Scientists as informers of public policy

I have just come out of the EGU session on geoethics and jotted down a couple of thoughts in my notepad. One of the speakers told us about placements their organisation (Geology for Global Development) coordinates help scientists learn about social and ethical issues and how they relate to their research.

He specifically mentioned teaching of participatory decision-making.

I feel that the concept of participatory decision-making is tricky for scientists. Science, as a method of inquiry, is based on the idea that there is a single result, a clear truth to be uncovered.

So then, scientists might have a natural tendency to think that experts would always make the best decisions, and that these decisions can be improved by increasing knowledge.

In the reality of public policy, on scales from family units to nations to the global community, there is no objective best policy. The consequences of each policy are dependent on the set of values and.opinions through which it is viewed. Essentially, people make things complicated. The natural world can be shown to behave according to certain laws. Approaches using game theory attempt to do the same with humans, but it is clear that social and cultural differences among humans affect their decision-making preferences. Essentially, as a wealthy white male living in the UK I am not in a position to define how the life of a woman in a drought-afflicted African country would be best improved.

In reality there is no ‘best’ policy, only consequences affecting different people in different ways. We need to map out these consequences (making use of scientific information, of course) to make an informed decision in a democratic fashion. This is why we need participatory decision-making.

Getting the balance right: talks, posters and more at #EGU2013

This post is a collection of thoughts about learning how to ‘operate’ huge conferences like EGU. It’s my first time doing this and I’m now starting to learn how to get the most out of it.

First I’d like to talk about the bread-and-butter of the conference, oral presentations (talks). I like going to talks. A well-designed 12-minute talk can give a great overview of the research, communicating novel methods and approaches as well as key results. However, a full day of 12-minute talks is quite draining. My EGU personal programme for the first two days was completely filled with talks and by the end of the first day I was craving a poster session. I enjoy chatting to people over their poster more than I do sitting down and listening to a talk. Going to a talk is a more passive experience. For me, posters are a little more ‘human’. There is time for detailed conversation on topics of your choosing, or you can take a less tiring approach and just wander down the aisles of the poster halls and skim the posters of interest.

Days at the EGU meeting are structured such that most talks happen in the morning and posters in the afternoon. I like this. It recognises that talks are more taxing and puts them early on, allowing people to relax later in the day. However, on Monday and Tuesday I spent the 1730-2000 slot in two excellent short courses on tipping points and predictability. I will write a separate post about them later because they really were superb, but the long and short of it is that I found them a very valuable use of my time. I didn’t miss the posters at all during these sessions because the lecturers were so engaging.

Talks and short courses: that’s been my EGU so far. This evening I will get the chance to go to some poster sessions, which I am really looking forward to.

And yet I’m still missing out on huge chunks of the EGU experience! My calendar is full of sessions running in parallel, all of which I would love to go to. I’ve missed out on some excellent Medal Lectures, which I have heard from friends are really nice (a break from the short, functional oral presentations). When I put together my programme (using the excellent smartphone app) I quickly realised I was going to have to get used to this feeling that I was missing out. Yesterday I tried to flit between sessions, aiming to attend specific presentations. This can be done, but it gets complicated quite quickly, and sometimes it can take a while to get to the new room.

Today I have taken a more laid-back approach. If a talk comes up that isn’t particularly relevant to me I will use that opportunity to ‘zone out’ and rest my brain. As the conference goes on I am finding downtime to be quite important! There is a park close to the conference centre which offers a chance to get some fresh air and relax. I find a relaxing lunch really helps me come back refreshed and ready to engage in the afternoon sessions.

My plan for this afternoon includes the remainder of the session on clouds, aerosols and radiation, the debate on fracking, and a whole host of posters on a remainder of topics. That’s the kind of mix I love: some talks relevant to my own work, something rather different with potential for great discussion, all rounded off by a walk around the poster hall. This evening I’ll be heading to the EGU Tweetup, a meeting for scientists interested in using Twitter and similar tools for science communication. See the #egutweetup hashtag for more details.

First impressions of #EGU2013

image

Arriving at the Vienna International Centre is a very exciting experience. I went to the conference centre yesterday to pick up a programme and to get my bearings, together with some other PhD students from Reading.

