It turns out there are a lot of issues with this claim if you really examine the facts. Lets do it!
The story was based on the fact that the National Weather Service has just changed the period used to calculate our climate--what they call the "normals." A thirty year period is used and each decade they shift it forward to cover the last thirty years. The official climate normals were 1970-2000 and are now 1980-2010.
The folks at the Seattle Times presented and discussed the results from Seattle (t is important to note is only ONE location in the region and might well not be representative!). Here are their results, showing the normals from the last three versions of the climate normals (1961-1990, 1971-2000, 1981-2010).
Lets look at precipitation first. Precipitation at Seattle dropped from the first normal to the second, and then increased for the new one. Seattle is getting wetter? Not so fast. First consider the differences between these values--only a few tenths of inch separate them and I am pretty sure these differences would not be statistically significant.
But there is something else. It is well-known fact among local meteorologists that something was wrong with the Sea-Tac rain gauge during the 80s and early 90s--the "SeaTac anomaly" in which the gauge reported too low, compared to gauges in the neighborhood. Recently it was fixed. This anomaly would clearly contribute to the trend.
Now if we can't be sure about Sea-Tac, what about the new normals for the other climatological stations in the region? Here are the annual normals for the last three versions for Hoquiam (hqm), Olympia (olm),Bellingham (bli), and Spokane (geg).
Completely different story! The annual precipitation has gone down! Stop the presses! New headline---the Northwest is getting drier! But I wouldn't say that either...these differences are surely not statistically significant, meaning they are small enough that they could be due to random variations that are not connected with any true trend.
What about the warming? Could there be a problem with that too? It turns out that if you look at other stations, one gets a mixed bag, some are cooling--like Hoquiam(hqm) and Bellingham(bli)-- and others show slight warming--like Spokane (geg). Here are some samples.
So what about the rising temperatures at Sea-Tac Airport? I suspect it is probably bogus due to sensor issues, the development around the airport, and the addition of the third runway. A careful comparison high temperatures at Sea-Tac and the temperatures of surrounding stations indicated a several warm degree anomaly at the airport (done by Mark Albright, past state climatologist). For example, the first 8 days of July, Sea-Tac was 2.3F warmer than two nearby stations.
The truth would not make a good headline: NO APPARENT TREND IN NORTHWEST CLIMATE DURING THE PAST SEVERAL DECADES.
The bottom line: be very skeptical about dramatic headlines about climate change--when you dig a bit, you may find the truth is more complicated.
Classic example of data dredging. Writing the story first and looking for the evidence later.
ReplyDeleteI don't live in Seattle but I happened to buy a Times in Moses Lake on Friday...I knew as soon as I saw the headline that you would probably be writing about it here! Thanks for doing that.
ReplyDeleteI have a couple of questions...first of all, it seems really weird to me that the NWS would pick a thirty-year block of time. Do you know why they chose that particular time period?
Also, if there are this many issues with modern equipment (inaccurate rain gauge, etc.) then can we place any faith in measurements that were taken a hundred years ago?
Excellent analysis, thanks for the additional details Cliff.
ReplyDeleteCliff,
ReplyDeleteIn my layman's point-of-view, I have seen this as a constant problem with official readings for major metropolitan areas. They put guages in concrete-dense areas (with the associated heat island effects) where I have noticed the impact on the weather which would leave these guages suspect.
Using Austin for example (as I am from there). If you watch many of the storms that come through the area, they may drench the surrounding hill country but as they approach Austin, the storms tend to fall apart or just as common split and go around the city. This has been attributed to Town Lake/Lake Travis and effects of the city (pollution/concrete/heat reflecting back up into the atmosphere).
As Austin is used as the major guage for the region, the numbers would seem to be skewed and not represent the "reality" of that particular day.
If this is the case, why put the guages in these places (other than secure access to the guages and a likely permanent location as airport tend to last a long time). And why rely on these locations vs other "better" locations?
Thanks, Cliff. Very nice!
ReplyDeleteWow, cool! I thought the article in the Seattle times was a bit incomplete. It didn't seem like they had enough data to make any real conclusions, as you have shown.
ReplyDelete"when you dig a bit, you may find the truth is more complicated."
ReplyDeleteI think people at both ends of the debate need to understand that.
Odd that the reporter didn't ask the people he interviewed whether his headline was right in a significant way.
ReplyDeleteSeems like a major oversight, if the story is supposed to inform the public.
I'm not a fan of the antepenultimate paragraph, either, what does this mean?
There's no simple explanation for the overall changes in Seattle or nationally — or even a consensus as to whether they're a big deal, given the often polarizing debate surrounding global climate change.
I think there's a strong consensus about some of the changes, and if I'm confused by the article, how can the general public unravel it?
