Guest Post by Willis Eschenbach
Well, the BBC, which as I understand it is an acronym for “Blindly Broadcasting Cra- ziness”, gives us its now-standard tabloid style headline, that
Climate change is ‘killing penguin chicks’ say researchers
Of course they’ve included the obligatory “awwwww-inspiring” picture, viz:
Naturally, the researchers didn’t say what the Beeb claimed. What they said was in their paper, Climate Change Increases Reproductive Failure in Magellanic Penguins, viz:
Statistical Analyses
We tested whether chick age, amount of rain, or low temperature affected a chick’s probability of dying during a storm using our 28 years of data with multiple logistic regressions.
Mmmm … testing to see whether more young chicks die in extremely cold, rainy weather … seems to me that even city kids would know the answer to that one.
In any case, how does this blinding insight into penguin mortality tie into climate? Glad you asked. It has to do with their model … or rather their models.
Figure 1. A list of the combinations of three predictor variables used in their twenty-one different models. These are used to model the odds of a penguin chick dying in a storm. The three predictor variables are age (a), amount of storm rain (r), and low minimum temperatures (l). Sadly, they did not archive their data … so this is just pretty pictures at present. Click the image to embiggen.
Their logic and observations go like this. They’ve noticed that the period during which the penguins lay their eggs has gotten longer over the last 30 years. Their hypothesis is that this will make them more vulnerable to the storms. Only thing is, how to prove it?
Why, make up a bunch of computer models of chick mortality, of course. Why not? Or as they say:
We simulated the effects of breeding synchrony on chick mortality in storms. We simulated the proportion of chicks likely to die in a storm on a given day by the hatching spread: for 13 days (the mean for 1983–1986) and 27 days (the predicted value for the early 2080s, based on an increase of 0.15 days per year; see results).
I do love the “extend a trend to infinity” logic of saying that by 2080 (or to be exact, the “early 2080s”) the Magellanic penguins will have a 27 day spread in their egg-laying dates … and using that same logic, we can be sure that by the year 2500 they will be breeding randomly throughout the year … but I digress …
So they simulated the chick deaths from storms, and then to connect that to climate change, they say:
Climate models predict that the frequency and intensity of storms will continue to increase.
Hey, that settles it for me. Since the data says there’s been a change in the length of their laying season, and since models say that the storms will kill more chicks if their laying season gets longer, and since they’ve included one sentence to establish that climate models predict more storms in the future, heck, their work is done.
It’s a beautiful chain of imaginary causation, the scientific version of the bumper sticker that says, “God said it – I believe it – That settles it!”, with “Models” in place of the Deity.
I have to say, this all seems to me like a huge waste of good data. These fine folks have done a solid, workmanlike job of collecting a very large mass of data over 28 years … but then they simply waterboarded the data until it confessed. One example of this is their choice of models.
First, while it is legit to try 21 models, at the end of that process the model you find should be pretty amazing, or else you’re just flipping coins until you get seven heads in a row and declaring victory … especially when you just keep adding parameters.
Next, they make a laudable effort to only use real-world variables in their models. For example they say:
We included all 2-way interactions except age × age squared because we did not want to include a cubic fit for age which is unlikely to have biological meaning.
I like that point of view, that the predictor variables should be real-world variables with physical or biological meaning, and age, rain, and low temperatures certainly fit the bill. Now that seems legit until you get to some of the combinations they use. For example, the model that they finally chose has the predictor variables of the following form.
A + A2 + R + A*R + A2*R + A2*L + A2*R*L
where “A” is age, “R” is rain, and “L” is low temperatures.
And that all looks logical … until we factor and simplify it, and we get
R + A (R + 1)+ A2 (L + 1) (R + 1)
So in fact, rather than the 7 variables they say they are using, in fact they are only using 5 variables:
A, R, A2, (R + 1), and (L + 1)
Unfortunately two of these variables that they are using, “rain plus one” and “low temperatures plus one”, have no conceivable physical meaning.
And that, in turn, means that their best model is actually nothing more than curve fitting using unreal, imaginary parameters without biological or physical meaning.
It is for this reason, among others, that I’m very cautious when I make models, and in general I don’t like combination additive-multiplicative models of the type they use. Yes, I’m sure that people can make an argument for using them … I’m just saying that such models make me nervous, particularly when they end up with eight or ten parameters as in their models.
