A Cool Look at Glaciers

Glaciers are the poster child of those who would have us believe that unstoppable global warming is in progress. We've been told that melting is accelerating just about everywhere, and that hundreds of millions will be affected in the future if glaciers continue to melt and reduce the boost they give to river flow in the dry season in areas like south-east Asia.

Glaciers hit the headlines in a big way in December 2009, when reports of a flawed prediction about Himalayan glaciers began to circulate on the internet. They quoted a statement in the IPCC 2007 AR4 report:

Glaciers in the Himalaya are receding faster than in any other part of the world (see Table 10.9) and, if the present rate continues, the likelihood of them disappearing by the year 2035 and perhaps sooner is very high if the Earth keeps warming at the current rate. Its total area will likely shrink from the present 500,000 to 100,000 km2 by the year 2035 

IPCC AR4 WG2 Ch10, p. 493

The statement was lifted from a 2005 WWF report (pdf) and contains a silly error; the current area was earlier quoted as "three million hectares" which is 300,000 squ.km (about right). This error, the unbelievable 2035 date, the fact that only a small number of glaciers has ever been surveyed, and the fact that a shrinkage from 500,000 to 100,000 km2 can hardly be described as "disappearance" were all reported by reviewers of the draft report, and amendments promised, but nothing was changed before publication.

In fact, Table 10.9 shows nothing of the kind; the average recession of the 9 glaciers listed is of the order of a few tens of metres. Simple arithmetic shows that if the rate continued, even the shortest would take around two hundred years to disappear. There are at least 15,000 Himalayan glaciers, and no one knows if the nine listed are a representative sample or not.

At the time,  Rajendra Pachauri, the chairman of the Intergovernmental Panel on Climate Change (IPCC), described an Indian government report that criticized the claim by the IPCC over the faster than expected melting of Himalayan glaciers, as “voodoo science”. If you'd like to read that report, titled "Himalayan glaciers: a state-of-art review of glacial studies, glacial retreat and climate change" and decide for yourself if it's “voodoo science”, you can find it here (pdf).

I found it an excellent, well-structured and informative read, even if the grammar is a little rusty in places. The above-mentioned chart (10.9) is notable, to my mind, for the age of much of the data; the latest data year ranges from 1957 to 2001, with most in the 80s and 90s. The "voodoo" report has much up-to-date (2009) data, in particular on the Gangotri Glacier, which feeds the Ganges (Ghanga in the report) river.

The linked AR4/WG2 page rightly identifies the importance of the Gangotri Glacier, and says "The 30.2 km long Gangotri glacier has been receding alarmingly in recent years", and quotes an average recession of 23 metres from 1985 to 2001. Note the recession rate and the length of the glacier (which also has several tributaries) - if that rate continued it would take 1300 years to disappear. The Indian report says

Gangotri glacier, it may be noted, had been showing a rather rapid retreat at an average of around 20m per year till up to 2000 AD, which had led to the imaginative prediction of the end of this glacier in [the] next 35 years or so. In actual fact, since 2001AD, rate of the retreat has come down considerably and between September 2007 and June 2009 this glacier is practically at stand still.

and provides tables of yearly recession, a graph, and photographs of the snout (lower end) from 2004 to 2009. It's clear to me Pachauri didn't actually read the report he dubbed "voodoo science".

The report surveys several glaciers in detail, and notes that the largest, the 74 km long Siachen glacier, "the second largest glacier known outside the polar and sub-Polar Regions", has receded very little between 1848 and 2008. The report points out that most Himalayan glaciers are receding, but that

Data that has been generated from the glacier studies, in the Himalayas, over the last 100 years or so, indicates that the glaciers, in the Himalayas, have been, by and large, shrinking and retreating continuously, barring a flip here and there, but the rate of retreat can not be considered as alarming/abnormal, especially in the last decade or so.

Studies of of glaciers in other areas such as the Andes show that some are receding quite rapidly, a few small ones have actually disappeared. However, attempts to show that this is linked to "Global Warming" are unconvincing; the temperature in many regions of the Andes has not increased markedly and in many cases has dropped. Some studies have shown that precipitation is the key; there's less snow to feed the "accumulation zones" high in the mountains. This is true for the much-discussed ice-cap on Kilimanjaro in Kenya. Its reduction in recent decades could not have been due to melting, as has been claimed; the temperature over the ice-sheet never rises above about -2°C. Recent studies show that the ice is sublimating, that is evaporating without melting. Snow precipitation has reduced over decades, and is insufficient to offset the ice loss.

In fact the response of glaciers to local changes in temperature, precipitation, wind speed and direction is complex, and in many cases the opposite of what might be expected. The"voodoo" report discusses these factors in some detail. It's not my only source of course, but it wasn't intended for just scientists, so it's relatively non-technical in language, and easy to read and understand.

Reports or media articles which generalise about glaciers and their fate are not, in my opinion being entirely truthful. To put it plainly, I've read a lot of unsubstantiated rubbish on the subject, and the volume is increasing daily.

The 2007 IPCC report has been described as the "Gold Standard" of climate science, so let's examine some of the the scientific prose in that Himalayan glaciers section (my bold).

Himalayan glaciers cover about three million hectares or 17% of the mountain area as compared to 2.2% in the Swiss Alps. They form the largest body of ice outside the polar caps and are the source of water for the innumerable rivers that flow across the Indo-Gangetic plains.

"The source of water" implies the only source, whereas in fact spring snow melt is the greatest source in the upper reaches of the rivers. Most published data does not, and cannot distinguish between the two. The page doesn't mention snow melt at all, neither does it mention simple rain catchment. It may come as a shock to some that there's still rainfall in high mountain areas even during the dry season on the plains. The internet is awash (no pun intended) with articles which claim that "The Gangotri glacier, which provides 70% of the water of the Ganges..." or similar wording, claiming that upwards of a billion people are almost entirely dependent on glacial runoff in the dry season. These claims are not just far from the truth, but totally untrue. If the glaciers provided that proportion of the flow in the massive rivers, they'd have disappeared millennia ago. The original source for this untruth is now untraceable, but the myth lives on.

It's a simple matter to estimate how much ice the Gangotri glacier loses during the dry season, and it may well contribute 70% to the flow in the upper few tens of km of the river's course, but the proportion falls with distance from the source. A refutation of the early demise of the Gangotri glacier, and an assessment of its  contribution to riverflow is worth a read. Some data on glacial runoff contribution and affected populations can be found in Modeling Glacier Contribution to Streamflow (2010).

What about those "innumerable rivers" - not a particularly scientific expression to my mind, and the Survey of India might be surprised, or even amused, to hear them described as such. There's more hyperbole on that IPCC report page:

The current trends of glacial melts suggest that the Ganga, Indus, Brahmaputra and other rivers that criss-cross the northern Indian plain could likely become seasonal rivers in the near future as a consequence of climate change and could likely affect the economies in the region. 

I've covered the glacier contribution to river flow, but "criss-cross"? I'd like to see two rivers crossing one another. It's good to end on a lighter note.

UPDATE: 27th Oct 2011

From Scientific American

New Research Casts Doubt on Doomsday Water Shortage Predictions

 By measuring the isotopes in river water, scientists have determined that mountain glaciers contribute less than thought to downstream water supplies

A growing number of studies based on satellite data and stream chemistry analyses have found that far less surface water comes from glacier melt than previously assumed. In Peru's Rio Santa, which drains the Cordilleras Blanca mountain range, glacier contribution appears to be between 10 and 20 percent. In the eastern Himalayas, it is less than 5 percent.

I rest my case.