Bryan's Blog
(Only the last ten entries are here, earlier entries can be found by using the Summary, Archive or Categories pages, or by using the calendar to go to previous months and/or years).
Oh no I can't keep up ..
As you have all noticed, I'm a tad busy at the moment, and blogging has been the obvious thing missing out. The other thing missing out is new technology:
Google Wave I hear you say? No ...
Google Fusion Tables API... oh no not another thing I can't pay attention to.
However, in finding out that I have to ignore another thing, I did follow the link to the NYT in Joe's post, to find a quote from Tony Hey that is SO RIGHT ON:
Tony Hey, a veteran British computer scientist now at Microsoft, said ... "In the U.K I saw many generations of graduates students really sacrificed to doing the low-level IT."
Except it wasn't, and isn't, just the UK!
So timely. I gave a talk today about the pressure of big data, from models and earth observation ... and having to take it all seriously from a national infrastructure point of view ... I'll eventually stick the talk up on my publications page (it's too big to upload from home), but the slides alone won't tell the story ...
Actually there is a lot of other timely stuff going on. I'd like to be talking about the live blogging from AGU ... but I have time to skim the content, but not take it and comment here ...
by Bryan Lawrence : 2009/12/16 : 0 trackbacks : 1 comment (permalink)
Clouds Revisited
I've just been reading "How well do we understand and evaluate climate change feedback processes", by Bony et.al. (2006) which appears in the Journal of Climate.) While I've delved into GCM cloud physics in the past, I've never really taken the trouble (beyond this) to get into cloud feedbacks in the climate sense, I've been happy to accept the received wisdom that cloud feedbacks are the dominant uncertainty in climate sensitivity, but that most folks (Lindzen apart) believe that despite their uncertainty, the sign at least is very probably positive, that is to enhance the affect of increasing CO2 on surface temperature.
This post is by way of notes from half a day following my nose down the rabbit hole, because for various reasons, I need to educate myself on the issue.
Storm intensity and Frequency
(Nothing to do with why I wanted to read the paper, but something I've been interested in for a while.)
Bony et al have an interesting figure which is actually a figure from an earlier paper by Carnell and Senior (1998), which got a mention in the TAR.
 |
As far as I can tell, this paper is close to the basis (see something like 90 citations) of the oft repeated statement that we expect storms to be less frequent but more intense in a future climate.
There have obviously been many follow up papers with other models, including for example, Leckebush and Ulbrich 2004) who in an analysis of GCMS and RCMs found (according to their abstract):
Although the overall number of modelled tracks is underestimated in the control period of the global model's simulation with present-day greenhouse gas forcing, compared to reanalysis data, realistic patterns of the track density over the investigation area are simulated.
Changes occur in particular with respect to the A2 scenario for extreme cyclone systems, while for B2 the changes are less pronounced. Especially over western parts of Central Europe, the track density of extreme cyclones increases for A2, accompanied by a tendency towards more intense systems. With respect to the A2 scenario, a tendency towards more extreme wind events caused by deepening cyclones is identified for several regions of Western Europe such as Spain, France, United Kingdom or Germany.
Additionally, the climate change signal in the regional climate model (RCM) HadRM3H is analysed. In accordance with the signal of the wind speed changes in the GCM simulation, the RCM reveals an increase of the 95th percentile of the daily maximum wind speed over extended parts of Western Europe related to the areas of increased track density of extreme cyclones under the A2 scenario. Changes with respect to the SRES B2 scenario are similar in their structure, but less pronounced in their amplitude.
A bit of googling resulted in an interesting powerpoint by Ruth McDonald which reviews a lot of similar studies ... and then I managed to get hold of Lambert and Fyfe, 2006, which has an analysis making the same point using a CMIP3 multi-model ensemble:
 |
Key changes in clouds in a future climate
(This is what I was after):
Several analyses show no consensus in the global response to clouds and cloud radiative forcing to a given climatic perturbation.
Two analyses showed that the global cloud feedback is positive in all models, but there are large intermodel differences in the magnitude of the feedback.
The frequencies of occurrence of different cloud types (both observed and moeled) are highly unequal and so the behaviour of certain clouds may matter more than that of others in explaining the range of feedbacks.
Several studies show that the responses of deep convective clouds and of low level clouds differ among GCMs.
