I find headlines like this frustrating and potentially misleading. It provides zero context. Do we know that this is abnormal? Forty years is a short period of time in the grand scheme of things. Six times more than what?
"The team was able to discern that between 1979 and 1990, Antarctica shed an average of 40 gigatons of ice mass annually. (A gigaton is 1 billion tons.) From 2009 to 2017, about 252 gigatons per year were lost."
I'm a climate skeptic. This type of research is why I'm a climate skeptic. 252 gigatons per year means a bit less than 0.001% of the mass of the ice sheet of the Antarctic continent. Or, to put it differently, about 1.8cm decrease in average ice height, when the current average is about 1.9km. They are telling me they have a method to estimate an average decrease of 1.8cm? Really? How can anyone claim such a thing with a straight face? Civilian GPS doesn't have this type of vertical resolution. You can achieve better resolution by the law of large numbers, however that is at the rate of the square root of the number of observations, and I doubt there are a tremendous number of observation points in Antarctica.
Most of these ice mass measurements are made using GRACE not physical height / GPS measurements. Ice mass loss appears to be accelerating by multiple forms of measurement.
Fine, it's GRACE. So GRACE has an accuracy of 0.001%? Really? Or rather, is there any estimation technique to measure the ice mass of a whole continent with accuracy of one thousandth of one percent?
Wait, actually they claim their accuracy is much higher, sometimes more precise than 0.0002% or even 0.0001% (I kid you not).
Here's the wiki link for GRACE [1]. It states that during 2003-2013 the ice loss was 67±44 Gt per year. That ±44 Gt represents ±0.00017%
Anyway, here's the actual abstract of the nature article [2]
"The Antarctic Ice Sheet is an important indicator of climate change and driver of sea-level rise. Here we combine satellite observations of its changing volume, flow and gravitational attraction with modelling of its surface mass balance to show that it lost 2,720 ± 1,390 billion tonnes of ice between 1992 and 2017, which corresponds to an increase in mean sea level of 7.6 ± 3.9 millimetres (errors are one standard deviation). Over this period, ocean-driven melting has caused rates of ice loss from West Antarctica to increase from 53 ± 29 billion to 159 ± 26 billion tonnes per year; ice-shelf collapse has increased the rate of ice loss from the Antarctic Peninsula from 7 ± 13 billion to 33 ± 16 billion tonnes per year. We find large variations in and among model estimates of surface mass balance and glacial isostatic adjustment for East Antarctica, with its average rate of mass gain over the period 1992–2017 (5 ± 46 billion tonnes per year) being the least certain."
For those not familiar, East Antarctica is much bigger than West Antarctica, so it's natural for any measurement to be least certain. However, take a look at the prior estimates from the GRACE studies [3]
"An early analysis of GRACE-based studies data indicated that the EAIS was losing mass at a rate of 57 billion tonnes per year and that the total Antarctic ice sheet (including WAIS, and EAIS coastal areas) was losing mass at a rate of 152 cubic kilometers (c. 139 billion tonnes) per year.[4] A more recent estimate published in November 2012 and based on the GRACE data as well as on an improved glacial isostatic adjustment model indicates that East Antarctica actually gained mass from 2002 to 2010 at a rate of 60 ± 13 Gt/y."
This ± 13 Gt accuracy listed here is well below 0.0001%
Yeah, but that same study showed that the rate of gains was slowing and more recent work by the same people using the same processes show it is now shedding mass.
> Fine, it's GRACE. So GRACE has an accuracy of 0.001%? Really? Or rather, is there any estimation technique to measure the ice mass of a whole continent with accuracy of one thousandth of one percent?
If some areas are losing/gaining meters then your 1.8cm average is irrelevant as it's not the resolution they're looking at.
Your argument is misleading because it assumes uniform ice loss across the continent, it's a straw man you create and then destroy. Time to turn that skepticism inwards.
>Your argument is misleading because it assumes uniform ice loss across the continent
I'm not assuming that. "Average" and "uniform" are two different things. I never said anything about "uniform".