We wandered amongst the half-prepared stalls and peeked into the rooms in which our presentations would be held. It was exciting to imagine the place jam-packed with scientists and bursting with new research ideas.

After a late lunch in the Innerstadt we headed back to the hotel to relax before going back to the conference centre for the opening reception. People flooded into the conference centre to meet old friends and make new ones. I was in bed quite early though, given I was presenting at 9:30 the following morning!

image

I presented my work at the Open Climate session. The audience was bigger than I had expected, but I think it went OK. It’s very easy to think of all the things one should have said once the talk is done. It wasn’t a disaster though, which means it counts as a success. This was my first time presenting at such a big conference and I am sure future presentations will be much improved by the experience.

I got some good feedback and comments after the talk, which was much appreciated. Now I can fully.enjoy the delectable menu of scientific delights EGU 2013 has to offer.

It’s going to be a great week. The Sun has even come out!

New chapters and foreign lands

By January of this year a major chunk of my PhD work was winding to an end. I had spent a long while looking at the effect of stratospheric aerosol geoengineering on the circulation of the stratosphere, which will form the second (and probably biggest) of the three results chapters in my PhD thesis. At the start of my project I had spent a lot of time reading about stratospheric dynamics and it’s now a somewhat familiar area to me. Well, almost. I still find some chunks counter-intuitive, and sometimes downright baffling. But at least it’s baffling in a familiar way.

We are always reluctant to move away from the familiar. But the work for my final results chapter required that I take the plunge into material and theory that was very much unfamiliar. I was going to look at the impacts of geoengineering on the tropospheric circulation. I eased myself into it by thinking about the tropospheric jet streams first. I at least have some grounding in this area. The dynamics of the midlatitude jet streams is somewhat similar to the dynamics of the stratosphere, and my undergraduate degree in Meteorology has quite a heavy emphasis on the theoretical underpinning of it all. The work on the jet was a nice transition.

Recently (over the past month or so) I have been thinking about precipitation. Especially tropical precipitation. Now, the Earth’s Tropics are meteorologically very different from the midlatitudes. In the midlatitudes the Coriolis Force is a significant effect and weather is determined by large scale wavelike motions producing depressions and anticyclones. Rainfall is mostly frontal in nature. In the Tropics the Coriolis Force is negligible. As a consequence we don’t usually see very large horizontal temperature gradients. This means we don’t see large, rotating weather systems. Rain comes from convective storms, on a much smaller scale than midlatitude frontal depressions. There is so much moisture in the air in the Tropics that the vertical temperature profile pretty much everywhere shows evidence of the release of heat when water vapour condenses to form rain. This forms a characteristic moist adiabatic temperature profile (see image below). Without a strong Coriolis Force this temperature profile is spread over the Tropical belt, so we see it even outside the rain-producing regions.

In order to interpret my model results I had to learn to think differently. Intuitions learned from midlatitude dynamics don’t apply this close to the Equator.

Learning new theory can be pretty intimidating. It’s difficult to know which paper to read first. Sometimes I find myself feeling paralysed. I have a pile of things to read but keep having to refer to different sources to understand terminology, or to get to the bottom of some ‘obvious’ physical understanding not fully explained in one piece of research. Then I took a different tack. I went to see one of the hundreds of other people working in the Department of Meteorology.

This department has experts on any conceivable area, and now, when I’m learning new theory, this is becoming invaluable. In a single hour with a researcher in tropical meteorology I ‘got’ it. I understood the fundamental differences between tropical and midlatitude thinking. Now I can read those papers with confidence. Now I understand the terminology, and a little of the intuition as well. Self-teaching works well (and is entirely necessary for a PhD student) but spending a little time with an expert can help one learn how to teach oneself. This is much the same as, how, when learning a foreign language, you must first learn enough to communicate on a basic level. Once you have that, you can begin to immerse yourself, to learn from conversation with native speakers. The amount of learning that goes on increases exponentially with time. You learn far more from native speakers. But you need to do that initial bit of work to access this higher plane of learning.

Talking of foreign lands, I will soon be off to the 2013 EGU (European Geosciences Union) General Meeting in Vienna. It’s a colossal conference (nearly 12,000 people attended last year) and I’m sure the experience will be educational, entertaining, confusing and exhausting. I’m sure I could list adjectives forever on that one. I will try to write some blog posts and Tweets during the conference, reflecting on what it’s like for a naive young PhD student to be launched into one of the world’s biggest academic conferences.

