years where it snowed more than 2", compared to 10 and 9 in the previous 2 periods.
I'm sorry that the numbers on the charts aren't very readable, I'm still trying to figure that out, so I'll do a little basic arithmetic and build a little chart to see if it makes it clearer.
There were 42 "measurable" snow events in April in New York's Central Park since 1870, with a total of 117 inches of snow.
Time Period Snow Amount % of Total Number of Events % of Total
1870-1918 62.29 53.2% 19 45%
1919-1967 39.33 33.6% 15 36%
1968-2013 17.89 15.3% 9 21%
Well there's your problem Nate! I really doubt if many of those people who were tweeting about how weird the April snow was, were around in the 1870-1918 period when most of the April snows fell during the 144 year data set, and where almost half of the measurable events took place. I doubt if many were around to remember the events before 1967 either.
What a strange coincidence that the drop off in the number and depth of April snow in NY kinda coincide's with the rising world temperatures. It's a bit amazing how if you look at some numbers in a different way you might reach a different conclusion. I guess we better not ignore the effects of climate change on weather patterns, any more than we should ignore the effects of steroids in baseball on home runs. We certainly don't want to be giving people the wrong impression.
UPDATE: Several comments below point out that I did not use good scientific statistical methods in presenting this data. They are absolutely right. I really did not intend for this diary to get published in a scientific journal. I used what I thought were good graphs to visualize how different the snow events of the past 48 years are compared to the previous 90+ years, and it's intended to be more of a Mythbusters approach than a hard science approach. You would never take this approach in any type of a science paper. So I need to issue a warning here. If there are any scientists reading this, DON'T try this at home! And if you try to use the above approach in a science paper, well, you kind of deserve what you are about to get.