Just a quick note of clarification. What is (financial) capital? And what is money? And which part of money is cash?

'Capital' in this context means net worth - i.e. the difference between the assets of a bank and the liabilities of the bank. I appreciate this might be confusing since it is an entirely different definition than physical capital - factories and such like.

'Cash': more precisely, 'Monetary Base' or M0. Cash or deposits of the Bank of England that can be freely convertible into cash. I refer to 'monetary base' colloquially as 'cash', because that is what it is!

'Money': 'Broad money' or M3. This definition contains bank deposits as well as 'monetary base'. Bank deposits are liabilities on the banks to pay the depositor cash on demand. I refer to 'broad money' colloquially as 'money'. Tim Joslin calls deposits 'electronic money' as far as I am aware.

N.B. Banks have two requirements governing their operation: they must have enough cash and they must have enough capital in order to lend.

I have come to the following draft conclusions:
a) The banking system is short of capital and it can't get it from joe public when the future outlook (and government involvement) is so uncertain
b) The non-banking system is short of cash, and can't get that cash from the banks without getting into more debt.

The banking system can be recapitalized by creating New Banks (see earlier posts), not recapitalising the old banks.

The non-banking system can get more cash but how? Printing money and some non-house inflation might be a good option.

What is a good carbon tax? Well a good start would be $100/tonne of CO2.
What does a carbon tax of $100/tonne CO2 (10c/kgCO2) actually mean??

Let's relate it to some other quantities:
Petrol has a 2.3kg of CO2/litre or 10.47kg CO2/gallon
So the tax adds 23c to a litre of petrol or $1.05 to a gallon of 'gas'

Crude oil has a carbon content of 0.43 metric tons CO2/barrel [http://www.epa.gov/grnpower/pubs/calcmeth.htm]. So the tax adds $43 per barrel of oil.

In electricity, the average carbon footprint is around 0.5kg CO2/kWh (0.4kg for gas 1.0kg for coal). So this tax adds 4c/kWh to gas electricity and about 10c/kWh to coal electricity.

How much revenue would this tax gather? If we each emit 20tonnes CO2 (US data) then this would make $2000 per person per year. This money could be spent by the government, used to reduce other taxes or used to reduce the government deficit. If that falls to 10tonnes, there would be $1000 per person per year.

If we say have a £50/tCO2 tax, this would at present make £500 per person, £500x60m=£30,000m=£30bn in total.

By comparison, here are the tax takes in 2006/7 (From http://www.hm-treasury.gov.uk/d/bud08_chapterc.pdf )
Income tax: £147.8bn NIC: £87.3bn VAT: £77.4bn Corporation tax £44.8bn

So a carbon tax at this level could replace approximately 40% of VAT.

This post just follows up from the last where I compared a bank to a supermarket which retains it's earnings. There is an important difference between the two, which helps to clarify the point that has been made.

As far as I can see, the difference between an earnings retained Sainsburys (if excess profit-as-cash isn't given back to shareholders but retained, say in a vault) and the desert-island bank - is that Sainsburys sucks in cash; whereas the desert island bank has an outstanding /obligation to receive cash/ which may or not be covered by any 'real' cash existing in the system (it is conceivable that there is not enough cash & deposits in the system to substitute for cash). Bank deposits /can /substitute for cash (until all the deposits are used up) but in this thought-experiment world, there can be a retained obligation for the guy to supply something (cash) which he can't get from anywhere since it does not exist outside the banks and the central banks. In that case there will be large amount of unrequited demand for cash (the debt obligation).

There appears to be only a few ways that cash can be got into the non-bank system, to supply the outstanding obligation. The main options are if the cash in the bank or central bank is used to buy up real assets. Presumably if the bank(-owners) have market power then they would be able to enforce as low cash-price for the assets in exchange for the outstanding debts (in other words there will be a fall in asset prices). In other words, the removal of the outstanding debt obligation /does not have to be /in the form of a default; it could be that the bank invests in property directly; or that the earnings are distributed to shareholders, who themselves invest in property. However, the bankruptcy choice is /more likely/ if asset values fall (and if the decision rest primarily with the debtor) because bankruptcy transforms the obligation to hand over the now-low-value asset rather than a costly cash-obligation.

