2010-10-13 20:22:18Feedback welcome: Do critics of the hockey stick realise what they're arguing for?
John Cook

john@skepticalscience...
124.186.160.198

The hockey stick, a reconstruction of temperature over the last 1000 or so years, is a much maligned graph. Critics of the hockey stick insist it underestimates past climate change. In particular, many insist that temperatures during the Medieval Warm Period were warmer than now. The next logical step from here is that if past natural climate change is comparable to today, then current climate change is also natural. The irony of this line of thinking is that if the Medieval Warm Period did turn out to be much warmer than currently thought, this doesn't prove that humans aren't causing global warming. On the contrary, it would mean the danger from man-made global warming is greater than expected.


Figure 1: Northern hemisphere temperature reconstruction from Moberg te al 2005 plus instrumental temperature measurements of northern hemisphere (HadCRUT).

To understand this, you first need to grasp the fact that climate doesn't change by magic. It changes when it's forced to change. When our planet suffers an energy imbalance (eg - the energy imbalance caused by rising CO2), it gains or loses heat. This change in heat is known as a radiative forcing or climate forcing. When our climate experiences a forcing, global temperature changes.

So to understand climate change over the last 1000 years, you need to look at the climate forcing over that time. The overall or net climate forcing is the combined effect of the drivers of climate over this time frame: mainly solar variations, changes in carbon dioxide and aerosols:


Figure 2: The combined radiative forcing from solar variations, CO2, volcanoes and aerosols (Crowley 2000).

The reason we see a hockey stick shape in temperature is because the climate forcing that drives temperature also shows a hockey stick shape. But from this data, we can do a lot more than compare shapes. We can determine how much global temperature changes when it's subjected to a climate forcing. This information is crucial in enabling us to predict how climate will act in future decades in response to rising greenhouse gas levels.

The temperature response to a climate forcing is known as climate sensitivity. Technically, climate sensitivity is defined as the change in global temperature if the planet experiences a climate forcing of 3.7 Watts/m2 (which is how much climate forcing you get from a doubling of CO2). The amount of positive feedback in our climate system determines how sensitive our climate is. If there's net negative feedback, the climate sensitivity will be less than 1.1°C. If climate sensitivity is greater than 1.2°C, our planet has net positive feedback. Climate sensitivity can be calculated by using temperature change over the past 750 years along with the change in radiative forcing (Hegerl et al 2000). Doing this yields the following result:


Figure 3: Climate sensitivity from palaeoreconstructions going back 750 years, combined with climate sensitivity calculated from instrumental records. The horizontal bars represent the 5 to 95% range, indicating a climate sensitivity range of 1.5C to 6.2C (Hegerl et al 2000).

When you combine the temperature record over the past millennium with climate forcings, you get a climate sensitivity around 3°C. In other words, net positive feedback that is consistent with the IPCC range of 2°C to 4.5°C. This positive feedback is the reason why we expect to see strong warming over the next century in response to the climate forcing from rising CO2.

However, if for some reason, temperatures over the Medieval Warm Period turned out to be warmer than previously thought, this means climate sensitivity is actually greater than 3°C. In other words, the climate response to CO2 forcing will be even greater than expected. So to argue for a warmer Medieval Warm Period is to argue for greater climate sensitivity and greater future warming due to human CO2 emissions.

2010-10-13 21:25:25Couple of comments
Ari Jokimäki

arijmaki@yahoo...
192.100.112.202

"In particular, many insist that temperatures during the Medieval Warm Period were at least or even warmer than now."

There seems to be something missing after "at least".

"The effect from volcanoes weren't included as we're interested in long-term climate change, not the cooling effect over a few years from a volcanic eruption."

When volcanic forcing is considered, the frequency of volcanic eruptions is more important factor than the cooling effect of a single eruption. If certain time period has high frequency of volcanic eruptions, you can have longer term climate change from volcanic forcing. In that sense I think your argument here misses the point.

2010-10-13 21:32:00Addition
Ari Jokimäki

arijmaki@yahoo...
192.100.112.202

By the way, the reference you use for forcings (Crowley, 2000) discusses volcanic forcing and says this among other things:

"There is increasing evidence (3, 7–10) that pulses of volcanism significantly contributed to decadal-scale climate variability in the Little Ice Age."

Here's the full text of Crowley (2000):

http://www.geos.ed.ac.uk/homes/tcrowley/crowley_science2000.pdf

2010-10-13 22:15:36Ok, I'm starting to see the value of blog review in this forum now :-)
John Cook

john@skepticalscience...
124.186.160.198
Thanks for the feedback, Ari. I'll look again at including the volcanic forcing in my forcing graph. I did initially but it results in these huge ugly spikes off the bottom of the graph. It's very distracting! I figured they only have a short-term influence so no one will miss them, right? Wrong! Maybe a 10 year average will smooth some of the nastier bumps out :-)
2010-10-13 22:35:49
Ari Jokimäki

arijmaki@yahoo...
192.100.112.202

It does look ugly. For a comparison, here's Hansen et al. (2007), who give different forcings and their net (for modern times) in figure 5 (NOTE, the file size is over 20 MB) - it looks rather spikey:

http://pubs.giss.nasa.gov/docs/2007/2007_Hansen_etal_3.pdf

2010-10-14 04:50:46some suggestions
dana1981
Dana Nuccitelli
dana1981@yahoo...
38.223.231.249

"In particular, many insist that temperatures during the Medieval Warm Period were at least as warm or even warmer than now" (whoops, Ari already caught this one).