RMetS meeting report: Lessons on Climate and Modelling from the Palaeo Record

Last week’s Royal Meteorological Society National Meeting took us on a whirlwind tour of the Earth’s climate in the deep past.

In his introduction, Alan Haywood described the timescales on which palaeo studies operate. Most of the Earth’s past has been spent in a ‘Greenhouse World‘ state, where there is no ice anywhere on the planet. The palaeo record takes us back to the Greenhouse World and the time of the dinosaurs, 40-80 million years ago. Then, 15 million years ago, ice began to grow at the poles and the Earth became an ‘Icehouse World’. Note this is not the same as an ‘Ice Age’. We are in the Icehouse World right now, because there is ice at the poles, but the ice extent is much lower than in an Ice Age.

Studying climate data from these periods can give us useful information about how the Earth’s climate system behaves. Crucially, as Alan Haywood pointed out, they provide an ‘out of sample test’ for climate models. Observations of the present-day and the recent past are used in the development of climate models, since a bare minimum requirement is that it should simulate the main recognisable features of the present-day Earth. One must be careful when comparing the results from climate models to these observations because the same observations could have been used in developing the model, generating a false sense of confidence in the model. Palaeo data are generally not used in model development.

There are some disadvantages to using palaeo data. The data themselves are often quite uncertain, and the coverage across the global can be very sparse indeed. These data aren’t really observations. They are based on proxies – indicators of previous climates such as remnants of tiny animals and bubbles trapped deep in ice sheets. Observations do not offer a gold-standard absolute truth. In reality, observations themselves are uncertain and this must be taken into account. The problem of observational uncertainty becomes even greater for the deep past. Several times during the meeting this theme emerged: if a model is shown to be different from proxy-derived observations, this isn’t necessarily evidence that the model is wrong. The best truth we can arrive at lies in a subtle synthesis of evidence from both models and proxies.

Aisling Dolan (University of Leeds) presented some multi-model simulations of the climate during the Pliocene (around 3 million years ago) from the PlioMIP project (Pliocene Model Intercomparison Project). She showed that there were substantial discrepancies between the models and proxies and among models themselves. The uncertainties in the models and the proxies were large enough to explain this discrepancy. This is one of those – irritatingly common – situations in climate science where little information can be gleaned. The large uncertainties don’t mean the models and proxies agree: merely that they cannot be said to be inconsistent with each other.

She mentioned something interesting about the PlioMIP experimental design. The point of a ‘MIP’ is the run different models under exactly the same setup so we can understand where models disagree and hopefully work out why. The PlioMIP setup involves using constant concentrations of greenhouse gases, solar activity and so on. These constant values represent an average over a long period. But, as Aisling pointed out, our observations of the Pliocene don’t represent an average over a long period of time. For example, the average carbon dioxide concentration used in PlioMIP could represent the concentration 3 million years ago, while the average solar activity could look like 3.2 million years ago. So we are not really giving models a fair chance here.

Matthew Pound (University of Northumbria) spoke about the Miocene, a very interesting period in the record where there were potentially low carbon dioxide concentrations but a very warm climate. He found that a model could not reproduce Miocene vegetation using low carbon dioxide. Once again, though, he pointed out the huge uncertainty in the measurements of carbon dioxide during this period. One reconstruction of global carbon dioxide concentrations in the Miocene puts them at 400 ppmv – at which level the model does a reasonable job.

‘Not inconsistent’…uncertainties strike again. But it is often the job of science to accurately determine the level of our ignorance rather than the level of our knowledge.

Palaeo data can help us understand the sensitivity of the Earth’s climate to changes in carbon dioxide. Mat Collins (University of Exeter) spoke about using these data to calculate this sensitivity. The IPCC express it as a global-mean temperature change for a doubling of carbon dioxide, and put the likely range at 2 to 4.5 degrees C. He noted an interesting conflict: palaeo data suggest climate sensitivity is rather high, whereas recent observations suggest it is on the low end of the IPCC range. There is still plenty to be done on the climate sensitivity issue.

