Bishop Hill Glacier loss of plot
Aug 15, 2014 Climate: WG2 This morning's must-read scientific paper comes from Science, where a team from the University of Insbruck led by Ben Marzeion has been looking at glacier recession. The results seem to have pleased the green fraternity, and a glance at the abstract shows why:
The ongoing global glacier retreat is affecting human societies by causing sea-level rise, changing seasonal water availability, and increasing geohazards. Melting glaciers are an icon of anthropogenic climate change. However, glacier response times are typically decades or longer, which implies that the present-day glacier retreat is a mixed response to past and current natural climate variability and current anthropogenic forcing. Here we show that only 25 ± 35% of the global glacier mass loss during the period from 1851 to 2010 is attributable to anthropogenic causes. Nevertheless, the anthropogenic signal is detectable with high confidence in glacier mass balance observations during 1991 to 2010, and the anthropogenic fraction of global glacier mass loss during that period has increased to 69 ± 24%.
An abstract written with headlines in mind, I'm sure you will agree.
The results are currently being trumpeted by, among others, the Carbon Brief:
69±24% of glacier loss since 1991 was caused by humans - Science
Chris Hope:
Man-made warming becomes main cause of glacier retreat, study says
and Bob Ward:
New study concludes about two-thirds of glacier mass loss worldwide in last 20 years is due to human activities
There are many others too.
So. let's take a look at how this startling conclusion was reached.
...we quantify the evidence for a causal link between anthropogenic climate forcing and observed glacier surface MBs [mass balance], not of individual glaciers but of all the world's glaciers outside of Antarctica combined. We then attribute the global glacier retreat since 1851 to natural and anthropogenic causes. We use a model of global glacier evolution that treats the MB of each of the world's glaciers contained in the Randolph Glacier Inventory (RGI)... individually, including a simple parametrization of ice dynamics leading to glacier hypsometry change... Forced by observed climate... the glacier model has been independently validated against both annual surface MB observations (fig. S1) and observed, temporally accumulated volume changes of hundreds of glaciers...
For each of 12 reconstructions of the global climate between 1851 and 2010, produced by general circulation models (GCMs) from the CMIP5 ensemble... we reconstructed the area and volume of each glacier in 1851... From this reconstructed glacier state, we modeled the evolution of each glacier forward in time. This forward model was run twice for each GCM: once subject to all known forcings (i.e., solar variability, volcanic eruptions, land-use change, anthropogenic aerosols, and greenhouse gas emissions; we call these model runs the FULL runs below), and once subject to only natural forcings (i.e., solar variability and volcanic eruptions; we call these model runs the NAT runs below).
So if I understand this correctly, we take the CMIP5 models - the ones that failed to predict the recent hiatus in global surface temperature rise and which, by mainstream climatologists' own admission don't include an important part of the climate system, namely heat transport to the deep oceans. Then, despite the fact these models have no known ability to reproduce regional climate either we use their output to feed glacier simulations, which include their own parametrisations. And from this conclusions are drawn.
It's models upon models upon models. And models, moreover, that are known to be unsuitable.
And from this our green friends expect us to accept their entirely uncaveated statements about glacier retreat.
Amazing.
Reader Comments (70)
'An abstract written with headlines in mind, I'm sure you will agree.'
In other words normal for climate science where 'impact' is far more important than facts .
We will know when the tide has really turned when we no longer see these types of papers has the money and interest has moved on.
Its models all the way down!
25 +- 35%?
This is a quote from an earlier comment by ATTP:
From the same person who refuses to read his host's 'Hockey Stick Illusion' (or any of the other good books by AM) and yet feels free to comment on their content in a totally unqualified way. This was sheer projection.
ATTP hasn't/won't even read 'Harryreadme' (I highly recommend it for its insight into modelling culture - and I read it all); so you can tell a man by the books/articles he does NOT read.
Bish, you are right to kick him/it out.
MikeN: Yes, that was my comment too. At 25 ± 35% human contribution to glacial loss, then there has been not been a statistically significant human component since 1851.
