EM. Suggest you check the width of your Irish eskers, compare with width of the Thwaites Glacier water-based areas and fast flowing ice streams, and revise arguments.
I have never disputed that the majority of ice melts when it contacts the ocean. What is important is the rate at which ice is transported to that location. This could well be controlled by basal water lubrication caused by geothermal warming. Your hangup, causing you to waste your time on ridiculous calculations, is to believe that ice loss would be caused by direct melting.
I note that you haven't resolved the problem of hydraulic gradients at the base of the glacier where it overlies a basement that slopes landwards.

"The paper mentions that the base of the ice is warmer, as would be expected". Gosh EM is that warmth due to increases in atmospheric CO2 or, heavens-to-Murgatroyd, could it be geothermal heat flow?

Supertroll

"I note that you haven't resolved the problem of hydraulic gradients at the base of the glacier where it overlies a basement that slopes landwards."

Actually I did, for my own amusement. I didn't think you would be interested.

Atmospheric pressure is 105 kilopascals or 10.5 tonnes/m2 and can raise a water column 10.3metres against a vacuum.The pressure exerted at the thickest part of Thwaites is dues to 2700m of ice thickness. The pressure on the water film at the base is 2700 tonnes/m2 or 2.7 megapascals. That will lift water by 10.3 ×2700 / 10.5 = 2640 metres.

This is enough to lift water from the bottom of the Thwaites basin (-1700m) to 940m above sea level, more than enough to overcome the hydraulic gradient.

EM. Please read what I wrote. The problem is not can you generate a gradient that will expel the subglacial water upslope and to the ocean. It is that if you do this, then seawater cannot flow downhill against that gradient. Thus there is no danger of seawater ingress or a consequent collapse of the ice cap.
Another problem was pointed out by GolfCharlie. If seawater does gain access by flowing beneath the icecap and downslope, what happens when it has melted some ice and cooled. What mechanism will move this water allowing more water access?

EM. BTW. Check out Darcy's Law. It shows that flow is determined by other factors than the hydaulic gradient, especially the cross- sectional area and the length the fluid flow. Because friction will be high in the thin layer of water
between the closely spaced floor and the ice base, even a large hydraulic gradient may fail to drive the fluid. I would be interested if you do a more appropriatecalculation.

Supertroll

Here's a description of my local esker field. Beragh is about seven miles southeast of my home in Omagh.

Note the dimensions of the eskers.

"Esker ridges are up to 1km long, 5-10m high and 10-40m wide and individual ridge fragments are separated by areas (100-500m wide) of non-deposition and/or erosion."

Plenty of room for bulk flow, so in this case the water did not have to squeeze along the base. The smallest of the Beragh eskers could drain the whole Thwaites basal annual melt (0.18 cubic km) at a flow rate of 0.1m/sec.

In my PPL days I flew over the area occasionally. You could see the ghost of a grid pattern from the eskers and moraines.

Incidentally, credit to Cyril Reilly, head of geography at my school. He needed staff to accompany field trips, so I walked over a lot of the local glacial terrain with him and his pupils.

Supertroll

I don't think Darcy's Law is relevant to Thwaites. The law describes water flow through porous rocks or sediments down head of water or gravity gradients under low pressure.

Water movement under Thwaites is through a network of channels and driven under high pressure from the weight of the overlying ice.

There is not much outward flow, 0.18 cubic km of meltwater/year. As I said earlier, this could be accommodated in one small channel. At the grounding line the pressure of the seawater and of the water layer under the ice are the same. Warm CBW can flow under the ice and push back the grounding line. Once the grounding line retreats onto the reverse slope seawater can flow downhill in large quantities under the glacier.

As golf Charlie pointed out, this water will accumulate under the ice. Leaving aside the melting, this will lift the ice off its bed, decrease the friction and increase the flow rate.

The concern is that a combination of melt, increased flow rate and seawater accumulating under the ice will cause the Thwaites basin to empty itself into the sea, with interesting effects on sea level.

Modified Darcy's Law applies for flow through narrow openings like fractures. If I remember correctly you need an asperity term.

As golf Charlie pointed out, this water will accumulate under the ice. Leaving aside the melting, this will lift the ice off its bed, decrease the friction and increase the flow rate.

The concern is that a combination of melt, increased flow rate and seawater accumulating under the ice will cause the Thwaites basin to empty itself into the sea, with interesting effects on sea level.

Oct 10, 2017 at 9:56 PM | Entropic man

It takes "about" the same volume of water to float the same volume of water.

I don't think this "concern" should be taken too seriously.