Thus deleveraging will cause a fall in asset values and either
a) a settlement of the outstanding debt due to the banks or the shareholders of the bank physically buying real assets in exchange for cash
b) default and the exact same process taking place, except that the banks get less for their money

in the same way that in an upturn the reverse is true (leveraging is accompanied by increasing asset prices and low default rates).

I'm not sure of the real world relevance of this primordial example but it does perhaps emphasize that giving banks cash in exchange for assets will not solve the problem, because the key is the cash shortage in the real economy and not within the banks.

A common argument* (to the effect that banking is a Ponzi scheme) is considered below**

[* Note: The argument considered is the one (imperfectly expressed) below. There may be other arguments claiming that banking in general, or in certain circumstances, is a Ponzi scheme, which may be true, but are not considered here.]

[** Second Note: I'm no longer sure that the argument is false. Although 'Ponzi scheme' is pretty imprecise word. The aspect I'm not sure about is that banks may have a monopoly over 'created' money and there may be the potential for a sort of 'short squeeze' on bank deposits/cash.]

When the rate of interest charged by banks on loans is greater than that charged on deposits, does banking constitute some sort of Ponzi scheme? Here is the argument:

Imagine a desert island consisting of one guy and a bank.

There is a certain amount of cash in a desert island economy (lets say £100) and it belongs to the bank. This is helpful because then the bank needs capital and liquidity to make loans and the guy wants some money. Let's say that banks need to have both capital (net assets) of £100 and liquidity (cash) of £100 in order to loan £1000 (both a liquidity requirement of 10% and a capital requirement of 10%).

So the bank on a desert island has £100 in cash. Through the wonders of fractional reserve electronic banking it can lend someone £1000 as an electronic deposit at an interest rate of (say) 10% per year. So long the guy doesn't want to withdraw this money in cash (aha the bank has only £100 in cash, so that would be impossible!) then the bank can create a deposit of £1000 and a outstanding loan of £1000. Let's say the deposit pays 5% per year from the bank to the guy and the loan pays 10% per year from the guy to the bank.

This guy is stupid and does nothing with his deposits. After a year, the deposit in the bank expands to £1050 and the loan outstanding amount expands to £1100.

The bank wants the principal plus interest back (£1100). But the guy has only £1050 of 'money' (deposits). In fact, since in our thought-experiment world, outside the bank there is only £1050 of money in total in the economy! So when all the debts are canceled by the deposits the bank is still owed £50. Money conservation (so the argument goes) implies that there is not enough money to pay back the bank, because amounts outstanding on the loan increase faster than outstanding amounts on the deposits!

Hence, banking requires an ever increasing amount of new money created to pay back the old.
So someone has to default. More debt is required to keep the thing going!! A Ponzi scheme!!

This is a very simple argument. However, is it true?

Is it true for any profitable institution? Say a supermarket? A supermarket makes profit - more money comes in than goes out. What does it do with the profit? Is it recycled, or do all the bank notes always end up with the supermarket? Usually the profit is either distributed to shareholders or kept as assets on the supermarket vault. Does any profitable institution eventually suck the whole economy dry of money? Or is that money usually given back to shareholders or used to buy real assets?

Is it the same with the bank? Can't the desert island bank just pay back it's shareholders, or buy some real estate. In this example, the profits from the rest of the economy are ploughed back to the bank, and the bank gets richer.

So in the original situation, the bank uses it's £50 profit to buy the guy's garage (in the open market), paying off his debt. Then there are no outstanding debts. Of course, since the bank made some profit at the small guys expense, the garage ownership got transferred from the small guy to the bank. The bank has ended up one garage richer and the guy has ended up one garage poorer. The bank makes the new asset cancel the existing debt obligation.

Now let's say that the man is poor, with no garages or other assets to sell. In that case he goes bankrupt and defaults because he does not have anything to pay back the bank. The bank has to write off the asset and the bank loses the excess asset that it thought it had.

So what happens depends on whether there are assets to transfer in exchange for the debt. But there is no 'money conservation problem'. There is only a default problem if the poor guy has nothing to sell to the bank, in which case the poor chap is bust. Otherwise the bank gets more stuff (not money).


