Glacier Retreat and Water Availability

This topic has spilled into Unthreaded.

The one comment that I would be inclined to make on this is that, if people are depending on water from glacier retreat in tropical and temperate settings, then it seems to me that their water supply would be equally diminished by glacier stabilization or advance, not just by glacier disappearance. What proportion of the water delivered to (say) India is from the annual precipitation-melt cycle and what proportion from “mined” meltwater from glacier retreat? Or other places?

During the Laurentide glacier retreat from North America, apparently the Ottawa River was equivalent in size to the Amazon.


  1. bender
    Posted Dec 31, 2007 at 7:27 PM | Permalink

    Dams are no damn good when there’s nothing to trap.

    Payne, J.T., Wood, A.W., Hamlet, A.F., Palmer, R.N. and Lettenmaier, D.P. (2004). “Mitigating the effects of climate change on the water resources of the Columbia River basin.” Climatic Change, 62(1-3), 233-256.

    Christensen, N.S., Wood, A.W., Voisin, N., Lettenmaier, D.P. and Palmer, R.N. (2004). “Effects of Climate Change on the Hydrology and Water Resources of the Colorado River Basin.” Climatic Change, 62(1-3), 337-363.

    VanRheenen, N.T., Palmer, R.N., and Hahn, M.A. (2003). “Evaluating Potential Climate Change Impacts on Water Resources Systems Operations: Case Studies of Portland, Oregon and Central Valley, California.” Water Resources Update, 124, 35-50.

  2. Philip_B
    Posted Dec 31, 2007 at 9:00 PM | Permalink

    Bender, I don’t know what your point is here. Susan said, “glacier retreat is real and serious, having significant implications for water availability in a number of places around the world” (pretty much what the IPCC says) and I asked for one place where this was occuring.

    Your latest link talks about changes in river flows due to changes in precipitation and snow accumulation/melt, then concludes with a vague statement that includes a reference glaciers in no way supported by the study (at least as described in the abstract).

    In all places glaciers are retreating (and precipitation being equal) summer river flows will be higher than were the glaciers not retreating, i.e. water availability increases. I don’t see why this is a difficult concept.

  3. Phil.
    Posted Dec 31, 2007 at 11:08 PM | Permalink

    Re #327

    In all places glaciers are retreating (and precipitation being equal) summer river flows will be higher than were the glaciers not retreating, i.e. water availability increases. I don’t see why this is a difficult concept.

    Because it begs the question of whether the precip is equal, which in many cases it isn’t, which is why the glaciers are retreating, for example Kilimanjaro.

  4. Philip_B
    Posted Jan 1, 2008 at 1:40 AM | Permalink

    Once the glaciers that feed rivers are gone, a significant decrease in water availability will follow, causing water shortages for populations and for crop irrigation. I don’t see why this is a difficult concept.

    This statement is clearly false (and it’s also not what you said originally). Glaciers feed rivers to the extent they retreat (lose ice). A glacier that is retreating makes more water available than a glacier that isn’t retreating. A glacier that has dissapeared entirely makes the same water available as a glacier that is neither advancing or retreating (precipitation being equal and on a annual basis).

    Andrew, I’m not going anywhere with this. I’m just pointing out that this piece of received wisdom is false. I could speculate on why the IPCC said this and why it is so frequently repeated, but that would take me into snippable or Zamboni territory.

  5. Ron Cram
    Posted Jan 1, 2008 at 10:18 AM | Permalink


    It would make sense to me that (all else being equal) as the glaciers melt, you get more water in rivers. That is, until the glaciers are gone – in the years without glaciers you would get less.

    Why don’t we establish that as the null hypothesis and then you show us the numbers and see if it is disproved?

    Or am I expressing an “unhealthy skepticism?” That seems to me to be a very odd term to use when so little data is in front of anyone. Let’s try not to be so judgmental.

    BTW, bender, you never did answer my question on the other thread about why ENSO years, like 1998, are so warm?

    BTW redux, Happy New Year to you. May your football team win!