The meeting closed with a panel discussion which summarised the day’s talks and gave the speakers a change to agree or disagree as they pleased. It was a fascinating discussion: informal, yet well-structured, engaging and informative. Probably the single best section of any RMetS meeting I have been to, in fact. Some of the discussion centred on the idea of finding, sometime in the past, an analogue for the climate of the present or the near-future. The idea is that, by studying the history of this period, we can gain some insights into current climate and how it may change. The problem here is that it is very unlikely that such an analogue exists. Humans have altered the Earth’s chemistry and biology in such novel ways that we leave quite a unique signature on the planet – leading some to call this age the ‘Anthropocene‘. Ed Hawkins, a scientist from my Department who was also at the meeting, added another point on Twitter.

Most of the time, the climate of the past varied quite slowly, and so at any point in time we could say the climate was roughly in equilibrium. In reality the climate always changes slightly, but most of the time it exists in a relatively stable state. Compare this to the present-day. We know the present climate is very much not in equilibrium, and the changes are likely to become more rapid in the future. Studying equilibrium climates (which may remain stationary for long time periods) may not tell us what we need to know about climate changes over short timescales.

Trade-offs between biofuels, pollution and human health

I came across a fascinating news item on Twitter the other day, publicising a paper  (Ashworth et al. 2013) about the effects of certain biofuel crops on air quality. This isn’t my area of expertise at all (assuming I have expertise, that is!) but I found it an excellent example of the trade-offs humans make in their interventions in the environment. In densely-populated and highly-industrialised Europe, where this study is focused, there is no such thing as ‘nature’ or the ‘natural environment’. Human and natural systems are intertwined. The study demonstrates how complex this relationship is.

Poplar – potential biofuel crop (C) di bo di

Isoprene emissions from woody biofuels impact human health

In a nutshell, the study showed that replacing crops or grassland with woody biofuel crops like ash, poplar and willow (which they call ‘short rotation coppice’) increases concentrations of a chemical called isoprene in the atmosphere. The vast majority of the isoprene in the atmosphere comes from plants. Isoprene is quite reactive and reacts with other chemicals in the atmosphere. If there is a lot of nitric oxide pollution around, isoprene reacts with it and forms ozone. Ozone is very important for shielding the Earth’s surface from harmful solar ultra-violet radiation, but it is best kept high in the atmosphere because it is also harmful to human health.

Nitric oxide and isoprene, therefore, is a bad mixture to have in the atmosphere. The resulting ozone is linked to asthma, bronchitis and heart attacks. Planting 72 million hectares of biofuel crops across Europe, the study estimated the isoprene-related ozone could cause between 690 and 1,890 additional deaths each year. Ozone is also harmful to crop growth, and they estimated crop losses with a value of $1-2bn (in 2010 dollars).

Other effects of isoprene

Isoprene is also implicated in the formation of secondary organic aerosol. These are tiny particles (‘aerosols’) which both absorb and reflect solar radiation. Most aerosols (except very sooty ones) tend to be more reflective, which means they cool the surface. A cloud of aerosols works much like a cloud of water droplets in this sense, providing a sunshade. These aerosols are ‘organic’ because they come from compounds produced by plants, and ‘secondary’ because these compounds first have to undergo some chemical reactions in the atmosphere before the aerosols are produced.

Secondary organic aerosols are still quite poorly understood and offer plenty of interesting research opportunities. We don’t even understand whether isoprene always increases the amount of secondary organic aerosol or whether it sometimes decreases it.

Choose your biofuel crops carefully

In my research for this post I came across another paper (Crespo et al. 2013) which looked at emissions of isoprene (and other so-called ‘volatile organic compounds’) from different types of biofuel crops. They showed that the problem of isoprene emissions is much bigger for woody crops, such as the ones used in the Ashworth study I mentioned earlier. Most plants emit isoprene but some do so much more than others.

[O]ur data suggest that the use of perennial grasses for extensive growing for biofuel production have lower emissions than woody species, which might be important for regional atmospheric chemistry.

The ‘perennial grasses’ they study are things like ‘elephant grass’, a fast-growing crop which is already being grown in the UK.