----------------------

Let's introduce growth into the economy.

Let's now say that the guy is a builder; he started off with one house with garage; however this time he does not just sit on his electronic money. This builder instead transfers £1000 in electronic money (deposits) to a brick supplier who supplies bricks and the builder builds a second house plus two sheds. He sells the second house (plus shed) to the brick supplier for £1050, leaving him with a second shed worth £50.

What is the financial situation?
The brick supplier starts with zero, and then received £1000 of deposits (in exchange for the bricks). This grows into £1050 of deposits (in the bank), and then is paid back to the builder leaving £0.

The builder starts with £0, gets a £1000 deposits from the bank in exchange for £1000 of debt. He pays £1000 of deposits to the brick supplier. After 1 year he receives £1050 of deposits back from the brick supplier. He then has £1050 of deposits and £1100 of debt. He uses the £1050 deposits to pay off £1050 of debt, leaving himself with £50 of debt (ie obligations to the bank). He then sells the second shed to the bank for £50 in cash. He deposits the £50 cash at the bank, paying off the deposit. The final situation is exactly the same as before, except that the bank now has a shed.

This is productive economic growth and is not a problem, albeit one where the product of the growth goes to the bank in exchange for finance.

There is no Ponzi scheme. There is a certain amount of financialization - transfer of assets to the financial sector, but no shortage of money.

The mistake is to see the obligations to the bank - and in particular the remaining obligation £50 to the bank as being a bit of 'negative money'. It is not. It is an obligation to the bank. This obligation can be paid back with any sort of asset, not necessarily a bank deposit or cash. The bank creates a net obligation to itself of £50 through it's hard-nosed practice of charging more interest on its loans than it donates on its deposit (and - we might add - it has a lot of sometimes free implicit government support). But it does not induce a money shortage. It just achieves, at the end, more of the assets in the economy to itself in exchange for being profitable.

How much solar could be got from the Indian desert?

Well, the 'Great Indian Desert' is 200,000km2. [ http://en.wikipedia.org/wiki/Thar_Desert ]
Let's conservatively assume 10W/m2 (10MW/km2) http://www.withouthotair.com That makes 2000GW or 1kW for 2billion people. So India can power it's existing energy consumption from the Thar desert, even with 2billion people ; but it can't power an american (10kW/person) level of consumption from this source.

Just thought I'd do some calculations specifically about the Gobi desert.

[The tone of this post has changed slightly from "wow china has a lot of people" to "wow there's lots of space in the Gobi, we can power the world from up there"; to "wow it's bitingly cold up there".... :)]

The Gobi is Northerly, high up and is dry in Summer. It's cold and snow-laden in winter however. It has an area of around 500,000 square miles (1,300,000 square km). [Source: http://tinyurl.com/d2t37m ]

[ http://en.wikipedia.org/wiki/Gobi_desert ]
Can we do Solar here? The temperature in winter seems to be cold. Around -15 to -25 Celsius. There might be a problem with water pipes involved in Concentrated Solar Power freezing. Even the mean annual temperature seems to be quite low - around zero e.g. at Sivantse (-2.5C) or Ulaanbaatar (3C).

So it's not clear that we can do CSP on the Gobi plateau. Maybe we should ask a (Chinese) engineer? You could try using anti-freeze as a coolant but mirrors and pipes are probably going to be unpleasant and expensive to erect in biting -20C to -30C temperatures and strong Siberian winds.

Using (hopefully cold-resistant) photovoltaics might be a better plan.
[ http://www.withouthotair.com ]
Let's say (generously) they achieve 5W/m2 [Similar to Bavarian Solar Park]

Let's say that 1million square km can be used. Then the Gobi can provide 5,000GW. That's about one third of total world energy consumption and 5kW for 1 billion people.

But it's still a bit chilly up there. And PV is likely to be expensive, (remember you've got to pay for the systems integration, not just the panels).

What about wind? 2W/m2, (assume only 0.5 million square km - if it's mountainous or even hilly then the valleys are less useful), 1000GW = 1kW per person for a billion people. Maybe not the whole story, but a significant input to China's energy consumption. It's high up in UlanBator and very windy.

