I wrote this post and then, after a discussion with Kevin Anchukaitis on Twitter, I took it down. I was tempted to not repost it, but thought I’d rewrite bits and have another go. I keep forgetting how controversial this topic can be.
It’s almost 20 years since the publication of the first hockey stick paper (Mann, Bradley & Hughes 1998). In case people don’t know, the hockey stick refers to millenial temperature reconstructions that look a bit like a hockey stick; a period of centuries during which temperatures appear reasonably flat, or cool slightly (the shaft), followed by a period of rapid warming starting in the mid-1800s (the blade). This rise also began at around the same time as we started using fossil fuels, providing evidence that our use of fossil fuels could be impacting our climate. The first reconstruction (MBH98) was Northern Hemisphere, relied mainly on tree-rings, and went back about 600 years. Later reconstructions are multi-proxy, global and extend back as far as 2000 years.
The hockey stick is rather iconic in the climate debate, with some going so far as to claim that it’s been debunked (it hasn’t) and others suggesting all sorts of nefarious intent. Michael Mann, one of the authors of the original hockey stick paper, has written a nice article about what’s happened since publishing the first paper, and about the need to speak out.
Of course, our understanding of millenial temperatures has improved greatly. The reconstructions extend further back than they did originally. The analysis methods have improved, so we have more confidence in the results. We can use many different proxies to reconstruct these temperatures, so have better spatial and temporal resolution. This allows for an improved understanding of variability and of the role of both internal and external perturbations. However, the big picture has changed little. A warm period about 1000 years ago, typically referred to as the Medieval Warm Period, a general cooling towards what is referred to as the Little Ice Age, and then the modern warm period, that appears unprecedented in the last 1000 years, or so.
I think it’s important to realise that all of this is part of the normal scientific process. Early work will have limited data and will often use methods that have not been tried before. With time, we collect more data, develop improved methods, and – consequently – improve our understanding. We might even conclude that some of the early work used methods that we’d no longer regard as suitable. This doesn’t suddenly invalidate the earlier work, or imply kind of some nefarious intent. It seems pretty clear that the big picture presented by the early millenial temperature reconstructions have stood the test of time. I hope I, one day, have a paper that is still relevant, and being discussed, 20 years after being published.
I’ve always quite liked this post by David Appell that points out that the broad shape of millenial temperature reconstructions (hockey stick like) isn’t surprising.