Take a look at the ice loss visualization [1] from NASA's website. On the west side there are ice losses as high as 300m (dark red), on the east side some ice gains of about 100m (light blue). All those losses and gains are estimated with some instruments. If you are in one of the areas where the ice loss was estimated to be about 100m, you look at your initial elevation, final elevation and take the difference. Will you elevation difference be -100m, or -98m? There will be some measurement uncertainty. Overall, uncertainties from different instruments tend to cancel out, that't the law of large numbers. Overall, I doubt we have that many numbers (or so little individual measurement uncertainty to begin with) to end up with a final measurement uncertainty that is so small.
Take a look at article [2] about the estimation of precipitations in Antarctica. It was published in March 2018 in the journal "Polar Science" owned by Elsevier, which as far as I can tell does not publish junk articles. The general tone is that there are large uncertainties in the estimation of precipitation in Antarctica.
"The study of Antarctic precipitation has attracted a lot of attention recently. The reliability of climate models in simulating Antarctic precipitation, however, is still debatable"
"The current method of data collection relies on measurement with limited temporal basis, with distances between measurements exceeding 1300 km (Knuth et al., 2010). This makes instrumental-based measurement of precipitation in Antarctic highly unreliable (Genthon et al., 2003)."
Of course, such an article does not get much press. How would one report it? "New research shows large uncertainties in our understanding of the mass balance in Antarctica" ?
The 0.00001 accuracy argument you are making seems like a red herring.
If I measure myself on a scale, do you discredit the measurement because the scale produces a value that is some fractional, fractional, fractional percent of the Earth's mass?
Why are you talking about Earth's mass? We are talking about Antarctica here. If I weight myself on a scale that has 0.2 pound precision (like the one I have), and then somehow I tell people I lost 0.002 pounds, some people will be quite skeptical of my claims. In the case of Antarctica, how is one supposed to quantify the accuracy of its net ice loss? By comparing it with the mass of an elephant of a blue whale [1] ?
It's a red herring because the total quantity of ice in the Antarctic is irrelevant, both to how the loss rate is measured and the climate significance.
It's like looking at someone who's hair is falling out and saying "look, they're barely losing any weight at all!"
GPS is not part of the measurement suite in OP, but better than 1.8 cm vertical resolution from GPS has been possible since the late 1990s, using after the fact orbit determination. It is now routine to get sub-cm vertical accuracy in GPS measurements [1, fig. 3, magenta line]. The horizontal displacements are known to about a mm.
These displacements are known so accurately that you can readily measure groundwater withdrawals, the extra mass related to of high atmospheric pressure over the GPS sensor (daily cadence), and post-glacial rebound (decadal cadence).
Your remarks below about the accuracy of the GRACE gravimetric measurement are also misguided. You can’t just choose a baseline mass and say it’s impossible to measure a change in that mass to some percent accuracy. The GRACE measurement depends on how well you can measure a distance between two spacecraft, and how well you can eliminate systematics from that difference. It does not depend on a relative mass.
Anyway, the paper uses a different technique to get ice fluxes. If you don’t know how they did the measurement, why are you so sure it is wrong?
Not clear how this ad-hominem enhances your argument.
>If you don’t know how they did the measurement, why are you so sure it is wrong?
I'm not "so sure", I'm just skeptical. I find it hard to believe that one can estimate the net ice loss over a huge and dynamic body of ice with an accuracy better than 0.001%, using any type of measurements conceivable.
You are saying that somehow the whole mass of Antarctica is irrelevant as a base mass to calculate accuracy? Why so? The ice naturally compresses and flows and this happens at continental scale. A large iceberg calving on the western coast of the continent may be offset by a relatively modest ice accumulation on the much larger eastern side. However, that accumulation on the eastern side will be conflated by the general compression that will lead to some ice being pushed towards the edges, etc, etc. There's no natural way to take a part of the continent out of the equation and say "well, we are interested only in this subset".
It was actually a fact, not an ad hominem. You clearly know less than you think you do about GPS. Ironically, it seems like the mark of a contrarian to assume that people are attacking you when they aren't and a lot of climate skeptics come off sounding that way to me.