Surely therefore wind (cheap; can be done now) is the one to go for? Wind is notoriously intermittent but my guess is that the Gobi is big enough and continental wind currents reliable enough to give some assurity.

My conclusion from this post is:

a) If renewables are reasonably feasible on the Gobi then they could power a large fraction (wind) or all (solar) of China's energy consumption.
b) However, the economic costs and engineering difficulties in the Gobi should not be underestimated, specifically the Siberian conditions (freezing temperatures, biting winds, windchill in the -40s permafrost?).
c) Wind seems the best option (can deal with mountainous terrain without huge costs), and (by carpetting every hill and plateau in the Gobi) could make a contribution of about 1000GW to China's energy needs
d) If Solar is possible up there, then Solar can power all of China, but I'm sceptical about the total systems costs of doing so at present,

It will be difficult for China and India to live off their own renewables

David Mackay has written in detail about the question 'Can Britain live off its own renewables' in his new book (http://www.withouthotair.com).

The broad answer to this question is that to make an impact renewable energy sources need to be 'country sized' and Britain does not really have 'country sized' areas to devote to energy generation, except perhaps offshore in the Atlantic or in foreign far-flung deserts (and even then there remain formidable technical, political and economic barriers).

Renewable energy sources have a power density around 1-5Watts per square metre. To achieve 5000Watts per person requires1000-5000-square metres per person. To power the entire population requires 50,000-300,000km2 of land (between one fifth and all the land in Britain).

Here are some interesting figures for other countries:

Country__Population__Area (km2)_Density(km-2)_Land pp (m2)
UK_______60,776,238_____242,900___________246________4000
China_1,323,324,000___9,596,961___________138________7250
India_1,103,371,000___3,287,263___________336________3000
----------------------------------------------------------------
USA_____301,140,000___9,629,091____________31_______32000
World_6,733,164,238_148,940,000____________45_______22000

Broadly speaking, this suggests that it will be difficult for China and India to live off their own renewables. Asia is where the future of the global economy (and of future emissions growth) lies. Of course, we see that for US (and in the future for the world in total) this constraint is not as important; but this shouldn't distract us from the need to find a solution that can be adopted in Asia too.

This the view that renewables are the whole answer may be somewhat US centric.
For reasonably local alternatives to coal in USA, China and India it may be that 'renewables' may be only part of the answer. Achieving a cost of electricity lower than coal must be our number one objective, but fundamentally we need an energy source that will suit India and China too.

A

Just a quick update about my favorite formats to use!

I've previously posted about the joys of (free, open source, easy to use) OpenOffice Writer over the dreaded (expensive, commercial, horrible) Microsoft Word; and the joys of (free, open source, easy to use) Zotero over the dreaded (expensive, commercial, horrible) Endnote.

However, the .doc format is, however, ubiquitous and if you start passing round .odt people get upset. But you don't need to use Microsoft Word in order to save in Word format! Open Office can save in Microsoft Word format (and PDF, Latex and MediaWiki for that matter)! You can also still use Zotero.

The particular referencing convention I tend to use is Author-Date (Harvard Reference Format 1) which stores the references as Endnotes, and I format using Bookmarks. Bookmarks are the format for use in word documents.

Comments: Zotero has an annoying habit of going back to Reference Marks when you click the set document preferences button. You need to ensure you stay on bookmarks if you do this. Zotero is fairly stable, but occasionally there are problems. It's best to backup your work regularly anyway.

A

This post is about the motivation for this blog. So it's an important post. Perhaps this post should have come at the start. Anyway, diving in... what are the key facts of the matter about climate change?

Firstly, we know there is an important natural greenhouse effect on Earth. The sun, which is hot (at an effective temperature of 5000C) emits electromagnetic radiation at a high energy (high frequency; low wavelength - mostly in the visible and ultra-violet spectra). The Earth, which is warm (about 14C) re-emits electromagnetic radiation mostly at low energies (in the infrared spectra). Some of this infra red radiation is absorbed by certain gases in the atmosphere (water vapour and carbon dioxide) and then re-radiated. Half of this re-radiating heat radiation goes back down to earth, leading to a higher equilibrium surface temperature. We can easily calculate what the temperature of the earth would be without greenhouse gases - about minus 15Celsius. So we know that the temperature of the Earth is about 30Celsius higher than it would otherwise be, due to the effect of these gases. It seems that water vapour and carbon dioxide are the two most important of these gases, although other gases such as methane are also important.