Changes in the water content of different types of clouds also differ among GCMs ...
Differences in cloud feedbacks in areas dominated by low-top cloud responses make the largest contribution to the variance in the global feedbacks, with
in the tropics, the response differs most between models in subsidence regions (which is consistent with low cloud anyway)
the large fraction of the earth covered with this type of cloud being an important contributor ...
At the same time however, we have reference to Zhang et al, 2005 which took me off on another riff, but that's a topic for another day.
Cloud Physics Parameterisations
(this is what I remain most interested in, science-wise, should I ever get any time to do any science myself ever again ....)
One of the things I'm looking forward to getting out out of metafor is a decent summary of what the current state of play in cloud parameterisations is in GCMs ... I've been out of it for just long enough that it's hard to get back in ... a souped up version of the following table (for just three models, from Wyant. et.al.2006) for all the CMIP5 models should be just the ticket:
 |
Papers I now want to read
For these I could only read the abstracts, due to the paucity of support for climate science in my institutional library (which is great for high energy physics, apparently, but crap for climate) ... I'm glad that I could get to most of the others via self-archived pdfs (yes, I could get to some via our library subscription!)
Uncertainty in Projections of UK Climate Change Resulting from Regional Model Formulation, Rowell, 2006
Changes in mid-latitude variability due to increasing greenhouse gases and sulphate aerosols, Carnell and Senior, 1998
(Note that the figures and table have been downgraded in quality by removal of information, consistent with my fair use policy.)
by Bryan Lawrence : 2009/12/14 (permalink)
The Back of the Envelope and the Removal of Guilt
Some of the most important things one learns in a physics degree are:
Dimensional Analysis (have I stuffed up anything obvious?)
Scale Analysis (what terms in this equation actually matter for this problem?), and
Start with an Envelope (know what your answer is roughly, before further calculation).
The last is pretty fundamental. I used to teach a course called "Nursing Mathematics" at my local polytechnic, for, you'll not be surprised to know, nurses. The entire point of the course was to give nurses the mental arithmetic tools (supplemented on occasion by an envelope and a pen) to know what the answers to most of their day-to-day calculations are, before using a calculator to get things right. You may or may not be surprised to know how important this is, folk have died because of incorrect factors of ten in IV flow rates ... Scale problems happen often with calculators, but less often when individuals have a grip on the scale of answers before attempting the "real" calculation (either from experience, or explicit "pre-calculation").
Which is a long winded way of introducing a post which both validates my original decision to buy an electric lawn mower instead of a petrol mower, and removes my guilt over not buying a push-mower.
I spend a lot of time thinking about this sort of issue, but never get down to writing it out. Well done "King of the Road". My current suspicion is that it would be a good thing to get an electric garden shredder, rather than pile the stuff on a regular basis in a car and drive to the dump ... one day, I will do that calculation ...
Meanwhile, along David MacKay's fabulous book, I feel I now have two "back to basics" places to go to find some numbers about practical ways of dealing with our energy futures.
by Bryan Lawrence : 2009/12/13 : Categories environment : 0 trackbacks : 1 comment (permalink)
Drawn into climategate
It's a long time since I bothered to write a letter to the editor, but the cru email controversy pushed me over the edge. Or to be more precise, Anne McElvoy's opinion piece did. Of course I would never have known anything about it, not normally having access to London Evening Standard , but their letters editor actually approached me to solict an intervention in the next edition (by which I assume that a) he was desperate, and/or b) he figured I might say something silly, or both).
In the event, I must say I was pretty inpressed with the way he elicted the following from me:
Anne McElvoy's opinion piece (25 November) strikes an unrealistic view of what has occurred in the University of East Anglia hacked emails controversy. Extended scientific conversations occur in parallel through various channels, and it's completely unreasonable to expect these conversations to be comprehensible based on a small subset without all the previous baggage. The point of scientific record and peer review is for scientists to stand up their claims and have them evaluated. Of course scientists have a prior view of what data could be telling them but the difference between scientists and most "sceptics" is scientists change their hypotheses when necessary. Most practising scientists will go out of their way to have an honest discussion about issues in their work. Among the sceptics are those who frame their questions in terms which can be addressed; but there aren't enough climate scientists or hours in the day to educate those who believe in the climate physics equivalent of a flat Earth.
with Most practising scientists will go out of their way to have an honest discussion about issues in their work being bit that they chose to highlight.