Furthermore, it seems you're still just launching your own arbitrary theories into the discussion and not responding to anything specifically wrong with the article.
> You are saying that somehow the whole mass of Antarctica is irrelevant as a base mass to calculate accuracy...
Yes, that is correct. The mass of Antartica, whatever that might mean, is not required in itself. Percent error in estimating that is not relevant. Remember that we are interested in mass changes only.
The GRACE measurement is based on explaining the relative position of a pair of orbiting spacecraft by mass changes in 4500 "mascons" -- equal-area spherical caps that blanket the globe (see figs 2 and 3 of [1]). The fundamental observation is the range and relative velocity of the pair of spacecraft, which can be very accurately determined. The mass within each mascon has an analytically-known relationship to these observations (eqs. 8, 9 of [1] - relating mass sigma to acceleration a).
It's a monthly measurement -- you accumulate a month's worth of orbits, and solve a least-squares problem to fit the range-rate data with the masses. This can also be viewed as maximum a posteriori estimation in the conventional way. See equation 13 of [1].
I hope this clarifies why it does not matter what the ice underneath is doing. The ice has mass, and therefore it affects the gravitational potential that the satellites operate in.
You are correct to be surprised. The measurement is surprising, and it has been revolutionary. Hundreds of papers have been published using it, and it has received scrutiny and undergone improvements for more than a decade. Three independent groups (JPL, UTexas, DLR) have worked on the full retrieval over this time. Lots of what we know about groundwater withdrawals, ice sheets, and more recently deep ocean currents, is based on this measurement.
There is an attribution issue at ice/water boundaries (sec. 6 of [1]). It is due to mixed pixels -- a mascon that is part water, part ice. The mass change observed at the mascon should not be spread evenly over the whole pixel. It is split in constrained way based on errors and priors. But as [1] explains, this is not an "order of magnitude" type error, it's just a correction.
Thanks for taking your time to explain all this. It's quite cool. I spent some time with your link and also with the wikipedia page about GRACE. This is quite an ingenious experiment. First I thought the two satellites were at different altitudes, one in LEO and one in, let's say GSO. The high altitude satellite perceives the Earth more or less like a mass point, while the low altitude satellite is much more sensitive to the distribution of mass, especially at the Earth's surface. You could in principle use several satellites at different altitudes, and do some sort of "triangulations". This way, you could potentially learn about the mass distribution deep inside the Earth. Anyway, the actual way it's done is also cool, a lump of mass, like a mountain pulls faster the leading satellite, Jerry, and the second satellite, Tom, lags a little behind. Later on, Tom goes over the same mass and catches on.
That being said, the complexity of this model sounds absolutely humongous. Complex models are exciting for mathy guys to work with, but are inevitably subject to more model risk. As coders say, the number of bugs is roughly proportional with the number of lines of code.
You're going to say that this numerical calculation is bug free. A numerical calculation with 4 thousand mass element (mascons) and distance measurements probably in the billions, and lots and lots of approximations and contributions from the sun and the moon and the planets. Of course you are not going to say it is bug free. Should we trust ice loss calculations coming from GRACE? To be honest, after I read your explanation and all other things about GRACE, I'm less skeptical. I don't fully trust it, but I don't fully distrust it either.
But as you said, this is not that relevant, since the actual result published was not based on GRACE, but rather on ice flow models. Now these models I take with quite a large grain of salt. Why? Models produce estimates, and the estimates generally include an uncertainty level. The uncertainty reflects only how the uncertainty in inputs propagates to the uncertainty in the final result, not the model uncertainty per se.
Take a look at the figure 1 in [1]. It shows the precipitation in Antarctica over 2.5 decades based on several models. The average number (labeled "multi-model mean") shows very little variability from year to year, and you could infer that very little uncertainty as well. However, this is just an artifact of taking the average of several not very highly correlated time series. The model uncertainty is better reflected by the wide range of the numbers produced by the different models. If you look at this range, it's about 200mm/y, which is 10 times the net ice loss we are talking about here.