That's the Times -- decide what the story is without much (or any) reporting, and write it anyway.
ReplyDeleteI signed on to ask the very same question that raja did...
ReplyDeleteSo, is there any practical solution, given the desire to have measurements that are both comparable to historical measurements AND representative of the region?
These airport readings should not be used to establish trends. I don't understand why, with all of our remote-sensing technology, we don't establish some more remote, highly accurate, highly reliable weather stations. Even better, make them gridded or at least separated by relatively small distances, so we can get some useful small-scale data. The RAWS and Sno-Tel sites, and others like them, are there, but they are not really that accurate and are separated by large distances.
ReplyDeleteCliff - By providing an honest scientific perspective in the global warming debate contrasted the alarmist misleading hype that pervades so many reports, you may start to reverse the credibility gap which makes us skeptics suspicious of the whole thing.
ReplyDeleteRight on, right on. It's stuff like this that makes me love this blog. Great analysis. Now somebody crunch the numbers and figure out if there is actually any statistically significant difference.
ReplyDeleteI've read this sentence 3 or 4 times. Did you mean to type cooling in this sentence?
ReplyDeleteIt turns out that if you look at other stations, one gets a mixed bag, some are warming --like Hoquiam(hqm) and Bellingham(bli)-
Bangfrog..yes that is a typo..thanks so much..fixed..cliff
ReplyDeleteCliff,
ReplyDeleteYou are getting a little hung up on this 30-year business.
While 10-year periods are too short for solid mathematical analysis, the most profound impacts of climate change have only been happening in the last decade.
I realize it's fun to be a bit of a contrarian, but really, we are facing extremely severe problems here.
How about getting into the (less comfortable) mathematics of short-term changes, particularly the occurrence of severe events?
Think of it this way: the new 30-year normals don't tell us where we are today. They tell us where we were 15 years ago.
Nasa Earth Observatory "Image of the Day" (July 6, 2011) recently had a piece on "U.S. Climate: The New Normal" showing two U.S. temperature comparisons: 1) January minimum temperatures from 1971-2000 and 1981-2010, and 2) July maximum temperatures from 1971-2000 and 1981-2010. (http://earthobservatory.nasa.gov/IOTD/view.php?id=51270)
ReplyDeleteUnlike the Seattle Times headline, I don't think the NASA image is presneted in a sensationalist manner, but it does suggest that the extreme temperature events (e.g., minimum Jan temps and July max temps) in the eastern pacfic are getting warmer and not cooler as you indicated.
When you say "the eastern Pacific has not warmed--in fact it has cooled", what area are you defining as the eastern Pacific? Also, are you basing your statement on annual averages? Decadal averages? Etc...
Not trying to be a wise guy, because I think the Seattle Times could have done a much better job, but I think it is somewhat contradictory to criticize the Seattle Times and then make a statement like "the eastern Pacific has not warmed--in fact it has cooled" without providing a sufficient amount of data.
itodd,
ReplyDeleteThose figures did not show the eastern Pacific ocean. If you look at the trends of the SST or air temperatures over the eastern Pacific, you will find no warming, in fact, cooling over the past several decades. I have examined this myself and published some graphics on this in my snowpack paper:
http://www.atmos.washington.edu/~cliff/Snowpack.pdf
check it out and see what you think..cliff mass
[not necessarily on topic please move or delete at will -- general question not written especially well, sorry]
ReplyDeleteOn an otherwise long since forgotten television program I saw abrupt turbulence pattern changes within a fluid between two spinning concentric cylinders used to illustrate potential concerns about prospective anthropogenic atmospheric changes. The increased velocity difference between the cylinders was said to be analogous to the increased amount of greenhouse gas and heating of the earth's atmosphere.
Ever since then, I've wondered about the stability of the larger atmospheric circulation patterns described by Hadley cells, Ferrell cells, and polar cells. Most sources I've encountered do not even appear to question these patterns, they are a given, like the existence of rocks or water or the atmosphere itself. Yet there are other patterns within our solar system -- Jupiter's atmosphere being a spectacular example.
so, my questions: Can we rely on the general atmospheric patterns remaining largely the same (save for perhaps a few extra degrees of shift one way or the other) throughout our lifetimes and those of numerous generations to come? Or is it possible that we may see abrupt transitions to more or fewer cells or instability so great that these latitudinal bands no longer adequately describe the broader atmospheric circulation patterns?
This comment has been removed by the author.
ReplyDeleteWhy do we keep insisting on thinking anything is "normal", especially in regards to weather? It seems like that kind of thinking only creates more stress in a universe where change is constant. It inhibits adaptation and by extension our evolution.
ReplyDeleteAnyhoo...just a though.