I find the interplay between the ‘natural’ and the ‘human’ especially interesting here. Humans may think they are doing something natural (or at least modifying the environment is a sympathetic way) by planting these crops. The ‘natural’ isoprene emissions combine with human pollution (nitric oxide) and produce ozone pollution, which affects human health, crop yields and ‘natural’ plant growth.

The working day of a PhD student

The results are in! Over the past month I have been logging my work patterns. I explained my motivation in a post before I started. I wanted to know how long I spent working, and what I was doing during working hours. The question was: can a PhD student get away with just working 9-5?

The answer is, broadly, yes.

How long do I spend working?

On an average working day I spent 8.57 hours ‘at work’. This is rather loosely defined, but essentially it means I am either at my desk on somewhere else on the university campus. I am not actually doing anything PhD-related for 2.17 of those hours. That brings my daily work time down to 6.4 hours.

Before I began, I estimated that I would spend 6 hours a day working. I may have overestimated my actual work time because I tended to round my working half-hours up (see the Methods section below).

The graph above shows the number of hours worked per day. The 8-hour line represents the standard length of a 9-5 working day. Some days were rather unproductive with large amounts of time ‘greyed out’ where no work was done. You can see at the start of the month, coming back from Christmas, I was very keen. I even did some reading on the weekend. Later on my working time waned slightly. You can also see that the second Friday was a very short working day. It snowed that day and I was finding it difficult to concentrate so I took the afternoon off!

What was I doing?

I spent the biggest chunk of my time coding. This applies to making plots as well as analysing data. Since climate modelling is necessarily computer-based, it’s no surprise coding comes out on top. Next comes my ‘nowork’ category, which I discussed above. I had just over two hours off per day. I rarely take clearly-defined lunch and tea breaks, but I suppose these two hours can represent them. In reality those hours were spent procrastinating on the Internet or doing other things not related to my PhD. This includes organisational tasks like booking transport to meetings and conferences, maintaining websites, and so on.

‘Understanding’ covered research I did with a specific purpose, like finding out how to do a particular type of analysis, or comparing my results with others. This explains why the time I spent ‘reading’ was comparatively small. My purposeful research obviously involved plenty of reading. The ‘reading’ category was specifically for reading new and interesting papers to develop my background knowledge, rather than to find a specific piece of information.

I was surprised how much time I spent in meetings. I think the proportion is artificially large because the Department held a Research Day in early January, when representatives of research groups give summaries of their work to the rest of the Department. Then again, I only went to an afternoon’s worth of talks. I value meetings. Obviously those with my supervisors are very important, but seminars are an excellent way to broaden my education. Recently I have also been thinking about acquiring new skills for analysis, and one of the best ways to learn about these is through seminars.

Finally, the ‘writing’ part: it’s very small, but that is because I haven’t really started my thesis in earnest yet. If I was at a loose end I would do a short stint adding some material or reorganising what I already had.

There is enough time

These results have shown that I haven’t been overly efficient during working hours. I don’t work very hard (in terms of work out versus time in), and yet my output is reasonable, as is the quality of my work. I worked slightly over the standard 8-hour day, but if I really wanted to I could trim this down without sacrificing quantity or quality. In fact, I might gain productivity by constraining my working day, because I’m likely to become less lazy and less easily distracted.

Methods

I used a Google Doc spreadsheet to record the time spent at work. I recorded my predominant activity in half-hour blocks. Work doesn’t fall neatly into blocks like this so I had to do some subjective re-jigging. For example, if I start work at 9:15, so I write in the 9:00 – 9:30 slot that I did some work? I went by the rule of thumb that, if I worked for more than half the time in the slot, I would record it. I categorised my time as follows:

  • understanding – active research, including reading, to find out specific things
  • coding – making plots or running models
  • meeting – talking to other scientists
  • writing – thesis or other smaller piece of writing; blog posts
  • reading – reading papers and other articles without a clearly-defined purpose
  • nowork – breaks, organisational stuff, admin

I think there is more information to be pulled out here. I actually find it quite interesting to just look at the colour-coded allocation of time in my spreadsheet rather than the graphs above. One can pick out, by eye, that I tend to do coding in long, uninterrupted blocks. I find it easier when I have got into the ‘flow’; plus it is simply time-consuming. Please do take a look at the spreadsheet. I would be interested in hearing comments on the results or the method, since I may be trying this again in the future!