Water vapour is the most important component of the greenhouse effect, but the concentration of water vapour in the atmosphere depends on temperature. The second most important gas carbon dioxide had an atmospheric concentration of about 280 parts per million by volume (ppmv) before human industrialisation (and after the ice ages where it dropped to 180 ppmv).

The most basic climate model w0uld suggest that a 100% effective greenhouse would raise (absolute) temperatures by 2^(1/4) or about 20%*250K=50 Celsius. This can be compared to the observed temperature increase of about 30 Celsius.

Simple models (similar to those of Arrhenius) suggest that the doubling of carbon dioxide concentrations from 280 to 560ppmv would raise global average temperatures by approximately 5C. Complex (General Circulation) models, and an analysis of the forcings in the ice ages (see previous post) suggest an average temperature rise of 3C for a doubling of CO2 concentrations, but with a probability range between about 1.2C - 6C (although there is non-zero chance of temperature rises above the maximum in this range).

Whilst there are arguments why the temperature might be less than the 3C, there are also arguments why it might be higher than this. How do observations of the whole world compare?

Next post I will discuss the observational record and what the estimates of climate sensitivity mean for us.

This week has been dominated by questions of replication and of what standards are required to serve the interests of transparency and/or science (not necessarily the same thing). Possibly a recent example of replication would be helpful in showing up some of the real (as opposed to manufactured) issues that arise. The paper I'll discuss is one of mine, but in keeping with our usual stricture against too much pro-domo writing, I won't discuss the substance of the paper (though of course readers are welcome to read it themselves). Instead, I'll focus on the two separate replication efforts I undertook in order to do the analysis. The paper in question is Schmidt (2009, IJoC), and it revisits two papers published in recent years purporting to show that economic activity is contaminating the surface temperature records - specifically de Laat and Maurellis (2006) and McKitrick and Michaels (2007).

Both of these papers were based on analyses of publicly available data - the EDGAR gridded CO2 emissions, UAH MSU-TLT (5.0) and HadCRUT2 in the first paper, UAH MSU-TLT, CRUTEM2v and an eclectic mix of economic indicators in the second. In the first paper (dLM06), no supplementary data were placed online, while the second (MM07) placed the specific data used in the analysis online along with an application-specific script for the calculations. In dLM06 a new method of analysis was presented (though a modification of their earlier work), while MM07 used standard multiple regression techniques. Between them these papers and their replication touch on almost all of the issues raised in recent posts and comments.

Data-as-used vs. pointers to online resources

MM07 posted their data-as-used, and since those data were drawn from dozens of different sources (GDP, Coal use, population etc. as well as temperature), trends calculated and then gridded, recreating this data from scratch would have been difficult to say the least. Thus I relied on their data collation in my own analysis. However, this means that the economic data and their processing were not independently replicated. Depending on what one is looking at this might or might not be an issue (and it wasn't for me).

On the other hand, dLM06 provided no data-as-used, making do with pointers to the online servers for the three principle data sets they used. Unlike for MM07, the preprocessing of their data for their analysis was straightforward - the data were already gridded, and the only required step was regridding to a specific resolution (from 1ºx1º online to 5ºx5º in the analysis). However, since the data used were not archived, the text in the paper had to be relied upon to explain exactly what data were used. It turns out that the EDGAR emissions are disaggregated into multiple source types, and the language in the paper wasn't explicit about precisely which source types were included. This was apparent when the total emissions I came up with differed with the number given in the paper. A quick email to the author resolved the issue since they hadn't included aircraft, shipping or biomass sources in their total. This made sense, and did not affect the calculations materially.