Here’s the strange part for me. Since they have good data on the length of the egg laying season, and good data on storms and chick deaths, why not just use the data to actually calculate the relationship between storm-related chick deaths and the length of the egg laying season? Perhaps I missed it, but I couldn’t find that calculation in all of their work. Instead, they make a complex model of the situation for which they already have data …
I see this as another tragic casualty of the ongoing climate hysteria. But I suppose I’m just being idealistic, and I’m overlooking the fact that in this current insane situation, it’s much easier to get funding if you say “Hey, I’m not just studying a bunch of birds that are too dumb to remember how to fly, I’m doing vital work on the climate crisis! Think of the grandchildren!” …
Finally, despite their whizbang model, I strongly doubt the researchers’ conclusion that the change in the length of the breeding season will lead to more chick deaths. Natural species survive in part because their methods of living and eating and giving birth are flexible, and they are able to change them in response to changing circumstances in such a way as to increase their odds of survival. The idea that the penguins are changing their breeding habits in the direction of communal suicide seems like … well, like an unusual claim that would require supporting evidence that is much more solid than a computer model with imaginary parameters to make me believe it.
Ah, well … onwards, ever onwards …
w.
N.B: If you disagree with me, please quote EXACTLY what it was that I said that you disagree with. A claim that I don’t know what I’m doing, or that I’m just wrong, or that I should go back to school, any of that kind of vague handwaving goes nowhere because I don’t have a clue what has you (perhaps correctly) upset … you could be right and no one will ever know it. So quote what you object to, that way we can all understand what you are referring to.
Gareth Phillips:
I write to support DS in his post at January 31, 2014 at 9:31 am.
Please note that the Government has to local outrage at the floods by agreeing to reinstate the needed dredging in the Somerset Levels. The floods have nothing to do with changed rainfall.
I addressed this on another WUWT thread and I copy that post to here to save you needing to find it.
Richard
=========
richardscourtney says:
January 27, 2014 at 10:56 am
Friends:
I write to provide a reminder – especially to non-Brits – of the clear message which needs to be presented to politicians and is provided by the flooding of the Somerset Levels.
The Levels were a swamp that was completely flooded most of the year except for a few, small islands. Indeed, it was by hiding in the levels that Alfred the Great ended up burning the cakes because searching the reed-covered marsh was impossible.
The Napoleonic Wars provided a need for additional grain and one response was to drain the Levels to obtain additional farmland. This conversion of the swamp to agricultural land was conducted in the period 1770 to 1833, and this paper describes it.
The drainage and water management are relatively recent and entirely man-made. The Levels will return to being a flooded swamp in the absence of proper maintenance and operation of the drainage and water management. So, the people who live on the levels KNOW they will be flooded if that proper maintenance and operation ceases. And they know the necessary dredging of the watercourses has been stopped.
Arguments about climate change and conservation are rejected by people who know their homes will be destroyed unless the dredging is conducted to ensure operation of the man-made drainage.
People who are faced with real threats to their homes and lives will reject politicians who use political scares as an excuse to ignore the real threats.
Richard
RomanM says:
January 31, 2014 at 8:18 am
Ah, very good, Roman, I was wrong on the question of the mean and SD when the numbers contain a stack of zeros …
Thanks as always for your insights. RomanM, for those who don’t recognize the name, is a man with very strong statistics-fu, one whose statistical advice and insights I always pay close attention to. For example, I hadn’t thought through all of the implications of their standardizing the data before using it, and as Roman points out there are some subtle problems with that as well.
For me, the depressing part is what Roman says, that
and
I had great hopes for PLoS when they came on the scene. But unless they start requiring their authors to post up their data and code, they’re just another part of what seems to be the Marvel Comics group of “scientific” journals that includes Science and Nature magazines …
Again, Roman, my thanks for your correcting my error above.
w.
AP says:
January 31, 2014 at 2:42 am
Interesting thought, AP. I saw no mention of that in the paper. Unfortunately, as RomanM said above,
w.
Frank de Jong says:
January 31, 2014 at 4:46 am
Thanks for the clarification, Frank. You raise an interesting point. However, I still don’t like the procedure, because of the units involved. Remember that they have normalized the temperature, meaning that it is no longer in degrees C, but in units of “standard deviations”.