Any links to the full paper?
I am not impressed with the results but would like to see the details
Innsbruck University have a bit of form when it comes to useless studies. A couple of years ago they made a painstaking study of why it floods so much in Tyrolean valleys.
Apparently the soil on the sides of Alpine valleys is quite thin and impacted by the weight of snow in winter. So the rain- and meltwater has nowhere to soak and runs off.
This was on the telly in Austria, it must have been a slow news day. My TV set was a lot slower after I threw the beer bottle at it.
I don't understand the +_ 25%. That seems like an undependable uncertainty. Can someone explain?
A brief web-search may reveal why Greenland and Antarctica are excluded.
One estimate of the glacial ice volume (excluding the Greenland and Antarctic ice sheets) is given as 170,000 (±21,000) cubic kilometres (12% uncertainty). Let’s use the higher limit of 191,000.
One estimate of the ice volume in Antarctica is 26,726,300,861,786 cubic km (using the US billion = 1.0E12).
So the paper and its models are working with 191/26,726,300,861 = 0.000000714652 of the world’s known ice.
Even if we use the UK billion (1.0E9) we still get 191/26,726,300 = 0.000714652 (i.e. 0.071%) of the world’s known ice.
Should we start worrying?
Melting Glacier a new buzz phrase to add to your Climate Change Bingo Card.
So the alarmist have lost the propaganda battle on melting icecaps. Arctic Shrinking getting hotter so they say Antarctic Expanding (trapping Eco Tourists in pack ice etc ) getting colder.
So open a new Climate Change front over Melting Glaciers.
So " Melting Glaciers "a new Climate Porn Starlet is reborn .
Potemkin models all the way down.
Mr. Physics is here a lot because the juice's dried up at his place.
Aug 16, 2014 at 7:43 AM OzWizard
I think you confuse UK and US usage. Still, I assume one must make allowances for your (implied by your monicker) location :-)
.............
Your Grace, were Your Grace to keep Anders banned, I, for one, should be grateful. He adds, IMVHO, little to the conversation.
With studies like this one, it's just a matter of time before schoolchildren will be insulting each other with phrases like "You're as dumb as a climate scientist!"
"This paper is silly/rubbish/wrong because it used models that failed to predict something that they were largely unable to predict."
Yet the paper goes on and makes predictions. So you back predictions made with no data being treated as if they are serious or useful.....for exactly what reason again?
This faux paper cherry picks to avoid dealing with where >90% of the world's ice is, misrepresents what is going on with glaciers (that pesky physical evidence of old forest remains being exposed by receding glaciers).
Next the climate obsessed will be echoing solemn reports about Tibetan glaciers melting by 2035 after all.
What is fun to watch is the recursive behaviors of the cliamte kooks. No matter how many times some alarmist claptrap gets debunked, they are ready to circle back to those claims. All it takes is some time, the stale old ideas getting washed up and given a new coat of cheap paint and the true believers will echo them again and again.
Watching Sky News this morning slow news day nothing about Cliff Richard just ISIS riots in St Louis and Legal Highs Legalisation.And finally some Greenies pointed a time lapse Camera at a North American Glacier.
Obviously after a money shot of some Glacier retreating backwards .
Then interrupted by cutesy fluffy Beaver walking up and licking the lens and knocking the camera over So they decided to put it up on YouTube.How sweat.
So sending researchers and TV camera crews to the poles is very expensive and even in the polar summer with continuous daylight still very cold and very dangerous. With continuous darkness and it's Star Wars Empire Strikes Back planet Hoff cold forget about humans going there in Polar Winter
BBC stock Climate Change footage David Shukman ,Roger Harrabin pieces to camera usual drivling platitudes standing in front of frozen lakes breaking apart and swimming Polar Bears and cliff face ice sheets collapsing into the sea all Climate Change money shots .Unfortunately not evidence of Global Warming but evidence of Seasonal Thawing.Next polar Winter the ice will be back with just no one around to measure it or film it.