  6. Michael Jankowski
    Posted Jan 1, 2008 at 11:16 AM | Permalink

    Re#305 Susann,

    Not sure how the WG4 discusses this, but go back to the 2001 IPCC TAR for something interesting:

    “…the timing of the onset of glacier retreat implies that a significant global warming is likely to have started not later than the mid-19th century. This conflicts with the Jones et al. (2001) global land instrumental temperature data (Figure 2.1), and the combined hemispheric and global land and marine data (Figure 2.7), where clear warming is not seen until the beginning of the 20th century. This conclusion also conflicts with some (but not all) of the palaeo-temperature reconstructions in Figure 2.21, Section 2.3 , where clear warming, e.g., in the Mann et al. (1999) Northern Hemisphere series, starts at about the same time as in the Jones et al. (2001) data. These discrepancies are currently unexplained…”

  7. Philip_B
    Posted Jan 1, 2008 at 4:19 PM | Permalink

    Re#363, Peter T, the reason I am interested is that it recently occured to me that Antarctic and Greenland ice accumulation is comparable to Tibetan plateau uplift. – elevation is less but timescale is similar. If the creation of the plateau(s) was the cause of the ice age then ice accumulation would appear to be a feedback intensifying the ice age.

  8. Steve Moore
    Posted Jan 1, 2008 at 5:06 PM | Permalink


    The Columbia River has 14 dams. I don’t see how changes in glacial melt will affect water availability. The only effect is a (possible) small seasonality change in net dam inflows.

    Two things:
    1. Those dams are there for hydropower and irrigation.
    2. There is a mandate to increase flows through the dams for salmon.

    Since #2 carries a lot of weight around here, the net effect will be a significant reduction in the power available for export outside the Pacific Northwest.

  9. bender
    Posted Jan 1, 2008 at 5:43 PM | Permalink

    Phil. is adding something of value. He’s pointing out that I’ve satisfied Philip_B request for just one example where melting glaciers could have an impact on people. I gave him several. Streamflow starts going up as glaciers melt, and then – if warming continues – it will go down. Agreed, we’re not there yet. There’s some melting still to happen. But predicting stream flow is not the rocket science that predicting atmospheric circulation is. The discussion then shifted, not through Philip_B’s contribution, but by others to whether or not the degree of melting has exceeded MWP melting.

    I offered to produce the numbers Steve M asked for – if he was interested in elevating it to the level of a thread. The numbers are there. “Denialist” does not simply refer to someone who doesn’t believe AGW. It refers to someone who will be unconvinced by data, no matter how solid the argument is. My advice is to drop the view that glaciers aren’t melting and that this will not have an impact. There are much better skeptical arguments to be advanced. But hey, if you love being wrong …

  10. bender
    Posted Jan 1, 2008 at 6:38 PM | Permalink

    Philip_B, what did you think of the line in the Schindler paper where he wrote:

    Glacial wasting in the eastern Rocky Mountains, which historically contributed 13–56% of summer flows in the Bow River at Banff (17), has advanced enough that glacial melt is now declining (18).

    17. Hopkinson, C. & Young, G. J. (1998) Hydrol. Process. 12, 1745–1762.

    18. Demuth, M. N., Pietroniro, A., Ouarda, T. & Yetter, (2002) The Impact of Climate Change on the Glaciers of the Canadian Rocky Mountain Eastern Slopes and Implications for Water Resource Adaptation in the Canadian Prairies (Geological Survey of Canada, Ottawa), open file 4322.

    Or did you not see that, as Phil. reasonably hypothesized?

  11. bender
    Posted Jan 1, 2008 at 7:52 PM | Permalink

    I went over the thread to retrace the development of “the” glacier argument, which actually morphed right before the point where I stepped in at #317 to respond to Philip_B’s untenable #315:

    Glacier retreat has no significant implications for water availability

    This was in response to Susann’s #305. Susann was the pivot, asking for data in support of Pat Keating’s #302:

    Oh, come on, Arthur, not that “glaciers are melting” cr*p from you! You are certainly smart enough to know that while some glaciers are melting, others are growing

    But Arthur was trying to make a point, not about impacts, but about the science of “warming in the pipe”, the Earth being out of thermal equilibrium, and how this affects the estiamte of climate sensitivity.