Data updates

In all of the data used, there are ongoing updates to the raw data. For the temperature records, there are variations over time in the processing algorithms (satellites as well as surface stations), for emissions and economic data, updates in reporting or estimation, and in all cases the correction of errors is an ongoing process. Since my interest was in how robust the analyses were, I spent some time reprocessing the updated datasets. This involved downloading the EDGAR3 data, the latest UAH MSU numbers, the latest CRUTEM2/HadCRU2v numbers, and alternative versions of the same (such as the RSS MSU data, HadCRUT3v, GISTEMP). In many cases, these updates are in different formats, have different 'masks' and required specific and unique processing steps. Given the complexity of (and my unfamiliarity with) of economic data, I did not attempt to update that, or even ascertain whether updates had occurred.

In these two papers then, we have two of the main problems often alluded to. It is next-to-impossible to recreate exactly the calculation used in dLM07 since the data sets have changed in the meantime. However, since my scientific interest is in what their analysis says about the real world, any conclusion that was not robust to that level of minor adjustment would not have been interesting. By redoing their calculations with the current data, or with different analyses of analogous data, it is very easy to see that there is no such dependency, and thus reproducing their exact calculation becomes moot. In the MM07 case, it is very difficult for someone coming from the climate side to test the robustness of their analysis to updates in economic data and so that wasn't done. Thus while we have the potential for an exact replication, we are no wiser about its robustness to possibly important factors. I however was able to easily test the robustness of their calculations to changes in the satellite data source (RSS vs. UAH) or to updates in the surface temperature products.

Processing

MM07 used an apparently widespread statistics program called STATA and archived a script for all of their calculations. While this might have been useful for someone familiar with this proprietary software, it is next to useless for someone who doesn't have access to it. STATA scripts are extremely high level, implying they are easy to code and use, but since the underlying code in the routines is not visible or public, they provide no means by which to translate the exact steps taken into a different programming language or environment. However, the calculations mainly consisted of multiple linear regressions which is a standard technique, and so other packages are relatively easily available. I'm an old-school fortran programmer (I know, I know), and so I downloaded a fortran package that appeared to have the same functionality and adapted it to my needs. Someone using Matlab or R could have done something very similar. It was a simple matter to then check that the coefficients from my calculation and that in MM07 were practically the same and that there was a one-to-one match in the nominal significance (which was also calculated differently). This also provides a validation of the STATA routines (which I'm sure everyone was concerned about).

The processing in dLM06 was described plainly in their paper. The idea is to define area masks as a function of the emissions data and calculate the average trend - two methods were presented (averaging over the area then calculating the trend, or calculating the trends and averaging them over the area). With complete data these methods are equivalent, but not quite when there is missing data, though the uncertainties in the trend are more straightforward in the first case. It was pretty easy to code this up myself so I did. Turns out that the method used in dLM07 was not the one they said, but again, having coded both, it is easy to test whether that was important (it isn't).

Replication

Given the data from various sources, my own codes for the processing steps, I did a few test cases to show that I was getting basically the same results in the same circumstances as was reported in the original papers. That worked out fine. Had their been any further issues at this point, I would have sent out a couple of emails, but this was not necessary. Jos de Laat had helpfully replied to two previous questions (concerning what was included in the emissions and the method used for the average trend), and I'm sure he or the other authors involved would have been happy to clarify anything else that might have come up.

Are we done? Not in the least.

Science

Much of the conversation concerning replication often appears to be based on the idea that a large fraction of scientific errors, or incorrect conclusions or problematic results are the result of errors in coding or analysis. The idealised implication being, that if we could just eliminate coding errors, then science would be much more error free. While there are undoubtedly individual cases where this has been the case (this protein folding code for instance), the vast majority of papers that turn out to be wrong, or non-robust are because of incorrect basic assumptions, overestimates of the power of a test, some wishful thinking, or a failure to take account of other important processes (It might be a good idea for someone to tally this in a quantitative way - any ideas for how that might be done?).

In the cases here, the issues that I thought worth exploring from a scientific point of view were not whether the arithmetic was correct, but whether the conclusions drawn from the analyses were. To test that I varied the data sources, the time periods used, the importance of spatial auto-correlation on the effective numbers of degree of freedom, and most importantly, I looked at how these methodologies stacked up in numerical laboratories (GCM model runs) where I knew the answer already. That was the bulk of the work and where all the science lies - the replication of the previous analyses was merely a means to an end. You can read the paper to see how that all worked out (actually even the abstract might be enough).