Of course I never had time to write anything that brief, the following is what I actually wrote (in a tiny amount of time, obviously while doing other things):
The Original
(decorated with a link to my favourite parody)Anne's opinion piece strikes a harsh and unrealistic view of what has occurred in this recent controversy.
My personal opinion is that extended scientific conversations occur in parallel using multiple mechanisms (phone, email, actual meetings etc), and we are only seeing a part of conversations where the correspondents have sensibly used shorthand like "trick" ... and it's completely unreasonable to expect these sort of conversations to be comprehensible in their entirety based on just the email subset without all the other baggage from prior and parallel conversations that's not explicitly included. Indeed, we can construct versions of reality from these subsets of actuality which are completely bogus, and that's what we are seeing happen ... (and the fatuousness of doing so is what some of the parodys that Anne is so dismissive of are trying to demonstrate, for example this).
Further, the entire point of the scientific record and peer review is for folk to be able to stand up their claims and have them evaluated (and to put effort into defending if that's appropriate). The process of doing science, however, involves the construction of hypotheses and their evaluation without documenting every blind alley to a publishable level for the benefit of other folk who steal notes or emails. Of course scientists have a prior view of what the data is or could be telling them, but the difference between scientists and sceptics is that scientists evaluate the evidence and change their hypotheses when necessary. There seems to be no evidence that the sceptics are doing this ... and there is no evidence that I'm aware of in these emails that Phil Jones or any of the other key participants are not.
Finally, most practising scientists will go out of their way to have an honest discussion about the facts and issues of their work, and it's my opinion that that is particularly true of those working in climate sciences. However, what we can't do is make the time for every argument from a community which does indeed seem to "deny" basic physics. Of course, amongst the sceptics are those who can and do frame their questions in terms which can and are addressed, but there aren't enough climate scientists or hours in the day, to educate those who believe in the climate physics equivalent of "a flat earth".
A few days later
I had clearly realised by writing something so long, I had made a hostage to editorial fortune, so I have to say, all kudos to their editorial team; their abstraction is a fair reflection of what I was trying to say (and by virtue of its brevity, probably a lot more effective).
However, despite thinking brevity is a good thing, I can't quite leave it alone because I think some things still need to be said. But, I can't do justice to what I want to say either, because I don't have time ...
In the best of possible worlds I'd like to spend time stressing and explaining how unreasonable this deconstruction of the CRU email is, but Gavin has already done that (context and original), and Anne didn't buy that.
I'd like to spend some time explaining just how time consuming debating the issue is when one has to spend most of the time dealing with basic scientific issues rather than the issue of global warming. Frankly, I don't think that's a good use of tax payers money, I'm paid to do climate science, not teach high school and/or undergraduate physics. Which is not to say I don't think more communication should happen, just that I'm not the best person to do it, and neither are most of my colleagues. But she'd probably think I was in my ivory tower, rather than just trying to be practical ...
I'd like to spend some time explaining what we know, and what we think, what we mean by probability and uncertainty, and give her that rigour, but like all communication, it takes two to party, and given that she appears not to have actually herself talked to a climate scientist about the import of these emails, I'm not sure she'd listen.
I'd like to explain why "deniers" is exactly the right word for some of the "sceptic" community, and why it's not sloppy to use slang and abbreviations when the other party to your communication knows what you mean, and you have no expectation that anyone else is ever going to see what you have written. If I really thought posterity was going to judge everything I wrote, I would be much less efficient ... is that lack of efficiency worth the price? Is it really? Of course, when I do try to communicate to a wider audience, then of course it's reasonable to expect a professional effort ...
And there is much much more.
However, it's not the best of possible worlds, because I haven't the time to take on all thse conversations and simultaneously do what I'm paid to do ... this week alone, I have no time that is not already committed to meetings and actions, and it's a rather typical week.