    I think the water availability question has been answered (Pat and Philip_B may want to argue that). But I think that is a side-issue. The real issue is the original one that Arthur raised: how much warming is “in the pipe”, what is it attributed to, and how do you use glacial data to answer the question? I would suggest refocusing on that one, and drop the streamflow question.

  12. Philip_B
    Posted Jan 2, 2008 at 5:18 AM | Permalink

    Re#434 This is the point Andrew made earlier and it’s a valid point. Glacial retreat (ice mass loss) will increase river flows. At some point the glacier reduces or dissapears and the additional flow declines or stops. The risk is that we become dependant on the increased availability of water.

    The statement ‘has advanced enough that glacial melt is now declining’ may be true, i.e. the amount of ice left to melt is smaller and hence less river flow results from melting. It may be that the cause is less recent warming relative to previous periods of warming results in less melting. I have no way of telling.

    I’d add that glaciers don’t store water, at least not in the sense a dam stores water, available for use when required. Ice in a glacier isn’t available for human use. Only when it melts does it become available (to humans).

  13. Philip_B
    Posted Jan 2, 2008 at 5:58 AM | Permalink

    I stepped in at #317 to respond to Philip_B’s untenable #315

    bender, I’m trying hard to keep this civil. I had a serious purpose in asking Susan or anyone to name one place where glacial retreat does have ‘significant negative (a word I originally omitted because I considered it was assumed) implications for water availability’ because it is arguably the strongest claim made concerning the adverse effects of GW. Statements like ‘a billion people are dependant on water from glaciers threatened by global warming’ are bandied around.

    I have yet to see any evidence I find persuasive that water avalability from a river system of any size will be negatively impacted by glacial retreat. In fact, we seem to agree that water availability impacts will be beneficial (i.e. water availability will increase), albeit temporarily (in the order of decades to centuries) in many cases.

  14. JohnB UK
    Posted Jan 2, 2008 at 7:10 AM | Permalink

    Say I have a savings account for ten years which broadly stays the same. I use a little each summer
    for holidays but top it up again in the winter from my regular monthly earnings.

    Then I change jobs and my monthly earnings are lower. As a result my savings account begins to empty.

    I have a problem coming. Which is the cause, my decreased earnings or my diminishing savings account?

    Isn’t this the situation with the glaciers?

    If the precipitation which has been feeding the glacier diminishes, then yes the reduced precipitation is a problem we can discuss, and yes we might need to be concerned. But our problem is the reduced precipitation, not any associated glacier reduction.

    If the precipitation remains the same and the downstream water demand does not increase we don’t have a problem do we?

    If the precipitation remains the same but the demand increases because of the temporary increase in flow due to the ice melt then yes we will have a problem when the glacier is gone, but we will have exactly the same problem if the glacier STOPS melting. Equally if the glacier begins to grow, and more of the precipitation is retained by the glacier, then we potentially will have a supply problem unless precipitation increases.

    If this is correct, then Glacier melt per se is not the problem. The related problems we might face in the future could be:

    Increased water demand fuelled by finite excess supply downstream of a melting glacier.

    Reduced precipitation

    Glacier growth.

    What am I missing?

  15. bender
    Posted Jan 2, 2008 at 7:27 AM | Permalink

    #521 And so you are in denial, because I gave you an example in North America. Flows in the Bow River are already declining and water quality is declining as a result of flow decline. The flow reduction is 13% in heavy precip years, 56% in drought years. The export economy (grain, cattle) and society is already overly dependent on the past increase in flow, a period which is over. You said it yourself: this is a risky situation. The impact is not yet economically significant (although Schindler could probably prove otherwise), but it is statistically significant. I gave you the one example you asked for. It’s a good example. If you need more numbers, they are our there.

  16. welikerocks
    Posted Jan 2, 2008 at 8:57 AM | Permalink

    “Except recently they’re doing it always, not just sometimes.”

    please bender,
    sometimes glaciers covered half of North America 3 miles thick. Farmers can want all they want, we can all want things to “stay the same” too but does that make this a reasonable debate? No. I could say change IS the only constant on Earth and I would be right. This conversation is no better than “thermo”. LOL

  17. Arthur Smith
    Posted Jan 2, 2008 at 9:38 AM | Permalink

    While the glacier discussion is a diversion, I guess I ought to comment in support of “bender” here.