I fail to see how adding standard deviations to degrees gives you anything real. Yes, if they did it before normalization, I could see it might possibly be right, because then you’re adding degrees to degrees as in your example above, and that has a real-world meaning.
But adding standard deviations to degrees? That just makes me nervous.
My best to you, and thanks for coming back to make your meaning clear.
w.
I note the location at Punta Tombo, at 44 degress S. In the northern hemisphere, this is like Eugene OR or Rapid City SD in the west or Portland Maine in the east.
I have lived the summers in southern Argentina, much more to the south, in the sheep ranching (pre-Mt. Hudson eruption in Chilean Andes) area between Comodoro Rivadavia (@45 S) and San Julian, in Santa Cruz Pvca., also on Argentina’s coast. It was down right Hot most of the time. Almost all of the time the wind was downright fierce, clocked at 160+ kms/hr on the worst days in Comodoro aeropuerto.
Post Mt Hudson eruption the foxes had no sheep to eat, so perhaps had reason to feast on Penguinos. It was estimated that Mt Hudson killed 30 million sheep in Santa Cruz Pvca. All the sheept estancias were bankrupted. My other suggestion is that the guanacos or the nandus or the Patagonian hare ate the penguins (just joking), but most likely they were bothered to death by the turistas.
David Boleneus says: @ur momisugly January 31, 2014 at 1:23 pm
Thanks for the local view point. The lost of the sheep would certainly send what predators survived out looking for other food. Like The Russian squirrel pack that killed a dog
Gail Combs says:
January 31, 2014 at 1:44 pm
True: The loss of the last of the sheep would require the the last of the predators to find other prey lest they themselves become the last food of the last lost-seeking late food-finders ….
Willis –
re: Original post and follow-up comments:
The subject paper of your post, “ Counting your Chicks Before They Hatch” is a paper that involves a number of complementary hierarchical statistical analyses that appear to involve both standard linear regression (e.g., Figure 6) and multi-variable logistic regressions. It appears that some of the analyses may have been performed using statistical categorical variables along with normalized (Z) continuous variables (for age, rain, etc).
I don’t have much recent experience with logistic regression, but I did use this method in the 80’s to analyze fatigue failure in superalloys based on defect sizes and location (surface vs internal), etc. Logistic regression uses mortality or failure (life) probability regressed against a set of predictor variables (either continuous or categorical). As such your normal understanding of variance (or stdev) may not apply in the way you think. Indeed, if you do a search on the original article, you will only find two instances of the phrase, “Standard Deviation”.
I have read this paper carefully several time and while I can’t claim to have fully grasped the entire presentation, I believe the statistical analyses were probably done properly . . . the authors used commercial (Stata2) software.
I do believe you are irresponsible for indicting a paper that you don’t appear to understand; and . . . yes you like specifics.
(1) Your factorization of the authors’ proposed variables (both first order and interactions terms) shows you don’t understand the standard linear model; furthermore your factored version alters the factors that the authors thought were important and in any event would be non-workable outside a non-linear scheme.
(2) You misinterpreted the effect of variables on mortality (e.g., X% +/- 1.5 X%) as being wrong. . . no the uncertainty of the effect is sufficiently large that it could have a larger (+) effect at one extreme (increasing mortality) or a lesser (-) effect at the other extreme (decreasing mortality).
(3) You original post and subsequent comments suggest you really don’t know anything about logistic regression.
I’ve enjoyed, and learned from many of your posts over the several years that I’ve been a WUWT devoted follower. But, I do believe that sometimes you react too quickly to errors in papers you review, without fully discriminating between what do you know versus what you don’t know.
Best regard
Dan Backman
I am a bit baffled. did I get this right? It is the time they choose to lay eggs that has changed a bit not the time it takes to get this done correct?
If this is the case then one would have to assume that they would have better survival since not as many of their offspring will be at their most vulnerable at the same time. To conclude that this makes them less safe implies there are less issues at the original range of dates. Unless it wasnt covered here that is not a case they made in any way so the obvious conclusion is that having a wider range of dates that individual chicks would be vulnerable would be better not worse.
The idea that having a wider range of dates opens them up to more storms, is obviously countered by the fact that having a wider of dates covered by various individuals also opens them up to more days without storms while at their most vulnerable point. Is this really not completely obvious?
