Monday 24 November 2014

Casual consensus and a lot of science

Have not been blogging much lately and the number of readers is suffering. So I need another post on the Climate Consensus; always a runner. At the Guardian, there is a discussion going on between 3 interesting people. John Cook, the benevolent dictator of Skeptical Science, Peter Thorne, the benevolent dictator of the International Surface Temperature Initiative (ISTI) and the benevolent dictator of And Then There's Physics (ATTP).

What I took from the discussion is that maybe we should not focus too much on communicating the consensus, still communicate it, but casually. That may actually make the consensus point more clearly.

If you do so casually, you produce less an impression that this could be a topic of debate, you thus create less false balance. If you mention the consensus casually, but focus on interesting points of scientific disagreement you also do not give the impression that the consensus means that everything is understood in sufficient detail.

AndThenTheresPhysics paints the dilemma:
"My understanding of the situation is something like this, though. A reasonably vocal group of people argue that there is a great deal of disagreement about climate science and that there is no consensus. Some other people then do a study to show that there is indeed a consensus, at least with regards to the basics. The first group of people then do their utmost to attack that result. Consequently another group of people do another study to show, once again, that there is a consensus. That too is then attacked so as to undermine that there is indeed a consensus. Then another study takes place, until we get to the point that even scientists are starting to question the whole consensus messaging because they perceive it as an attempt to communicate climate science through consensus messaging only (which I don't think is the intent, even if it might seem that way). That scientists are now criticising this then gives those who would rather there was no consensus (or that people thought there was no consensus) more ammunition to attack the various consensus projects."
Peter Thorne replies:
"But how do you break catch-22 here? Its not clear that continuing round the circle achieves anything other than getting dizzy. There are many interesting scientists on blogs and twitter (yourself and skepticalscience included) communicating in varied ways with nuanced messaging that perhaps gives a better sense of the science and the process than repeated articles on a consensus at the Guardian ever could."
"My point would be that while it is important to communicate the consensus it is at the same time a significant mistake in my personal view to obsess upon it or make it even the central strand of any discussion upon the issue."
Clearly we should avoid given the wrong impression that there is no consensus on the basics, but maybe just a half sentence with a link is enough on the internet and the rest of the message can focus on science. In the mass media, which is what Cook is thinking of, a simple message focussed on the existence of a consensus on climate change among climate scientists may well be effective. I am no expert, but Cook's arguments sound convincing.

Even if one accepts that consensus messaging is most effective to convince the population that there is a problem, that still does not mean that scientists should do this. Scientists have their own aims, skills and interests.

Peter Thorne:
Carbon dioxide and gases active in the IR spectrum we have known for over a Century will act to warm the atmosphere. On that you will find as close to consensus amongst qualified experts as in any field.

But that understanding is far from the end of the story and as you get to more and more nuanced questions there is no longer the degree of unanimity or consensus. If we knew everything there is to know then you wouldn't see several thousand papers a year appear in the peer reviewed literature on the subject providing new insights and building the knowledge base. Nor would you see repeated assessment activities such as IPCC. The issue with saying there is consensus repeatedly is that people then think, mistakenly, that all aspects of the science are settled. This is very far from the case.
As scientists we are naturally also very aware of the problems that are not solved. That is what we work on every day. That is the fascinating part. Just because the main lines are solid does not mean that we should no longer expect important changes in our understanding of the climate system and that all we need are applied studies on the impacts of climate change.

It seems to me somewhat premature to invest much manpower into studying how cauliflower, leek or wine with grow within some small region of Germany, France or Luxembourg. Not only because that needs predictions at a very local scale, which are much more difficult than large-scale predictions, but also because there is still important work to do the fundamentals of climate science. I would personally mention the influence of non-climatic changes on trends in extreme weather and I would love to see a global station network making climate-quality measurements, designed to avoid introducing non-climatic changes.

Not that I would argue that we should not do any impact studies. Doing so gives a first view of where the main societal problems may lie, it helps us see where the difficulties of impact studies are and to develop the tools that are needed to make reliable impact studies and estimate the corresponding uncertainties. However, maybe we do not need to study every vegetable for every province at his stage.

The International Surface Temperature is building an open global temperature dataset with good provenance relying on volunteers and some support by NOAA. For the first time on a global scale the ISTI will validate the algorithms to remove non-climatic temperature changes. Generating the validation dataset is supported by the MetOffice, but mainly performed by volunteers. My own smaller-scale validation study was a volunteer project, with some travel funding by COST. Peter is co-chair of GRUAN, a network of climate-quality radiosondes. A rather sparse network. And governments around the world are pruning the station network, regularly destroying some of the longest series we have. I am surely bias, but I would put my priorities in the data the science is founded on and not with cauliflower.

Yes, it is warming, but how much, where and when. If that changes due to our better understanding, we can redo all the studies for every vegetable and province. And that is just temperature. Many more aspects of the climate are changing. And those are just examples from my field. Other climate scientists could likely make a similar list of important questions that still need to be resolved. Especially, when it comes to adaptation local skill is important and hard to get. Investing in basic science may well save a lot of money for unnecessary adaptation measures.

Perhaps Peter Thorne said it better than I have:
"If we want to make truly informed and effective adaptation and mitigation decisions it is incredibly dangerous to contend that there is a consensus on anything more than the most general abstract aspects such as that Carbon Dioxide emissions are causing warming."