Do you see why I'm a bit skeptical?
But anyway, thanks again for taking your time with me. I learned something cool today.
I appreciate the reply. I happen to know some of the GRACE team, so I'm personally predisposed to respect their results. But, it's also significant evidence that the retrieval was re-implemented by 2 other teams, one US, one German, with similar results. All 3 datasets are available to the community for comparison.
About fig. 1 of [1]: Of the spectrum of important Earth system models, precipitation is the least well-described. For good reasons: highly non-gaussian, very local in nature, highly dependent on nonlinear condensation/temperature properties. Even in the continental US, precipitation is not well-described. Your remark about deciding on a bogus "uncertainty" of the mean precipitation in fig. 1 is absolutely correct.
But it's of no use to be "skeptical about models" based on poor results of one class of models. The whole reason that paper was published is to draw attention to that fact! They literally say this in their conclusion. ("Most CMIP5 and reanalysis models are unable to simulate a consistent spatial and temporal precipitation pattern for the Antarctic.")
I have come to believe that skepticism is somewhat empty as a value. That's why you have experts who know what the state of the art is, and what to be skeptical about.
> But it's of no use to be "skeptical about models" based on poor results of one class of models
This is not just "one class of models". This is the gain side of the gain-loss equation needed for the estimation of ice balance in Antarctica. If that side has an uncertainty 10 times higher than the stated result, then how can I not be skeptical?
>That's why you have experts who know what the state of the art is, and what to be skeptical about.
"Trust the experts" lead to the 2008 financial crisis. "Trust the experts" was what string theorists were telling us until Lee Smolin exposed them. "Trust the experts" is ultimately what religion is.
>skepticism is somewhat empty as a value
That's a surprising (for me at least) point of view for a scientist.
I'm quite sure you wanted so say something else. Something like in your experience, those who describe themselves as climate skeptics (I mean in what other context do people declare themselves skeptics? I'm a skeptic about the dark matter as well, by the way) are simply obstructionists, or ignorants, or have a political agenda or something else. Since we are all more efficient forming Bayesian priors to work in life with, your Bayesian prior in this case is that "skepticism is an empty value". And that's quite fair. I happen to agree that a large part of "climate skeptics" are quite toxic. The way the Trump administration treats the EPA, or science in general seems quite deranged. But that doesn't mean you can't be genuinely skeptic. In this case skepticism has nothing to do with values.
I simply approach this debate and I use my own heuristics. I don't know any climate scientist to use heuristics such as human character, you seem you do. For me, I apply other heuristics, based on my day to day work with models. I reached other conclusions so far than those that you reached.
> we expect multi-meter sea level rise from Antarctica in the coming centuries.”
1) Nonsense modeling. 20 year predictions from the 1990's turned out to be laughable, now they're predicting 200 years ahead. How about they try 2 year models, maybe they could get those right?
2) Can't see why I should care about someone's beach property in 2428. Even if their models were correct, which ch they aren't.
Most people live paycheck to paycheck and these lunatics want to tax and spend their hard earned money to prevent fantasy scenarios. Truly appalling.
23 comments
[ 0.16 ms ] story [ 59.6 ms ] threadI'm a climate skeptic. This type of research is why I'm a climate skeptic. 252 gigatons per year means a bit less than 0.001% of the mass of the ice sheet of the Antarctic continent. Or, to put it differently, about 1.8cm decrease in average ice height, when the current average is about 1.9km. They are telling me they have a method to estimate an average decrease of 1.8cm? Really? How can anyone claim such a thing with a straight face? Civilian GPS doesn't have this type of vertical resolution. You can achieve better resolution by the law of large numbers, however that is at the rate of the square root of the number of observations, and I doubt there are a tremendous number of observation points in Antarctica.
https://www.nasa.gov/mission_pages/Grace/index.html
Wait, actually they claim their accuracy is much higher, sometimes more precise than 0.0002% or even 0.0001% (I kid you not).