    In the examples “bender” has posted, having a large glacier providing summer melt water helps stabilize river flow. With “equilibrium”, the glacier grows in the winter from snowfall, and then shrinks equally in the summer, effectively transferring that amount of precipitation from one season to the other. It grows during years of heavy precipitation, and shrinks equally during years of drought. Yes of course the total amount of precipitation is the same (though one of the consequences of global warming is changing precipitation patterns, so that’s another issue) but its availability for human use is quite different. When the glaciers disappear, that buffering effect is largely gone. There’s still some seasonal effect from winter snow cover melting in spring of course – until it stops snowing altogether.

    That’s what’s significant here – the loss of glaciers increases the variability of water supply even beyond changing precipitation patterns – that means real economic costs, for those who care about such things.

  18. phil
    Posted Jan 2, 2008 at 12:06 PM | Permalink

    Global vs. Local

    I would submit that one needs to differentiate between global and local effects in regard to glaciers and water availability. The global effect would seem to be fairly clear. When the Earth enters into an ice age, a very large amount of water is “locked up” in permanent ice caps and is no longer available to circulate on a yearly basis. The Earth becomes a very inhospitable, cold and dry place. In a sense, it is “freeze-dried.” When the Earth comes out of an ice-age, as has been happening for some time now, the opposite should be true: a great deal of water is liberated to, once again, circulate on a yearly basis.

    In a parallel sense, a colder world will circulate less moisture, because a colder atmosphere will evaporate less water from the oceans and other bodies of water to be returned as precipitation to feed rivers and streams. A warmer world would have the opposite effect.

    On a global basis, one would expect these two effects to reinforce each other, resulting (when averaged globally) in larger sustainable river flows.

    On a local basis, however, I would submit it is much more difficult to see what the net effects would be (i.e. wetter vs. drier conditions in a given place). The global models at present appear to have technical difficulties with increasing resolution (see various threads here). Consequently, whether the sustainable flow of a particular river system (i.e. Kilimanjaro, the American West), is going to be greater or lesser with global warming I would submit is hard to tell.

  19. TonyN
    Posted Jan 2, 2008 at 12:20 PM | Permalink

    Re: Steve M’s post

    Glacier melt as a contribution to water resources in India. here

    Bob Bradnock is a geographer at the School of African and Oriental Studies, London Uni and an India Specialist. He also gives devastatingly restrained and rational sceptical lectures on climate change.

  20. Francois Ouellette
    Posted Jan 2, 2008 at 1:02 PM | Permalink


    Is there a good reference on global glacier melting or advancing? From what I’ve read (some papers, but also Pielke Sr.’s blog), our knowledge of glaciers is very partial. Many remain unmeasured. Some are retreating, some are advancing, but I don’t know if there is a global estimate of net water budget, and would such a global estimate make any sense? In the end, it seems to me to be a problem of attribution. Are the melting glaciers melting because of GHG’s? And if they started melting before 1900, what do you blame the melting on? And how do we explain the advancing glaciers? If it’s a local or regional effect for the advancing glaciers, then it’s also a regional effect for the retreating ones.

    I agree that retreating or advancing glaciers change the water availability. But if it also occurs naturally, it’s also something we have to cope with. It’s a problem for human beings, not for “the Earth” in general. Our planet has had more than its share of glacial periods, and nature in general has proved very resilient, including Homo sapiens. But changes in water availability have always been a problem for living creatures. In Australia, many of the early settlers arrived at a time when there was more water. They built settlements, villages, that later had to be abandoned, not because of AGW, but only because the climate had gone back to “normal”, ie. very, very dry.

  21. Steve McIntyre
    Posted Jan 2, 2008 at 1:08 PM | Permalink

    #54. The issue in this thread is NOT whether glacier retreat is or isn’t unprecedented – there are other discussions – but the impact of glacier retreat on water supplies.