Quality communication and education takes time, lots of it, and that's why I wish for more quality journalism ... and, if society wants it, more professional scientifically educated communicators to work in the interface between the scientific coal face and the public and their policy engine of state.
by Bryan Lawrence : 2009/11/29 : 0 trackbacks : 2 comments (permalink)
erase to transparency
The easy way (with The GIMP): somehow select an area (choose by fuzzy selector contiguous colour is one good option), then, if your image hasn't got an alpha channel, add one (right click, choose transparency, add an alpha channel) ... then color to alpha (right click, choose transparency, color to alpha).
by Bryan Lawrence : 2009/11/09 : Categories gimp : 0 comments (permalink)
Still here
I'm still here ... but my day job is all consuming at the moment. Blame CMIP5.
by Bryan Lawrence : 2009/11/09 (permalink)
my hero
Everyone needs to understand this:
A 50%-good solution that people actually have solves more problems and survives longer than a 99% solution that nobody has because it's in your lab where you're endlessly polishing the damn thing. Shipping is a feature. A really important feature. Your product must have it.
It applies to science too ... politics, you name it ...
(Citation: Joel on Software via Tim Bray).
by Bryan Lawrence : 2009/09/29 : Categories computing ndg metafor : 0 trackbacks : 0 comments (permalink)
Reading in 2009, 15-18: Teenage action fodder for summer
Even the unobservant readers of this blog will have noticed the silence by now. It's workload folks ... something has to give. I'll be back ...
Meanwhile, for your amusement, and my records, here's the first half of my summer reading ... as a snapshot.
All four of these came for 20p each at the village Fete (you'll be right if you were to suppose that I didn't just buy four books for 80p, there is a fifth, and I just finished it, but there were a few in between I need to tell you about, and I'm not ready for that yet.
The thing these four have in common is that they're more teenage boy action fodder. I chose this lot on the grounds that I expected to spend lots of the summer with only five minutes here and there to read, and it was pointless to try and read something which required anything approaching attention. I also chose them because the rest of the books on the stand sucked even more ...
Anyway, here they are, starting with three from Jack Higgins:
The Keys of Hell and the Day of Reckoning. These are the same book, oh no they're not, they've got different covers, and the characters have different names, and the country they're set in is different, and the events are different but they are the same book! As I read the second, I kept having deja vu. Surely I just read this? The word formulaic jumped into my mind and would not let go ... even James Bond is less formulaic ... oh, the plot? Well I've almost deliberately forgotten already, but basically it's superhero agent type bloke beats mafia (or equivalent) butt (in one country or another ) ... yawn. Still somehow I read them, so perhaps despite my age, there's a dose of brainless teenager in me yet, I could have stopped, but they were so easy to read in a low intensity sort of a way ...
Fortunately Thunder Point did actually have something approaching a plot, and although one of the same superhero agent dudes was in this one, somehow he came across more plausibly (or maybe my ability to suspend disbelief had been suped up by a day or two of warm sunshine) ... anyway, if you have to read one of these three, read this one - it's on safer ground for Higgins being vaguely related to WWII.
The other one was a different kettle of fish.
Bernard Cornwell's Sharpe has been fighting the battlefields of the early 19th century for a wee wihle now, on paper and on the tellie, and I've been an avid reader and collector since the beginning. Yes, this too is boys own stuff (it might seem like this is a recurring theme, but be fair, a lot of what I read isn't), but it's historically accurate and when I was a teenager I was right into the history of the Napoleonic wars, so it's fiction set in a period for which I once knew the history pretty well.
Anyway, this one was Sharpe's Fortress It's typical fare, Sharpe is mistreated by erstwhile colleagues because he's not a gentleman, makes friends with some real men, dallies with a female (incredibly peripherally for a relatively modern book), beats various foes by his own bloodymindedness and martial skill, and conquers a fortress to boot. No wonder he was on the tellie.
I must be getting older though, so while I enjoyed it for what it was, I'm getting less keen on following Sharpe, the books seem more formulaic (There's that word again), and I'm finding it harder and harder to enjoy books that are primarily about kicking butt without some more interesting twists.
Anyway, amazing what happened when I pitched up at a School Fete, and there were piles of books for 20p each isn't it? I ended up buying a bunch nearly all the same (the fifth book was very different ... but I'll leave that as a tease).
by Bryan Lawrence : 2009/09/17 : Categories books : 0 trackbacks : 1 comment (permalink)
The relationship between collecting metadata, and the optimum size of a child's plate of food
I've said it before, and no doubt I'll have to keep saying it, but the word metadata is understood by nearly every individual differently. This has a number of consequences, starting with defining (in any given) case, what comprises metadata. The problem is nicely encapsulated a recent email on the Galeon list. I hope Gerry Greager wont mind me further publicising his statement:
A lot of otherwise really sharp folks tend to define everyone's data and metadata by their own prejudices, including me. After all, MY data's easy to identify and define, and I can see how YOUR data should be identified and defined, too. What? you don't agree with me? How dare you?