When you talk about temperature changes caused by forcings, you could reference the 'how do we know CO2 is causing warming' advanced rebuttal.

"The overall or net climate forcing is the combined effect of the main drivers of climate over this time frame:"

I would change this to "The overall or net climate forcing is the combined effect of the drivers of climate over this time frame: mainly solar variations, changes in carbon dioxide and aerosols".  Because the net forcing is mainly, but not only due to those three factors.

When you start talking about climate sensitivity, you might want to reference the 'climate sensitivity is low' rebuttal.

"climate sensitivity is defined as the change in global temperature if the planet experiences a climate forcing of 2.7 Watts/m2"

It's 3.7 W/m2.  5.35*ln2.

"If there's net negative feedback, the climate sensitivity will be less than 1.1°C"

Why 1.1°C?  Isn't the warming from a doubling of CO2 alone 1.2°C?  Regardless, you should explain where this number came from.

"In other words, net positive feedback that is consistent with the IPCC range of 2.5°C to 4.5°C."

IPCC range is 2 to 4.5°C.

You could also add a bit about forcing efficacies (as discussed in the climate sensitivity rebuttal linked above), since technically climate sensitivity isn't a one size fits all factor, which is how you're treating it here.  But that just depends on how technical you want to get.  You could just add a sentence that the efficiacies of different forcings vary, but none are much larger than CO2, and many (like solar) are lower.  Or you could just leave it out, but I got criticized for that in the original version of the climate sensitivity rebuttal.  It's a valid point, but a little complex.

Overall it's good, just need to clean up these details.  This is a very imporant skeptic contradiction which is well-worth highlighting.

2010-10-15 14:47:42suggestions for the graph
jyyh
Otto Lehikoinen
otanle@hotmail...
80.186.85.135
include the negative forcings on top, so the highest line would be 'the T without negative forcings', adding natural variations like ENSO could be around the actual temperature (though this should be inverted). the final image should include the lines for actual T and the projected T from forcings, but the image should be quite large since they are not easily separated during the last 60 years :-).
2010-10-21 07:13:31Goddard
dana1981
Dana Nuccitelli
dana1981@yahoo...
38.223.231.249

Steve Goddard reacts to this blog post with his usual "open mouth, insert foot" routine.  He calls John's Moberg+instrumental graph "BS" and "whacked" because he assumes it uses land+ocean data, which of course it doesn't.  Rather than investigate possible alternatives, Goddard immediately goes into attack mode, embarrassing himself by basically calling John a fraud for using the correct data (since Moberg's, like most temperature reconstructions used NH land-only proxies).

When his errors are pointed out in the comments, Goddard of course refuses to admit any fault.  He falls back on the position that he was merely "calling him" on not specifying that the HadCRUT data was land-only.

I can't believe this ignoramus has the balls to run a climate blog.  Or that anybody actually reads it.

2010-10-22 20:39:32Steve's post
John Cook

john@skepticalscience...
124.186.160.198
To be honest, I'm a little annoyed with myself that I made the tiny error of omitting the "land" in my caption. It enabled Goddard to go on a big rant completely missing the point of my post, which was about proxy data and climate sensitivity. Instrumental data was irrelevant to the point I was making. The lesson here is to be aware that climate skepticism is at it's heart cognitive dissonance. When you write a blog post, skeptics will seize on any tiny detail to distract themselves from the main point you're making. I've seen it happen time and time again with my own blog posts. So when you write your material, and proofread others' content, try to anticipate what will distract skeptics from your take-home point. Endeavour to remove as many stumbling blocks as possible.

that's why this forum is such a blessing. Having a bunch of friendly eyes who can scrutinise our articles is a wonderful resource.

2010-10-26 00:35:00
MarkR
Mark Richardson
m.t.richardson2@gmail...
192.171.166.144

Something we haven't covered here is Spencer's internal variability stuff.

 

I personally think he has a point: internal forcing sounds perfectly possible, although its effect on recent warming and past warming hasn't been fully demonstrated. It _could_ mean that you get a "natural" medieval warm period in the past with a relatively low climate sensitivity.

But climate science isn't put together from individual pieces of evidence, but it relies on multiple independent sources. The synchronicity of temperature & orbit over many Milankovitch cycles for example is almost certainly not due to random internal variability (the probability of that happening must be effectively 0!*), and the fact they support modern climate sensitivity estimates suggests that Spencer's ideas on internal variability, whilst possibly correct, fail to demosntrate the low sensitivity he wants.

 

 

*I mean 0, not 1. The exclamation mark is for emphasis :P