Nate Silver's New Site Stirs Climate Controversy
First time accepted submitter taiwanjohn (103839) writes "One of the first articles on Nate Silver's highly anticipated data-driven news site used flawed data to make its conclusions, according to some of the nation's top climate scientists. Silver's FiveThirtyEight published its first article about climate change on Wednesday, entitled 'Disasters Cost More Than Ever — But Not Because of Climate Change.' But climate scientists are condemning the article and its author, Roger Pielke Jr., saying he ignored critical data to produce a 'deeply misleading' result. The crux of Pielke's article is this: Extreme weather events are costing us more and more money, but that is not because climate change is making extreme weather more frequent or intense. The reason we are losing more money, rather, is because we have more money to lose. Pielke came to this conclusion by measuring rising disaster damage costs alongside the rising global Gross Domestic Product. He also cited a U.N. climate report, along with his own research, to assert that extreme weather events have not been increasing in frequency or intensity."
Refreshing there is some common sense creeping into this global warming/climate change/the new name when the current one looses its umph. Naturally the pro-we-ignore-the-earths-climate-has-changed-over-millions-of-years crowd cry foul. I cannot ever recall a group of scientists like these folks be so opposed and go to the lengths they do to squelch any and all dissenting views. That is not science but fanaticism.
My karma is not a Chameleon.
He also cited a U.N. climate report, along with his own research, to assert that extreme weather events have not been increasing in frequency or intensity. Aren't extreme weather events and their relative energy levels easy to gauge and track? Why is this controversial? Either there are more extreme / extremely powerful events, the average energy level increasing, or there aren't. Im sure that (like economists do for inflation), factors that are constant and not constant (like solar output) can be factored.
Most media sites are internet hacks now, posting stories for clicks, that's all they care about (alla Newsweek) and guess what, reddit is their big secret! Nate Silver was one of the very few that stuck to the data, and was trustworthy. But that was then, this is now. He has to prove his new venture is going to be accurate and truthful and not just another HuffPost type bullcrap.
~~~Please pass the salt, I hate unsalted MD5s
Most of the complaints about the original article are suspect, and the primary and most solid complaint (technological innovations in structures) has in fact been accounted for by Pielke.
I would trust someone vetted by Nate Silver a great deal more than anyone posting on a highly partisan site like "Think Progress" - the goal of FiveThirtyEight being to bring real and carefully considered data to have a conversation based on science, not emotion.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Nobody is ignoring natural climate fluctuations. Nobody. The fact that the climate fluctuates naturally does not argue against anthropogenic climate change any more than the fact that the weather changes from day to day argues against the existence of seasons. How about you come back when you have an argument that hasn't already been debunked based on evidence here:
https://www.skepticalscience.com/argument.php
I'm guessing you ignored the part where he fudged the data he did use, and ignored a whole other pile of data and criticisms of his previous analyses, in order to produce this result?
I gather you didn't read to the end where the update explained why none of what you said is true?
538's original mix, sports and politics, are both essentially spectator sports. The major interest in entertainment and people watch for the narratives. Seeking to drive interest (and appease partisans) media come up with false narratives that ignore data. This creates a lot of low hanging fruit for 538 to take the data and point out the narratives are wrong.
I think that 538 has made the mistake of believing that this low-hanging fruit exists elsewhere. When you have multiple groups of writers all trying to generate the best 2.5 hours of cable news punditry every week you're going to get a lot of easily debunked BS. When you try to apply that same once over data analysis to areas of serious scientific study you're going to be the one spewing BS.
I hope Silver can find some additional areas of news that are in real need of analysis because trying to do original scientific research in a news article won't end well.
I stole this Sig
The day before this article came out, the AAAS released a report on The Reality, Risks and Response to Climate Change, and seems to be starting a publicity push on the topic.
Here's what I see -- the majority of scientists believe that there are real problems with global warming, even if there may be some cyclic effects (heat kills off all the humans, they stop causing problems, everything cools back down).
So instead we have groups trying to sow disinformation with questions about the incidence of some severe weather events (are we just monitoring better and catching more, in part because humans are in more places, or are they actually increasing), and are the increases in intensity statistically significant?
And at this point, I've seen some data that might've been tainted (eg, temperature monitors that have had buildings encroach), but the general concensus is that yes, storms are getting worse.
I'm not going to say his results are completely bunk, as he's likely right in that some of the problems can be explained by how and where people build (eg, in the flood plain -- but the flood plain was resurveyed and is growing in my area ... that might be because of silting up of rivers from construction, it could be because of increased rainfall))
Where I do fault the article is for referencing a 'recent' UN report that hasn't been released yet (website says "The Summary for Policymakers will be released on 31 March 2014"), so we can't actually get to the underlying data that he's basing his claims on.
Build it, and they will come^Hplain.
The evidence claims that so far, there hasn't been an increase in monetary cost of natural disasters relative to GDP. I'll let other, more informed people tackle this factual issue.* My problem is purely based on faulty logic; at the end of the article, the author extrapolates that this trend of disaster damage being correlated with and caused by increases in GDP will continue indefinitely. But the only evidence cited for the conclusion that climate change won't ever cause increased natural disasters actually says that US tropical cyclones won't significantly increase in frequency and severity for several decades; I found nothing about winter storms/polar vortex, crop loss due to drought, sea level rise, etc. and I'm not even sure how accurately you can extrapolate to tropical cycles in other places, not to mention many of us hope to still be around in several decades. I appreciate that Nate Silver is a great statistician, but this is going to go downhill really quick if the conclusions of articles posted on his site are only tangentially related to the actual statistics.
*The other disappointing thing is that the author has claimed this before, has been refuted, and hasn't changed his argument even so much as to mention the points made by various people who had rebuttals.
Pielke isn't a actual scientist, he's a political scientist who doesn't understand real science enough
He worked at the National Center for Atmospheric Research as a REAL scientist for eight years. Possibly using his mathematics degree, you realized he had one?
You weren't actually basing your understanding on who he was based solely on what someone trying to discredit him painted him as... right? Right?? Sigh.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Extreme weather cost more due to several factors:
It's just a trade-off between ability to get an income and the risk of suffering extreme conditions.
If the weather gets worse - it's harder to tell since you have to cover the weather for decades to see through the statistical noise. If storms hits a coastal region on average every 10 years it may still be a month between the storms and then 20 years of silence.
If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.