The corollary of this position I'd state as:
It's really easy for me to create my metadata, and time consuming and unrewarding to create the metadata you need. How dare you ask me to waste time doing it? Oh, by the way, can you please create the metadata I need to consume your data? What, you don't want to unless you're a co-author? Bizarre!
(In this context, consumption goes well beyond merely being able to load and manipulate the data, the meaning and context matter too ...)
This has interesting consequences for those of us trying to collect metadata within projects, like, for example metafor. There, one of our goals is to document the models used and simulations produced in CMIP5. That means, we're going to be asking the modelling groups to enter metadata about those models and simulations, and it's going to be time consuming to do so. I expect many will consider it not of direct benefit to them (that said, I hope just as many, if not more, will recognise direct benefits). Indirect benefits should be obvious: the better documented we make these models and simulations, the better the interpretations and derivative science should be, particularly when those intepretations and derivations are done by those outside the normal community of model data users, by folks who need that extra metadata to be sure of what they are doing (or even to do it at all).
Ok, so I think I make a cogent argument about benefits, so where does childrens eating behaviour come in? Well, I think when one is trying to gather metadata, we're in the same boat as parents are with young children: if you put too much food on the plate, kids just dabble round the sides and don't each much. Put the right amount on the plate, and kids gobble it up. Too little, and you're back to "don't each much".
So, when asking for metadata, it's crucial to ask for just the right amount, enough for a large proportion (but not all) the potential data consumers, but not so much that the task of producing it puts off the metadata producers, and you end up getting little or none of what you need. (And don't ask so little, that you end up getting little or none of what you need.)
Returning to metafor, the question I keep asking myself is: "Should we ask that, will it put folk off answering at all?" The problem of course is, knowing what the answer is. Again, like children, we need to try and second guess how much capacity and desire there is ...
With a metadata entry tool, we're in the even more complicated situation of a children's party: we have different children (with different capacities and desires), so we have to work out the average plate size, but allow for second and third helpings for those with large capacity. That is, we need to guess how much metadata we can reasonably ask for in the average without making it look too large, but make it possible for those with the interest and desire to give us much much more information within the same structures.
The answer is further complicated by the situation. To push this analogy even further, In the case of CMIP5 we have the advantage of a peer-induced pressure (for increased metadata production), just like at that party, where there is peer-induced pressure (for increased food consumption).
by Bryan Lawrence : 2009/08/26 : Categories CMIP5 metadata metafor : 0 trackbacks : 0 comments (permalink)
contrails
Silence is golden, so this must be a golden blog ... the last month I've either been head down over the keyboard slaving on the development of a tool to collect descriptions of the models used and simulations produced for CMIP5, or on holiday. More of the latter soon.
Meanwhile, as I sat in the backyard on Sunday afternoon, supping a Grove Mill Sauvignon Blanc, my head fell backward as it is want to do, and I gazed at the contrails decorating our sky. They often decorate our sky, which means amongst other things that a) we seem to sit under the entry flightpaths for a lot of routes into Europe, and b) our sky is often clear enough to see them ... but that those two factors are not unrelated. Sometimes our sky isn't clear enough to see contrails, because it's full of clouds made up of ... dispersed contrails!
 |
Looking at them reminded me of their radiative importance, and influence on climate. The literature used to be full of stuff about it, ten years ago, but things seemed to have moved on. A couple of numbers stuck in my head though: Mannstein (1999) found in an algorithmic analysis of AVHRR derived contrails a global mean annual contrail coverage of 0.5%, with significantly higher values in some times and places (and this is itself a factor of about 50-100% less than other manual analyses). I vaguely remember figures as high as 6% for central Europe at some times (but can't track down the reference quickly). One also has to remember that these numbers have to be an underestimate, because as they spread out they become less detectable, and obviously the presence of other clouds can both mask such contrails or, indeed, be caused by them.
by Bryan Lawrence : 2009/08/25 : Categories climate : 0 trackbacks : 3 comments (permalink)