EM. Darcy's law applies so long as flow is laminar, which I assumed it would be in a narrow melted slot at the base of a glacier. Even at Reynolds Numbers between 1 and 10, another term can be added (Forchheimers) which is able to account for the non-linear behaviour of the pressure difference vs flow rates.
Whatever is the case, your calculation does not account for frictional energy losses which will decrease flow. An extreme case I know, but imagine trying to push syrup through two close-spaced glass sheets. You might be able to do it with small sheets but exceed a certain size and flow would stop. In the case of draining water from beneath a glacier you are moving a thin water film for tens of kilometres. Frictional losses will be high reducing discharge from that you calculated.

https://phys.org/news/2017-06-thwaites-glacier-ice-loss-quickly.html

"The melt rate of West Antarctica's Thwaites Glacier is an important concern, because this glacier alone is currently responsible for about 1 percent of global sea level rise. A new NASA study finds that Thwaites' ice loss will continue, but not quite as rapidly as previous studies have estimated."

The inaccuracy of the previous studies, is blamed on computer models.

The problem with all computer models programmed for the benefit of Climate Scientists, is that they are programmed to believe that natural processes have never caused variations before, and natural variations tend to vary.

Climate Scientists need to reprogramme themselves, if they ever want to produce reliable models.

Oct 11, 2017 at 11:51 AM | Supertroll

The frictional losses you describe may not have been programmed into the models that are being relied on.

Supertroll

What I calculated was the rate at which geological heat flow would melt ice from the base of the glacier, a maximum of 0.18 cubic kilometres/year. How quickly it then flows out to sea is another matter.

From your comments, your own mental model is of uniform water flow through a uniform thin film between the base of the ice and the bedrock.

I think the flow is not uniform, but intermittent. I note the presence of sub-glacial channels and sub-glacial lakes, which empty suddenly. My own view is that water melting from the base of the ice accumulates in the lakes and then empties down channels when each lake reaches a critical volume. This is intermittent flow in large quantities in large tubes, so Darcy's Law does not apply.

Where Darcy's Law applies is when the ice melts and the water has to move through a thin film to the channels or the lakes. This brings up an interesting conundrum.

If the friction limiting water flow is high, you get a thicker water film which better lubricates ice flow. If the friction limiting water flow is low, you get a thinner film, more resistance to ice flow and more frictional heating. Too much Darcy's Law friction and water accumulates in the film where it melted, which makes the film thicker and lowers the friction. Sounds like an equilibrium maintaining a fairly constant film thickness.

One final implication. If the flow out from under the grounding line is intermittent, then it easier for CBW to sleep in.

EM. Those channels of yours must occupy a volume. Imagine the situation before your episodic discharge. How do you create your channels? Do they suddenly form? How does the ice that was there before the channel formed melt? Presumably you now call upon geothermal heat or have you dispensed with that? Pray also explain how global warming affects the base of the glacier more than a kilometre down.
Another problem for you, most of the glacial floor is ice on rock, only channel like features are wet based. How does the whole glacier slide off into the sea?

Oct 11, 2017 at 1:36 PM | Entropic man

But the theories and maths you quote, as used in the computer models have been proved to over exaggerate. If your maths is ok, where are the mistakes?

Oct 11, 2017 at 2:58 PM | Supertroll

The "science" of rheology
https://en.m.wikipedia.org/wiki/Rheology
confirms it is not perfect physics, under laboratory conditions

I expect it would be difficult to model, even if observed data was available from the void that exists.

In Virtual Reality Climate Science, anything COULD be possible, and computer models can be used to make up anything required, that isn't there.

Supertroll
Look at the channels link again. The authors discuss two types of channel.

Concentrated channels are set into the ice, like eskers. They occur where the slopes are steeper, the basal water flux is greater, basal shear stress is greater and closer to the sea.

Distributed channels are set into the sediments below the ice. They occur where the slopes are shallow, when basal water flux is lower, basal shear stress is lower and further inland.

The concentrated channels could be produced by flooding as sub-glacial lakes empty, the intermittent flow I described and linked. Water flowing fast and in large quantities finds a weak point and rapidly erodes a channel through the ice. Such a channel might well be transient. When the flow stops, the pressure of the ice above flattens out the channel until it disappears.

Distributed channels could be produced when the normal basal water movement finds a weak spot in the underlying sediment and washes out a shallow channel along a line of least resistance.

How does heat reach the base of the glacier apart from geothermally? Remember moulins? Ice melts at the glacier surface due to relatively high temperatures, rain or surface darkening. The water flow downwards through cracks and crevasses, carrying heat to the base.

How does the whole glacier slide off into the sea? Given the reverse slope, it probably won't. Some will slide to the sea. Once the surface elevation drops and the grounding line retreats onto the reverse a slope, most of it will probably melt in place.