Dan, thanks for your detailed comments. My reply follows:
DanMet’al says:
January 31, 2014 at 5:55 pm
Mmmm … Dan, I’m not sure how familiar you are with climate science papers. In any case, I fear that the fact that they used commercial statistics software increases rather than decreases the odds that there are problems.
I believe my factorization of the terms is absolutely legitimate. I note that in RomanM’s analysis above, he found no problem with it. However, he did point out further problems I hadn’t thought of regarding the standardization process … you sure you want to defend this penguin paper’s statistics against RomanM?
And I assure you, I understand the standard linear model quite well. I can give you the underlying math for how to do the analysis in whole or in part. I’ve done hundreds and hundreds of them, both using commercial statistics software, as well as using functions for linear modeling that I’d written ab initio, in order to be sure that I understood them.
Next, if you understand math then you’ll understand that factoring changes nothing. It makes absolutely no difference whether I write
(x – y)^2
or whether I write
x^2 – 2xy + y^2
This is one of the problems with just picking random combinations of interactions of factors. You can end up using combinations which sound reasonable but have no physical meaning … as in this case.
You are correct, I was wrong in that … however this was my attempt in a comment to further understand what they’d done. It had nothing to do with the head post, which stands on its own.
Didn’t you say just a few lines above that you were going to be specific? That is totally and completely vague. It’s nothing but mudslinging based on your uncited, unsupported belief. Sorry … that just loses you points.
Thanks, Dan. Let me cut to the chase here.
They’ve collected a bunch of great data. Then, rather than analyzing the data, they used the data to make a model. In all, their model has a total of 8 parameters.
Then they did a linear extension of a 28-year trend in egg-laying to the year 2080, to get their claimed egg-laying period at that point.
Then they applied their whiz-bang model to the results of the 28-year trend that they’d extended out 66 years, and proclaimed their conclusions.
So that was their method … are you telling me that you are defending that on the basis that they used commercial statistical software? Even if their statistics were impeccable, and they’re assuredly not, are you defending that analysis?
Next, I could likely show many more errors. But as RomanM lamented above, after pointing out the statistical errors that he could see:
and
They did not archive either their code or their data. As a result, their “scientific” study is nothing of the sort. Neither you nor I can replicate it, check it for errors, see if their logic is correct, examine their code for bugs, or see if their data actually supports their claims.
And as a friend of mine said … that’s not science, that’s just advertising.
Are you defending that?
Next, what they didn’t do, and what they have to do, is to give us the null hypothesis, and then falsify it. In my opinion, the null hypothesis is that the lengthening of the egg-laying period is an adaptation that will increase their chances of survival, or at worst has no effect on survival. After all, that’s what evolution does. Successful creatures respond to changes in their environment in such a way as to increase their chances of survival … and with over a million Magellanic penguins on the planet, which breed in a variety of locations, my null hypothesis is that the Magellanic penguins know what they’re doing.
Now, to falsify that null hypothesis, the authors need to throw away their model and just analyze their data. I don’t care what their model says. What does the data say? My guess is that what they’d find is that their dataset is too short (only 28 seasons) to determine the answer to the question. I say that in part because storm deaths are a low probability event (9%), so you need more years to get good statistics. But like I said … we don’t have the data, so it’s just advertising.
Finally, surely you are not defending their claim that global warming will increase the number of storms. That is the basis of their whole paper, and they don’t even try to justify it … but even the IPCC doesn’t believe that nonsense any more. The planet has been warming for well over a century, and in all that time, there has been no increase in any type of weather extremes.
SUMMARY:
• They have extended a short-term trend to 2080. That alone disqualifies the study as science.
• They have made a hugely complex model (8 parameters). They have not tested the model out-of-sample, or if so, I can’t find it in their paper …
• Despite not being tested for even five years out-of-sample, they are applying their model to the putative 2080 conditions.
• They have claimed that there will be more storms as a result of global warming, when there is no evidence of that anywhere, nor is it supported by the IPCC AR5.
• The authors have not archived either their data as used or their code as used. As a result, there is no way to replicate their study, which makes it just an anecdote, not science in any form.
Dan, I said the paper was bogus. Whether my factorization of the statistics is correct makes absolutely no difference to that conclusion, and I stand by it strongly.
w.