Here's the wiki link for GRACE [1]. It states that during 2003-2013 the ice loss was 67±44 Gt per year. That ±44 Gt represents ±0.00017%
Anyway, here's the actual abstract of the nature article [2]
"The Antarctic Ice Sheet is an important indicator of climate change and driver of sea-level rise. Here we combine satellite observations of its changing volume, flow and gravitational attraction with modelling of its surface mass balance to show that it lost 2,720 ± 1,390 billion tonnes of ice between 1992 and 2017, which corresponds to an increase in mean sea level of 7.6 ± 3.9 millimetres (errors are one standard deviation). Over this period, ocean-driven melting has caused rates of ice loss from West Antarctica to increase from 53 ± 29 billion to 159 ± 26 billion tonnes per year; ice-shelf collapse has increased the rate of ice loss from the Antarctic Peninsula from 7 ± 13 billion to 33 ± 16 billion tonnes per year. We find large variations in and among model estimates of surface mass balance and glacial isostatic adjustment for East Antarctica, with its average rate of mass gain over the period 1992–2017 (5 ± 46 billion tonnes per year) being the least certain."
For those not familiar, East Antarctica is much bigger than West Antarctica, so it's natural for any measurement to be least certain. However, take a look at the prior estimates from the GRACE studies [3]
"An early analysis of GRACE-based studies data indicated that the EAIS was losing mass at a rate of 57 billion tonnes per year and that the total Antarctic ice sheet (including WAIS, and EAIS coastal areas) was losing mass at a rate of 152 cubic kilometers (c. 139 billion tonnes) per year.[4] A more recent estimate published in November 2012 and based on the GRACE data as well as on an improved glacial isostatic adjustment model indicates that East Antarctica actually gained mass from 2002 to 2010 at a rate of 60 ± 13 Gt/y."
This ± 13 Gt accuracy listed here is well below 0.0001%
[1] https://en.wikipedia.org/wiki/Gravity_Recovery_and_Climate_E...
[2]https://www.nature.com/articles/s41586-018-0179-y
[3]https://en.wikipedia.org/wiki/East_Antarctic_Ice_Sheet
If some areas are losing/gaining meters then your 1.8cm average is irrelevant as it's not the resolution they're looking at.
Your argument is misleading because it assumes uniform ice loss across the continent, it's a straw man you create and then destroy. Time to turn that skepticism inwards.
I'm not assuming that. "Average" and "uniform" are two different things. I never said anything about "uniform".
Take a look at the ice loss visualization [1] from NASA's website. On the west side there are ice losses as high as 300m (dark red), on the east side some ice gains of about 100m (light blue). All those losses and gains are estimated with some instruments. If you are in one of the areas where the ice loss was estimated to be about 100m, you look at your initial elevation, final elevation and take the difference. Will you elevation difference be -100m, or -98m? There will be some measurement uncertainty. Overall, uncertainties from different instruments tend to cancel out, that't the law of large numbers. Overall, I doubt we have that many numbers (or so little individual measurement uncertainty to begin with) to end up with a final measurement uncertainty that is so small.
Take a look at article [2] about the estimation of precipitations in Antarctica. It was published in March 2018 in the journal "Polar Science" owned by Elsevier, which as far as I can tell does not publish junk articles. The general tone is that there are large uncertainties in the estimation of precipitation in Antarctica.
"The study of Antarctic precipitation has attracted a lot of attention recently. The reliability of climate models in simulating Antarctic precipitation, however, is still debatable"
"The current method of data collection relies on measurement with limited temporal basis, with distances between measurements exceeding 1300 km (Knuth et al., 2010). This makes instrumental-based measurement of precipitation in Antarctic highly unreliable (Genthon et al., 2003)."
Of course, such an article does not get much press. How would one report it? "New research shows large uncertainties in our understanding of the mass balance in Antarctica" ?
[1] https://svs.gsfc.nasa.gov/30880
[2]https://www.sciencedirect.com/science/article/pii/S187396521...
If I measure myself on a scale, do you discredit the measurement because the scale produces a value that is some fractional, fractional, fractional percent of the Earth's mass?
[1] https://www.washingtonpost.com/news/energy-environment/wp/20...