Oct 11, 2017 at 6:25 PM | Entropic man

You link to this from July 2013:

Evidence for a water system transition beneath Thwaites Glacier, West Antarctica
by Dustin M. Schroeder Stanford University

but ignore this from June 2017:
https://phys.org/news/2017-06-thwaites-glacier-ice-loss-quickly.html

"The melt rate of West Antarctica's Thwaites Glacier is an important concern, because this glacier alone is currently responsible for about 1 percent of global sea level rise. A new NASA study finds that Thwaites' ice loss will continue, but not quite as rapidly as previous studies have estimated."

Golf Charlie

Yes,I saw it. Encouraging if it's correct.

Oct 11, 2017 at 7:36 PM | Entropic man

Perhaps you should not cling to some of the dire warnings produced by those funded to produce scary stories?

https://phys.org/news/2017-06-thwaites-glacier-ice-loss-quickly.html was produced by Helene Seroussi, a scientist at NASA's Jet Propulsion Laboratory in Pasadena.

It is of course possible that NASA Scientists are worried about their employment, and the change in US Political climate is reflected in a change in some of what NASA would like to be concluded from the data that they produce, and has been interpreted by others.

And now, the ability of volcanic activity to cool the planet

https://wattsupwiththat.com/2017/10/11/ancient-pacific-northwest-eruption-blocked-out-sun-cooled-planet/

Interesting debate about topics of which I know nothing.

When we started debating global warming, we launched into the subject without discussing the data. We have touched on the meaning (or not) of average global temperature, but what about the measurements that underpin the global datasets? In fact, all of the Western datasets are based on HCN, itself a massaged and homogenised beast that was created from some long forgotten measurements. Then the Met Office, NASA GISS and NOAA set about creating their own beasts from that.

What do you think of GISS rewriting the database every month? We know what the average temperature for, say, 1925 is today. We can look up what it was last month, but we must wait until next month to find out what it is then. Is this science?

Then, during the pause, we have records being claimed every month based on increases of thousandths of a degree. The original measurements were to the nearest degree Fahrenheit.

The increases were almost certainly the result of the massaging. The different datasets are based on different strategies. If I remember correctly, the Met office leaves grid cells empty if there is no data with which to fill them. I believe that GISS smears temperatures from existing thermometers that could be thousands of miles South. Then we have the different strategies for handing sea, ice and snow cover.

Then, another big can of worms is UHI. Then we have the massive reduction in stations. Today, most stations are at airports, subject to aircraft exhausts, tarmac, concrete and air conditioned buildings.

Then we have the measurement itself. BOM, the Aussie Met office has been caught breaking the rules by using one second measurements. This makes the maximum temperature derived from a series of measurements more prone to electrical or interference spikes which may not even represent the temperature.

Some people claim that the above datasets are not fit for purpose and that the original raw data was discarded. Certainly, normal scientists should be appalled with the practises of climate science. The satellite records do not go back so far, but I would imagine that they are infinitely more reliable. There have been problems of orbit decay, but I believe that these are understood and corrected.

These is an argument that we should only consider the satellite record for the reasons discussed above, but NASA GISS, NOAA and the Met Office would lose much of their evidence for warming.

Oct 17, 2017 at 9:04 PM | Unregistered CommenterSchrodinger's Cat

The datasets aren't fit for purpose as they are themselves a model in which the models. are compared with (I believe in a circular fashion at times).

It would be really easy to test the dataset models though by removing each station in turn and seeing how the dataset model predicts the temperature of the removed station. I reckon the sparsely populated polar regions might get have some dramatic errors in the model. Does anyone know if this has been done at all?

Does anyone know if this has been done at all?

Oct 17, 2017 at 11:35 PM | Rob Burton

Does anyone know if models are subject to any form of Peer Review, or external audit?

What do you think of GISS rewriting the database every month? We know what the average temperature for, say, 1925 is today. We can look up what it was last month, but we must wait until next month to find out what it is then. Is this science?

1. You discover that there is more data available than there was before. You should, therefore, update your estimate. Maybe it's not that you didn't know about this other data, but that it just wasn't possible to use it before, but is now. Maybe it hadn't yet been converted into a suitable format). Maybe it comes from a country that hadn't made it available before.

2. You develop an improved technique for analysing the data. You should, clearly, reanalyse your entire dataset. Scientists regularly develop new ways of analysing datasets. It's perfectly normal to do so.

3. Maybe the data for 1925 hasn't changed, but the data used to define the baseline does. Remember, these are anomalies, so are measured relative to a baseline. If the baseline changes, the anomaly values change. It doesn't mean that the temperature in 1925 changed, but that the temperature against which it is being measured changed.

4. There are other possibilities, none of which suggest that NASA is changing these data values for any reason other than to present the best estimate of the - in this case - global surface temperatures.

The key point is that good science involves updating things when new information is available. Not doing so would be wrong.