Willis Eschenbach said January 30, 2014 at 12:28 pm
If nobody posts serious scientific objections to some piece of “B.S.” (to use your term) that’s being passed off as real science, lots of people believe it
Thanks for the ink
I expect you did a great job of smashing the Penguin Chick story. (I didn’t read it) And it must be satisfying to have the talent to do so, but the way to win at Whac-A-Mole is to pull the plug on the machine.
The paper says, “Young chicks between 9 and 23 days old were particularly vulnerable to hypothermia, as they were too young to have fully grown their waterproof plumage but already too big to seek shelter under their parents’ bodies”
This UMICH website says differently: “The young are constantly cared for and brooded for 24 to 29 days, after which the parents spend extended periods of time foraging and will return to the nest every 1 to 3 days.” http://animaldiversity.ummz.umich.edu/accounts/Spheniscus_magellanicus/
In terms of counting your Penguins before they are hatched, there seems to be a major problem: The paper says there are roughly 400,000 penguins, this paper from 2008, Boersma, says differently, http://depts.washington.edu/turbopen/wordpress/wp-content/uploads/2013/08/B580707.pdf 2008
The largest breeding colony of Patagonian (Magellanic) penguins, at Punta Tombo, Argentina, had approximately 200,000 breeding pairs in October 2006—a decline of 22% since 1987.
Looks like they are doing well these days! Even better if you read the Argentine tourist board site: http://www.patagonia-argentina.com/en/punta-tombo,
“Punta Tombo is the most important Magellan Penguins colony within continental Patagonia. More than a million and a half of them arrive at this place every year in order to breed.”
There are even more here: http://www.welcomeargentina.com/puertomadryn/punta-tombo-reserve.html
“Saying that about two million Magellan penguins were expecting us is a sweet lie; the truth is that they gather there from September to April in order to nest, mate, incubate their eggs and feed their offspring, presenting a show unique on the continent.”
Of course they are selling the area to tourists, but the people who did the study were paid to sell disaster.
“The team noted that not all rainstorms killed the chicks. Of the 233 storms that occurred over the course of the study period, only 16 resulted in chick deaths.”
So there was an average of just over 8.5 storms per year, less than 7% of those storms killed chicks, but that becomes an ecological portent of doom brought on by climate change.
Steve Case says:
January 30, 2014 at 1:17 am
Steve Case says:
February 1, 2014 at 2:10 am
Steve, when you figure out how to pull the plug on climate alarmism, I’m your man … I’ve never figured it out, so I just do what I can. You seem to think that doing what I can is somehow bad and wrong … but since you haven’t found the plug to the whack-a-mole machine, what do you suggest doing?
I suppose I could follow your path, and go around insulting and ragging on those people who are actually doing something while we wait for Steve to find the plug and pull it, but somehow, that doesn’t seem too productive …
w.
Willis Eschenbach, I appreciate what you do. Please keep it up.
But I have to say something else as I can’t just be a cheerleader (I haven’t got the legs).
Perhaps pulling the plug isn’t the right idea. Perhaps improving the standard of Climate Science is the right idea – as you are trying to do. So my question is “how can improvements be built into Climate Science?”
May I suggest some basic rules for peer review that need to be ticked off before a paper is accepted. things like:
•All model variables to have some clearly defined meaning in the real world.
•All combinations of variables to be factored and simplified to check rule 1.
•Release the data.
•Refer to the real world observations to check the model.
And that’s just what I got from this paper.
Such a list would drive up standards, in my opinion, and you have the platform that could start the discussion.
“. We’re talking about the southern coast of Argentina, right? Violent weather and torrential downpours? Cape Horn?? Tierra Del Fuego? Named 3 or 4 hundred years ago. That place? One of the stormiest, most unsettled weather areas on the planet. Seriously?”
No, we’re talking about the coast of Patagonia, in the rain-shadow of the Andes with a desert climate and approximately the same annual precipitation as Phoenix, Arizona.
It seems to me to be a case of too much education. I believe my lack of knowledge of statistics may allow me to see what is involved here more clearly. I won’t comment on synchrony, only the author’s assertion that “events [storms] appear to be increasing as climate changes, further stressing the population.” In fact, one might conclude that just the opposite is happening.