It's like looking at someone who's hair is falling out and saying "look, they're barely losing any weight at all!"
GPS is not part of the measurement suite in OP, but better than 1.8 cm vertical resolution from GPS has been possible since the late 1990s, using after the fact orbit determination. It is now routine to get sub-cm vertical accuracy in GPS measurements [1, fig. 3, magenta line]. The horizontal displacements are known to about a mm.
These displacements are known so accurately that you can readily measure groundwater withdrawals, the extra mass related to of high atmospheric pressure over the GPS sensor (daily cadence), and post-glacial rebound (decadal cadence).
Your remarks below about the accuracy of the GRACE gravimetric measurement are also misguided. You can’t just choose a baseline mass and say it’s impossible to measure a change in that mass to some percent accuracy. The GRACE measurement depends on how well you can measure a distance between two spacecraft, and how well you can eliminate systematics from that difference. It does not depend on a relative mass.
Anyway, the paper uses a different technique to get ice fluxes. If you don’t know how they did the measurement, why are you so sure it is wrong?
[1] https://www.unavco.org/data/gps-gnss/derived-products/docs/H...
Not clear how this ad-hominem enhances your argument.
>If you don’t know how they did the measurement, why are you so sure it is wrong?
I'm not "so sure", I'm just skeptical. I find it hard to believe that one can estimate the net ice loss over a huge and dynamic body of ice with an accuracy better than 0.001%, using any type of measurements conceivable.
You are saying that somehow the whole mass of Antarctica is irrelevant as a base mass to calculate accuracy? Why so? The ice naturally compresses and flows and this happens at continental scale. A large iceberg calving on the western coast of the continent may be offset by a relatively modest ice accumulation on the much larger eastern side. However, that accumulation on the eastern side will be conflated by the general compression that will lead to some ice being pushed towards the edges, etc, etc. There's no natural way to take a part of the continent out of the equation and say "well, we are interested only in this subset".
Furthermore, it seems you're still just launching your own arbitrary theories into the discussion and not responding to anything specifically wrong with the article.
Yes, that is correct. The mass of Antartica, whatever that might mean, is not required in itself. Percent error in estimating that is not relevant. Remember that we are interested in mass changes only.
The GRACE measurement is based on explaining the relative position of a pair of orbiting spacecraft by mass changes in 4500 "mascons" -- equal-area spherical caps that blanket the globe (see figs 2 and 3 of [1]). The fundamental observation is the range and relative velocity of the pair of spacecraft, which can be very accurately determined. The mass within each mascon has an analytically-known relationship to these observations (eqs. 8, 9 of [1] - relating mass sigma to acceleration a).
It's a monthly measurement -- you accumulate a month's worth of orbits, and solve a least-squares problem to fit the range-rate data with the masses. This can also be viewed as maximum a posteriori estimation in the conventional way. See equation 13 of [1].
I hope this clarifies why it does not matter what the ice underneath is doing. The ice has mass, and therefore it affects the gravitational potential that the satellites operate in.
You are correct to be surprised. The measurement is surprising, and it has been revolutionary. Hundreds of papers have been published using it, and it has received scrutiny and undergone improvements for more than a decade. Three independent groups (JPL, UTexas, DLR) have worked on the full retrieval over this time. Lots of what we know about groundwater withdrawals, ice sheets, and more recently deep ocean currents, is based on this measurement.
There is an attribution issue at ice/water boundaries (sec. 6 of [1]). It is due to mixed pixels -- a mascon that is part water, part ice. The mass change observed at the mascon should not be spread evenly over the whole pixel. It is split in constrained way based on errors and priors. But as [1] explains, this is not an "order of magnitude" type error, it's just a correction.
[1] https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/201...
But to repeat -- the fundamental result of the OP does not rely on GRACE (nor does it rely on GPS). It relies on ice sheet velocity models.
That being said, the complexity of this model sounds absolutely humongous. Complex models are exciting for mathy guys to work with, but are inevitably subject to more model risk. As coders say, the number of bugs is roughly proportional with the number of lines of code.