Bottom line

Despite minor errors in the printed description of what was done and no online code or data, my replication of the dLM07 analysis and it's application to new situations was more thorough than I was able to do with MM07 despite their more complete online materials. Precisely because I recreated the essential tools myself, I was able to explore the sensitivity of the dLM07 results to all of the factors I thought important. While I did replicate the MM07 analysis, the fact that I was dependent on their initial economic data collation means that some potentially important sensitivities did not get explored. In neither case was replication trivial, though neither was it particularly arduous. In both cases there was enough information to scientifically replicate the results despite very different approaches to archiving. I consider that both sets of authors clearly met their responsibilities to the scientific community to have their work be reproducible.

However, the bigger point is that reproducibility of an analysis does not imply correctness of the conclusions. This is something that many scientists clearly appreciate, and probably lies at the bottom of the community's slow uptake of online archiving standards since they mostly aren't necessary for demonstrating scientific robustness (as in these cases for instance). In some sense, it is a good solution to a unimportant problem. For non-scientists, this point of view is not necessarily shared, and there is often an explicit link made between any flaw in a code or description however minor and the dismissal of a result. However, it is not until the "does it matter?" question has been fully answered that any conclusion is warranted. The unsatisfying part of many online replication attempts is that this question is rarely explored.

To conclude? Ease of replicability does not correlate to the quality of the scientific result.

And oh yes, the supplemental data for my paper are available here.

Share This

To start with, I'll ask a question, one suggested by friend of mine:

Why be alarmed about climate change?

There is a question about this question. Is this the right question to ask? 'Alarmed' is an emotional word, imbued with connotations of panic. Even in the case of a major war; it is likely that public address announcments would ask us not to be alarmed, but instead to behave calmly and rationally. In other words, even in the event of a likely catastophe, alarm might be an inappropriate response. The argumentative strategy of the 'straw man' (or 'paper tiger') is to attack a opponent position which is falsely weak.

'Alarm' (or it's cognate 'alarmism') is also a word which is used as a group label in climate change debating circles and labels are notoriously divisive. So instead I'll ask a different question:

Why be concerned about climate change?

Now it's possible that my choice of words is also viewed as political and contentious. I could try to ask a purely scientific question. But I'm looking for a choice of words that hints at emotion (in the sense of moving someone) without being itself emotive.

However, I think the question could be contrued as being a leading one. So I could ask the question 'should we be concerned about climate change?'. This feels like it is a leading question in the direction of thew answer 'no'. It also has the contentious word 'should'. However, it is at least roughly neutral, but could be made more so by replacing 'should' by a question of 'justification'. So this is where I shall start my next discussion.

Is concern about climate change justified?

But this question could be answered 'of course - we should have some concern about climate change', but the level of concern would be a long way down the list of those concerns which are viewed as relevant. So perhaps a better question is:

What level of concern about climate change is justified?

I think this is a useful question. It's still emotional to some extent; using the word 'concern' is a personal rather than an objective measure; however 'justified' is an objective sounding word.

The difference between a single calculation and a solid paper in the technical literature is vast. A good paper examines a question from multiple angles and find ways to assess the robustness of its conclusions to all sorts of possible sources of error — in input data, in assumptions, and even occasionally in programming. If a conclusion is robust over as much of this as can be tested (and the good peer reviewers generally insist that this be shown), then the paper is likely to last the test of time. Although science proceeds by making use of the work that others have done before, it is not based on the assumption that everything that went before is correct. It is precisely because that there is always the possibility of errors that so much is based on 'balance of evidence' arguments' that are mutually reinforcing.

So it is with the Steig et al paper published last week. Their conclusions that West Antarctica is warming quite strongly and that even Antarctica as a whole is warming since 1957 (the start of systematic measurements) were based on extending the long term manned weather station data (42 stations) using two different methodologies (RegEM and PCA) to interpolate to undersampled regions using correlations from two independent data sources (satellite AVHRR and the Automated Weather Stations (AWS) ), and validations based on subsets of the stations (15 vs 42 of them) etc. The answers in each of these cases are pretty much the same; thus the issues that undoubtedly exist (and that were raised in the paper) — with satellite data only being valid on clear days, with the spottiness of the AWS data, with the fundamental limits of the long term manned weather station data itself - aren't that important to the basic conclusion.