The link between climate change and chick mortality claimed by the authors is difficult to establish because no climate data is given. In order for change we must establish a base period against which to compare. The base could be an average of a period within the study, or a period immediatley precdeeding the study or some other period. The authors do not do this. We are left with a few scraps of information to use.
The authors cite Haylock MR, Peterson TC, Alves LM, Ambrizzi T, Anunciação YMT, et al. (2006) Trends in total and extreme South American rainfall in 1960–2000 and links with sea surface temperature. J Climate 19: 1490–1512. doi: 10.1175/jcli3695.1 as their source that change has occurred:
At the Trelew airport weather station (43° 12′ S, 65° 16′ W), about 90 km north of Punta Tombo, precipitation in storms became heavier: the amount of precipitation from wet days (days with at least 1 mm of precipitation), the number of consecutive wet days, the number of days with at least 20 mm of precipitation, and the percentage of total precipitation from days with more than the 99th percentile of rain all increased [32].
In this paper it is noted that records exist from 1957 for Trelew. One of the goals of the conference was to develop standards for data collection. “Fifty-four stations were deemed to be of sufficiently
high quality and to have sufficient observations to be used to assess changes for the period 1960–2000.” I am not qualified to judge this paper but I do not think the data or conclusions relative to the Punta Tombo area are complete and convincing.
What is not shown is (and I am assuming the TuTiempo.net database is the same as the one used for this paper) that the earlier (1957-1973) records are of little value.
1957-58 do not appear to inlcude precipitation detail.
1959-64 no data
1965 is only partial information
1966 no data
1967 starts in April with many dates blank
1968 No data Jan through March
1969-1972 no data
1973 No precipition indicated in January through May period.
1974-Current Good records with few dates lacking detail.
Data concerning weather conditions was accumulated daily according to the authors. We do not have that data. Because of the proximity it is reasonable to conclude that any climatic change at Punta Tombo would be reflected at Trelew. Since we are comparing change, I suggest looking at the period from 1974 when decent records begin through 1982, the period prior to the beginning of this study. And then comparing to the most recent 5 year period.
A few points allow us to narrow the data to review. It is worth noting that the vast majority of days in the summer period get no rain. The birds arrive from Brazil in September/October and eggs hatch in November and early December according to the authors. The birds leave Punta Tombo by March. No chicks died from the effects of precipitation beyond 44 days and chicks ” were most likely to die in rainstorms when they were between 9 and 23 days of age…Seven chicks older than 30 days died in storms, all in nests that had running flood water.” Therefore, only rain in the months November, December and January is relevant.
At Punta Tombo, we collected weather data daily, usually before 0800 h. We recorded precipitation (±0.1 mm) using a manually-emptied plastic rain gauge, and minimum and maximum temperatures (±1°C) using a minimum-maximum recording thermometer for the previous 24 hours. We defined a storm as a period of consecutive days with measurable rain, ranging from one to six days (165 of 233 storms lasted one day and 50 lasted 2 days; only 18 lasted more than 2 days). For storms lasting more than one day, we added the rain for the consecutive days.
and
From 1983–2010, 206 known-aged chicks (8% of 2482 chicks alive during a storm) died in or after storms. No chicks died in 217 of the 233 storms (93%). Sixteen storms (14 in December, 2 in November) in 13 of the 28 years killed between <1% and 70% of the chicks. One chick died in a storm with only 1.2 mm of rain (with a low temperature of 3°C), but 97.6% of chicks killed experienced storms with at least 10 mm of rain.
and
Precipitation ranged from 0.1 to 142 mm per storm with a mean of 98.8 (±51.3) mm of rain during the breeding season and low temperatures ranged from 1° to 18°C. Storms that killed chicks averaged 29.4±36.1 mm of rain (N = 16 storms). Storms when no chicks died averaged 7.3±13.8 mm (N = 217 storms; t231 = 5.3, P<0.0001).