Citizen and scientist

When a mitigation sceptics doubt something basic, I find it natural for a normal citizen to answer, well there is a clear consensus on this topic, that is enough for me to hold this view, convince the scientists first before bothering me as a non-expert. That is just a shorthand for: I do not want to discuss this with you, I expect this to be pure nonsense and we are not the right persons to discuss this.

I would prefer it if a scientist (in the right field) would answer by providing the evidence. This is our role in society, even if it is not the most effective strategy to convince the population there is a problem.

The answer is mainly to make clear to the casual reader that there is an answer; one should have no illusions that this will lead to a productive conversation with the mitigation sceptic. If the answer is too good, the mitigation sceptic will change the topic and try the same loop somewhere else.

Even as scientist you do not have to jump all hoops. If someone claims CO2 is not a greenhouse gas, or that the temperature is decreasing or we will soon enter an ice age, there is nothing wrong with asking such fools to first convince their political allies Anthony Watts, Jo Nova or Roy Spencer, who officially reject these claims.

Depending on your role, you communicate differently. Thus maybe the Guardian debate was partially about how people see as their role.

As a citizen I feel that the arguments of John Cook make a lot of sense and I guess that it is necessary and effective to communicate to the publication that there is a consensus within climate science about the basics and that we are performing a dangerous experiment with the climate system our livelihoods depends up on.

In my role as scientist further aspects become important and one should make sure that the consensus message does not give the wrong impression that we already know everything sufficiently accurately and that all we need are impact studies for cauliflower.



Related reading

Why we need to talk about the scientific consensus on climate change. The Guardian article that inspired this post.

Five reasons scientists do not like the consensus on climate change. My first try to explain why scientists may not like communicating consensus and why these arguments do not hold water.

"Blinded by Science: How 'Balanced' Coverage Lets the Scientific Fringe Hijack Reality". A repost of an oldie from 2004 that is still current by Chris Mooney in Columbia Journalism Review.

The BBC will continue fake debates on climate science. "When a new member of our solar system was discovered, there was no debate with an astrologer claiming there was no empirical evidence, because you could not see VP113 with the naked eye."

The value of peer review for science and the press For science peer review is a filter. For the press an anchor.

On consensus and dissent in science - consensus signals credibility You have to pick your fights. On topics where there is dissent there is clearly work to do. Where there is a consensus finding a problem is hard, but thus also most rewarding.

The Tea Party consensus on man-made global warming. The people that call the climate consensus based on scientific evidence "group think", have a very strong consensus themselves, without much scientific evidence. Group think?

Wednesday 5 November 2014

Participate in the best validation study for daily homogenization algorithms

Rachel Warren is working on the validation of homogenization methods that remove non-climatic changes from the distribution of daily temperature data. Such methods are used to make trend estimates for changes in weather extremes and weather variability more accurate.

To study this, she has just released a numerical validation dataset. Everyone is invited to apply their homogenization method to this dataset. It looks to be the most realistic validation dataset produced up to now. Thus it promises to become an important paper for the homogenization community.

Rachel wrote about her study in a post at the blog of the benchmarking group of the International Surface Temperature Initiative. I hope it is okay that I republish it here below.

She is not the Rachel Warren of the Hip-Hop Dance Workout, that would be too much healthy fun for scientists, but Rachel Warren, the statistician from the University of Exeter. Hopefully the healthy smile on her photo makes up for the fun. And the interesting results of the study. VV



Release of a daily benchmark dataset - version 1
by Rachel Warren

Kate Willett's blog post from 6th October gives a detailed over-view of the benchmarking process that forms part of the ISTI's aims. It is hoped that in the long term these benchmarks will not only be produced at the monthly level, but also for daily data.

This post announces the release of a smaller daily benchmark dataset focusing on four regions in North America. These regions can be seen in Figure 1.


Figure 1 Station locations of the four benchmark regions. Blue stations are in all worlds. Red stations only appear in worlds 2 and 3.

These benchmarks have similar aims to the global benchmarks that are currently being produced by the ISTI working group, namely to:
  1. Assess the performance of current homogenisation algorithms and provide feedback to allow for their improvement
  2. Assess how realistic the created benchmarks are, to allow for improvements in future iterations
  3. Quantify the uncertainty that is present in data due to inhomogeneities both before and after homogenisation algorithms have been run on them

A perfect algorithm would return the inhomogeneous data to their clean form – correctly identifying the size and location of the inhomogeneities and adjusting the series accordingly. The inhomogeneities that have been added will not be made known to the testers until the completion of the assessment cycle – mid 2015. This is to ensure that the study is as fair as possible with no testers having prior knowledge of the added inhomogeneities.

The data are formed into three worlds, each consisting of the four regions shown in Figure 1. World 1 is the smallest and contains only those stations shown in blue in Figure 1, Worlds 2 and 3 are the same size as each other and contain all the stations shown.

Homogenisers are requested to prioritise running their algorithms on a single region across worlds instead of on all regions in a single world. This will hopefully maximise the usefulness of this study in assessing the strengths and weaknesses of the process. The order of prioritisation for the regions is Wyoming, South East, North East and finally the South West.

This study will be more effective the more participants it has and if you are interested in participating please contact Rachel Warren (rw307 AT exeter.ac.uk). The results will form part of a PhD thesis and therefore it is requested that they are returned no later than Friday 12th December 2014. However, interested parties who are unable to meet this deadline are also encouraged to contact Rachel.

There will be a further smaller release in the next week that is just focussed on Wyoming and will explore climate characteristics of data instead of just focusing on inhomogeneity characteristics.