One quick point about the reconstruction methodology. These methods are designed to fill in missing data points using as much information as possible concerning how the existing data at that point connects to the data that exists elsewhere. To give a simple example, if one station gave readings that were always the average of two other stations when it was working, then a good estimate of the value at that station when it wasn't working, would simply be the average of the two other stations. Thus it is always the missing data points that are reconstructed; the process doesn't affect the original input data.

This paper clearly increased the scrutiny of the various Antarctic data sources, and indeed the week, errors were found in the record from the AWS sites 'Harry' (West Antarctica) and 'Racer Rock' (Antarctic Peninsula) stored at the SCAR READER database. (There was a coincidental typo in the listing of Harry's location in Table S2 in the supplemental information to the paper, but a trivial examination of the online resources — or the paper itself, in which Harry is shown in the correct location (Fig. S4b) — would have indicated that this was indeed only a typo). Those errors have now been fixed by the database managers at the British Antarctic Survey.

Naturally, people are interested on what affect these corrections will have on the analysis of the Steig et al paper. But before we get to that, we can think about some 'Bayesian priors'. Specifically, given that the results using the satellite data (the main reconstruction and source of the Nature cover image) were very similar to that using the AWS data, it is highly unlikely that a single station revision will have much of an effect on the conclusions (and clearly none at all on the main reconstruction which didn't use AWS data). Additionally, the quality of the AWS data, particularly any trends, has been frequently questioned. The main issue is that since they are automatic and not manned, individual stations can be buried in snow, drift with the ice, fall over etc. and not be immediately fixed. Thus one of the tests Steig et al. did was a variation of the AWS reconstruction that detrended the AWS data before using them - any trend in the reconstruction would then come solely from the higher quality manned weather stations. The nature of the error in the Harry data record gave an erroneous positive trend, but this wouldn't have affected the trend in the AWS-detrended based reconstruction.

Given all of the above, the Bayesian prior would therefore lean towards the expectation that the data corrections will not have much effect.

The trends in the AWS reconstruction in the paper are shown above. This is for the full period 1957-2006 and the dots are scaled a little smaller than they were in the paper for clarity. The biggest dot (on the Peninsula) represents about 0.5ºC/dec. The difference that you get if you use detrended data is shown next.

As we anticipated, the detrending the Harry data affects the reconstruction at Harry itself (the big blue dot in West Antarctica) reducing the trend there to about 0.2°C/dec, but there is no other significant effect (a couple of stations on the Antarctica Peninsula show small differences). (Note the scale change from the preceding figure — the blue dot represents a change of 0.2ºC/dec).

Now that we know that the trend (and much of the data) at Harry was in fact erroneous, it's useful to see what happens when you don't use Harry at all. The differences with the original results (at each of the other points) are almost undetectable. (Same scale as immediately above; if the scale in the first figure were used, you couldn't see the dots at all!).

In summary, speculation that the erroneous trend at Harry was the basis of the Antarctic temperature trends reported by Steig et al. is completely specious, and could have been dismissed by even a cursory reading of the paper.

However, we are not yet done. There was erroneous input data used in the AWS reconstruction part of the study, and so it's important to know what impact the corrections will have. Eric managed to do some of the preliminary tests on his way to the airport for his Antarctic sojourn and the trend results are as follows:

There is a big difference at Harry of course - a reduction of the trend by about half, and an increase of the trend at Racer Rock (the error there had given an erroneous cooling), but the other points are pretty much unaffected. The differences in the mean trends for Antarctica, or WAIS are very small (around 0.01ºC/decade), and the resulting new reconstruction is actually in slightly better agreement with the satellite-based reconstruction than before (which is pleasing of course).

Bayes wins again! Or should that be Laplace?

Update (6/Feb/09):The corrected AWS-based reconstruction is now available. Note that the main satellite-based reconstruction is unaffected by any issues with the AWS stations since it did not use them.

Share This

Roger Few at the University of East Anglia is leading a new network focussing on the interconnections between environment and human health in the context of poverty and development faced by poorer or more marginalized regions and communities around the world