The key metric is the number of storms of 10mm or greater between November 1 and January 31 and then to determing if there been an increase in frequency of occurrence. Here is what I have found:
1974 none
1975 3 storms (53.09mm Dec, 24.13mm and 18.04mm Jan)
1976 2 storms (28.10mm and 39.39mm Dec)
1977 1 storm (20.84mm Dec)
1978 none
1979 3 storms (10.16mm and 13.97mm and 10.67mm, Nov)
1980 4 storms (13.97mm Nov, 12.95mm Dec, 10.16 and 12.45mm Jan)
1981 1 storm (131.06mm Jan)
1982 none
***
2009 none
2010 none
2011 2 storms (135.89mm Dec, 10.92 Jan)
2012 2 storms (11.19mm Dec, 12.95 Jan)
2013 none (note: Data for Jan 2014 from wunderground.com)
Comment: I guess a bar graph would help demonstrate the pont.
What about temperature data?
For the period 1974-1982
Mean Average Temperature (sum of the mean avg temp of 3 months divided by 3):
19.04 (ranging from 17.97 to 20.03)
Mean Maximum Temperature (sum of the mean max temp of 3 months divided by 3):
26.72 (ranging from 24.90 to 28.87)
Mean Minimum Temperature (sum of the mean min temp of the 3 months divided by 3):
12.08 (ranging from 11.00 to 12.90)
For the period 2009-2013
Mean Average Temperature (sum of the mean avg temp of 3 months divided by 3):
19.83 (ranging from 19.37 to 20.93)
Mean Maximum Temperature (sum of the mean max temp of 3 months divided by 3):
28.03 (ranging from 26.73 to 29.87)
Mean Minimum Temperature (sum of the mean min temp of the 3 months divided by 3):
12.79 (ranging from 12.43 to 13.30)
Notes: Prior to 1982 temps are listed as whole numbers only. 2011 was exceptionally hot, reaching 30 degrees on 45 of 92 days. Dec 1976 appears to have been remarkably cold. ( I am keeping it simple and did not weigh for the discrepancy in days per month. It is an indicative number only anyway.)
What about precipitation? Total reported precipiation for Nov, Dec and Jan for each year.
1974 23.13mm
1975 112.54mm
1976 83.36mm
1977 43.96mm
1978 11.70mm
1979 41.93mm
1980 60.98mm
1981 175.51mm
1982 15.46mm
average per year: 63.17mm
2009 27.69mm
2010 16.24mm
2011 163.57mm
2012 46.22mm
2013 .80mm
average per year: 50.90mm
Notes: A storm in 1981 dumped 131.06mm and one in 2011 dumped 135.89mm. Argentina is suffering a serious drought currently.
Days with precipitation:
The days with ANY measured precipitation, even less than 0.1 mm, are are recorded as averaging 20.22 from 1974 through 1982 (range of 13 to 34) and 7.8 from 2009 through 2013 (range of 5 to 10).
In the It's Worse Than We Thought Category:
Climate models predict that a storm of 40 mm that was expected to occur at Punta Tombo every 20 years in the late 20th century, will occur every 7–15 years by 2081–2100 [7].
Data from Trelew:
12/20/75 53.09
1/27/82 131.06
1/2/84 80.01
1/2/85 100.08
1/10-12/91 99.07
12/10/06 117.09
11/15/09 131.06
12/28/11 135.89
Statistical analysis is wonderful. Of course, starting with good data is important. I am not so sure much earlier data is "clean".
In a BioScience article from 2008 P. Dee Boersma relates that penguins arrived in Punta Tombo in the 1920s, grew rapidly and that the population of the colony peaked at some point around four decades ago.
As FAH stated above the "paper appears to be a classical example of “overfitting.” What occurs here is that there is a huge of amount of weather data. Precipitation is indicated for less than 20% of dates. Most of those amounts are in tenths of a mm. A very few dates show significant rain of 10 mm or more. A tiny fraction achieve 40mm, but, of that fraction most are 3 to 5+ inches. Occasionally it will rain a lot and when it does chicks will likley die. No regression analysis needed. No indication of AGW climate change shown.
The Trelew 40+mm storm table in my post above contains incorrect dates. I apologize for the error. Thank you for allowing me to post here. It should read:
From the paper: “From 1983–2010, 206 known-aged chicks (8% of 2482 chicks alive during a storm) died in or after storms. No chicks died in 217 of the 233 storms (93%).” (my emphasis). Most of the 206 storm casualties took place in just 2 years, 1991 and 1999 (fig 2a). Starvation and predation were far more significant dangers to the penguin chicks in most years.
It seems to that there is a very interesting paper to be written (using the same data) on the remarkable resilience of penguin chicks to rainstorms.