You're going to say that this numerical calculation is bug free. A numerical calculation with 4 thousand mass element (mascons) and distance measurements probably in the billions, and lots and lots of approximations and contributions from the sun and the moon and the planets. Of course you are not going to say it is bug free. Should we trust ice loss calculations coming from GRACE? To be honest, after I read your explanation and all other things about GRACE, I'm less skeptical. I don't fully trust it, but I don't fully distrust it either.
But as you said, this is not that relevant, since the actual result published was not based on GRACE, but rather on ice flow models. Now these models I take with quite a large grain of salt. Why? Models produce estimates, and the estimates generally include an uncertainty level. The uncertainty reflects only how the uncertainty in inputs propagates to the uncertainty in the final result, not the model uncertainty per se.
Take a look at the figure 1 in [1]. It shows the precipitation in Antarctica over 2.5 decades based on several models. The average number (labeled "multi-model mean") shows very little variability from year to year, and you could infer that very little uncertainty as well. However, this is just an artifact of taking the average of several not very highly correlated time series. The model uncertainty is better reflected by the wide range of the numbers produced by the different models. If you look at this range, it's about 200mm/y, which is 10 times the net ice loss we are talking about here.
Do you see why I'm a bit skeptical?
But anyway, thanks again for taking your time with me. I learned something cool today.
[1] https://www.sciencedirect.com/science/article/pii/S187396521...
About fig. 1 of [1]: Of the spectrum of important Earth system models, precipitation is the least well-described. For good reasons: highly non-gaussian, very local in nature, highly dependent on nonlinear condensation/temperature properties. Even in the continental US, precipitation is not well-described. Your remark about deciding on a bogus "uncertainty" of the mean precipitation in fig. 1 is absolutely correct.
But it's of no use to be "skeptical about models" based on poor results of one class of models. The whole reason that paper was published is to draw attention to that fact! They literally say this in their conclusion. ("Most CMIP5 and reanalysis models are unable to simulate a consistent spatial and temporal precipitation pattern for the Antarctic.")
I have come to believe that skepticism is somewhat empty as a value. That's why you have experts who know what the state of the art is, and what to be skeptical about.
This is not just "one class of models". This is the gain side of the gain-loss equation needed for the estimation of ice balance in Antarctica. If that side has an uncertainty 10 times higher than the stated result, then how can I not be skeptical?
>That's why you have experts who know what the state of the art is, and what to be skeptical about.
"Trust the experts" lead to the 2008 financial crisis. "Trust the experts" was what string theorists were telling us until Lee Smolin exposed them. "Trust the experts" is ultimately what religion is.
>skepticism is somewhat empty as a value
That's a surprising (for me at least) point of view for a scientist.
I'm quite sure you wanted so say something else. Something like in your experience, those who describe themselves as climate skeptics (I mean in what other context do people declare themselves skeptics? I'm a skeptic about the dark matter as well, by the way) are simply obstructionists, or ignorants, or have a political agenda or something else. Since we are all more efficient forming Bayesian priors to work in life with, your Bayesian prior in this case is that "skepticism is an empty value". And that's quite fair. I happen to agree that a large part of "climate skeptics" are quite toxic. The way the Trump administration treats the EPA, or science in general seems quite deranged. But that doesn't mean you can't be genuinely skeptic. In this case skepticism has nothing to do with values.
I simply approach this debate and I use my own heuristics. I don't know any climate scientist to use heuristics such as human character, you seem you do. For me, I apply other heuristics, based on my day to day work with models. I reached other conclusions so far than those that you reached.
I linked this press release because it has a link to the PNAS article, which is free and describes the findings in greater detail.
1) Nonsense modeling. 20 year predictions from the 1990's turned out to be laughable, now they're predicting 200 years ahead. How about they try 2 year models, maybe they could get those right?
2) Can't see why I should care about someone's beach property in 2428. Even if their models were correct, which ch they aren't.
Most people live paycheck to paycheck and these lunatics want to tax and spend their hard earned money to prevent fantasy scenarios. Truly appalling.