Good FOI then ,request the filming dates of all the BBC stock Polar AGW footage.It's all logged by the archivist and even written on the clapper board.All shot in the Polar Summer months.
However less challenging filming Glaciers in Canada Norway and Lord of the Rings location New Zealand much easier putting a time lapse camera on top of a hill pointed at a local glacier preferably in the middle of summer. Usual trick be rather vague about the dates it was filmed over and the frames per hour or per day Just how many years will they leave the camera up there rolling .
Time lapse Photography will show Glacier rate of flow and it's width but it won't show its depth and it's volume.also if a Glacier appears to contract it could be blocked at it source by more freezing ice at higher altitude.So time lapse photography ambiguously proves nothing.
So Glaciers are now the new alarmist propaganda weapon ,so expect Roger Harrabin in nice Timberland Grizzly Adams boots and jacket and a Davy Crockett fake fur hat pointing at a Glacier prattling on.With a Forrest Ranger near by with an M16 just in case he gets eaten by Yogi and Booboo.
And we can all mark Gacial Retreat off our Climate Change Bingo Cards on next edition of Newsnight.
If only that little Beaver had wiped its arse on the time lapse lens camera instead of its nose on Sky News this morning it would have been so much funnier.
I can only conclude that CO2, among its other miraculous properties, can travel back in time since even this article admits glaciers started melting as the Little Ice Age ended.
Sorry for the long post, but I've got a few issues with the methodology in this paper, and I think that is important to point a few sources of uncertainty that have been neglected. The whole paper is an exercise in calibrating a model with too many parameters to the desired output.
(Marzeion et al. 2014) say "The model thus does not attempt to capture the full energy balance at the ice surface, but relies on air temperature as a proxy for the energy available for melt. "
However Huss et al. (2009) has shown that the strong Alpine glacier melt in the 1940s was due to enhanced solar radiation and that this change in solar radiation question the stability of temperature index melt models over decades.
Temperature and precipitation are prescribed by the CRU CL 2.0 dataset plus anomalies introduced by CMIP5 models. However these models have large biases with uneven regional distribution (John & Soden 2007), their local application to the scale of a single glacier is questionable.
(Marzeion et al. 2014) paper cannot account for non temperature-related influences on the mass balance of glaciers. There are many parameters of questionable quality, and therefore the uncertainty of the results is underestimated.
An increase of 20% in precipitation compensates for a 1°C increase in temperature in the European Alps (Braithwaite & Zhang 2000) and precipitation is the variable with the largest error in the data set they use: (New et al. 2002).
Thus it is very difficult to know if the mass balance change is attributable to changes in temperature or errors in the estimation of precipitation. (Marzeion et al. 2014) use a constant lapse rate for precipitation:
" While mean model skill shows a weak decrease with increases of the precipitation lapse rate γprecip (Fig. 5, panel b), the mean rmse tends to decrease with increasing γprecip (panel c). Most importantly, the model has a non-zero mean bias for γprecip >~ 2%/100m and γprecip >~4%/100m. We therefore set γprecip = 3 %/100 m."
But it is well known that precipitation lapse rates are not constant and very difficult to estimate (Thornton et al. 1997; Sevruk & Mieglitz 2002; Lehning et al. 2011).
Likewise for temperature. they estimate lapse rate from the CRU CL 2.0 ("The temperature lapse rate γtemp is estimated at each glacier location by regressing temperature of 3 × 3 CRU CL 2.0 grid points around the location of the glacier onto zCRUclim.").
But is not clear how the estimate the lapse rate over the actual glacier. This is rather complicated as it changes seasonally and depending glacier size and wind conditions (for large glaciers during the ablation season it is a dry adiabatic lapse rate due to katabatic winds when geostrophic winds are low)
The lapse rate, therefore can change from slightly positive to -0.0098°C m^-1 see for example (Greuell et al. 1997; Smeets et al. 1998, 2000; Greuell & Smeets 2001).
Then there is a lot of scope for calibration and uncertainty in their estimation of threshold for melt and threshold for solid precipitation (the first between -2°C and + 1.5C and the second from 0°C to +5°C). They set a fixed value, but the fact is that the actual values can change from day to day and from glacier to glacier. etc. etc.
Braithwaite RJ, Zhang Y (2000) Sensitivity of mass balance of five Swiss glaciers to temperature changes assessed by tuning a degree-day model. Journal of Glaciology, 46.
Greuell W, Knap WH, Smeets PC (1997) Elevational changes in meteorological variables along a midlatitude glacier during summer. Journal of Geophysical Research, 102, 25941–25954.
Greuell W, Smeets P (2001) Variations with elevation in the surface energy balance on the Pasterze, Austria. Journal of Geophysical Research, 106, 31,717–731,727.
Huss M, Funk M, Ohmura A (2009) Strong Alpine glacier melt in the 1940s due to enhanced solar radiation. Geophysical Research Letters, 36, n/a–n/a.
John O, Soden BJ (2007) Temperature and humidity biases in global climate models and their impact on climate feedbacks. Geophysical Research Letters, 34, L18704.
Lehning M, Grünewald T, Schirmer M (2011) Mountain snow distribution governed by an altitudinal gradient and terrain roughness. Geophysical Research Letters, 38.
Marzeion B, Cogley JG, Richter K, Parkes D (2014) Attribution of global glacier mass loss to anthropogenic and natural causes. Science.
New M, Lister D, Hulme M, Ian Makin (2002) A high-resolution data set of surface climate over global land areas. Climate Research, 21, 1–25.
Sevruk B, Mieglitz K (2002) The effect of topography, season and weather situation on daily precipitation gradients in 60 Swiss valleys. Water Science and Technolgy, 45, 41–48.
Smeets CJPP, Duynkerke PG, Vugts HF (1998) Turbulence characteristics of the stable boundary layer over a mid-latitude glacier. Part I: a combination of katabatic and large-scale forcing. Boundary Layer Meteorology, 87, 117–145.
Smeets CJPP, Duynkerke PG, Vugts HF (2000) Turbulence characteristics of the stable boundary layer over a mid-latitude glacier. Part II: Pure katabatic forcing conditions. Boundary Layer Meteorology, 97, 73–107.
Thornton PE, Running SW, White MA (1997) Generating surfaces of daily meteorological variables over large regions of complex terrain. Journal of Hydrology, 190, 214–251.
"Sorry for the long post, but I've got a few issues with the methodology in this paper, and I think that is important to point a few sources of uncertainty that have been neglected. The whole paper is an exercise in calibrating a model with too many parameters to the desired output."
The whole paper is a dishonest exercise in curve-fitting.
The basis of the claim is that:
1. CO2-sensitive models curve-fitted to observed temperatures match observed temperatures with CO2 turned on and fail to match it with it turned off.
2. Temperature-sensitive glacier models curve-fitted to observed temperatures match observed glacier changes when driven by climate model outputs with CO2 turned on and fail to match when driven by outputs with it turned off.
It's a trivial result, and the consequence of their methodology. The method would yield the same conclusion if you used astrological predictions instead. You create a model by regressing world events against the orbits of the planets, setting the constants to get the best fit, and then you observe what the same model outputs if you take one of the planets away. When you find that your model no longer shows the same rise in unemployment when you remove Saturn from the picture, you can calculate exactly how much of that unemployment is being caused by the position of Saturn in the heavens.
It's a ridiculous joke of a paper. And the smoking cherry on top is the fact that they're apparently getting the 'observations' they're using to check the models against reality from the database of climate data that 'Harry' was trying to fix when he was writing the famous 'HARRY_READ_ME.TXT'.
Perhaps they ought to read it.
John Marshall, "How many glaciers were actually visited? There are several thousand."
The best estimate is about 160,000 glaciers of which perhaps just 100 have been studied in any detail and maybe another hundred given some cursory study.