    As I noted before, if water supply in some areas is from glacier retreat, then, if glaciers stopped retreating, wouldn’t that have as bad an effect on water supply as the glacier disappearing altogether? The issue is then the dependence on areas implicitly using “mined” water from glaciers.

  22. Paddy L
    Posted Jan 2, 2008 at 1:15 PM | Permalink

    I am surprised that no one has cited Ollier, “The Greenland-Antarctica Melting Problem Does Not Exist”. Cliff Ollier is with the School of Earth and geophysical Sciences, University of Western Australia. this commentary upon a working paper by Alan Carlan (NCEE Working Paper #07-07) can be found at:
    where it is reproduced with the author’s permission.

    Ollier makes a powerful case for the proposition that these massive ice sheets cannot melt in a short period nor yield the volume of water that Hansen predicts.

    I am interested in reading comments from those who have read Ollier’s paper.

  23. Sean Egan
    Posted Jan 2, 2008 at 1:20 PM | Permalink

    Bender #37
    about you Bow river example

    ” In the lower stretches of the Bow, the situation is more dire: In average years we extract 68 percent of the average annual flows, and in low flow years, the percentage of flows diverted reach 80 to 90 percent.


    … Contrary to popular myth, the majority of the Bow’s waters does NOT come from glacial melt. In fact, less than one percent of annual flows can be attributed to glaciers. But this does not mean that glacial contribution to the river’s flows are insignificant. For thousands of years, glaciers have been an important source of water in the Bow River. This is a particularly true during the summer and early fall months when glacial melt makes up a greater portion of the flows of the river.”


    Looks like the Bow river folk see it as a water management problem including how the dams flows are regulated rather than a supply problem.

  24. Steve McIntyre
    Posted Jan 2, 2008 at 1:36 PM | Permalink

    I think that another distinction needs to be made with glaciers. Some glaciers appear to move quite rapidly – I’ve noticed this with some Chile glaciers where ice cores don’t recover much history because they move fast. If there were no glacier, it would still snow and there would still be annual runoff as in other areas.

  25. Phil.
    Posted Jan 2, 2008 at 1:47 PM | Permalink

    Re #60

    The point is the advance/retreat of glaciers depends on the mass balance, basically precip-sublimation-melt, there’s more than one way for it to happen. As I’ve pointed out above Kilimanjaro seems to be retreating because of a precip deficit that’s true of others too, you can’t just assume as someone did way back that there’s no problem because the precip will stay the same.

  26. Lance
    Posted Jan 2, 2008 at 1:55 PM | Permalink


    While I think your analysis of the glacier argument is dead on I’d be careful indulging in the use of the word “denial”. If you think someone is ignoring evidence then I suggest you point them to it. If they do not give it the weight you think it deserves then ask them to give reasons for their opinion. If you feel they have ulterior motives then politely ask if that is the case.

    I see no useful purpose in using this tainted word. Denial, when used in this context, is just a weasel word for being a liar or delusional.


    Thanks for giving props to the Wolverines. Exciting game to watch although if Michigan doesn’t provide four turnovers (two inside the five) the Wolves blow the Gators out.

    Nice to send Lloyd out with a win.
    Of course you can’t spell Lloyd with out two l’s.

  27. Mike B
    Posted Jan 2, 2008 at 2:16 PM | Permalink


    Thanks for providing the facts on the Bow River.

  28. bender
    Posted Jan 2, 2008 at 2:25 PM | Permalink

    The one post that Steve M forgot to carry over from unthreaded was the most important one – the one that contains the numbers on flow rates in the Bow River attributable to glacial melt:

    Sean Egan, I’m under no illusions about the degree of melt or impact on water flow. My involvement in this thread is merely to reply to Philip_B’s request for proof of just one case where glacial melt was affecting flow in a negative way. Don’t argue with me. Argue with the numbers and the people who obtained them.

    Lance, Phil already pointed out in two instances where evidence that was provided was not being consulted. I’m not going to repeat what he’s said already.

    As I’ve said, water availability is one issue. Water quality is another. Larry asks to be splained what’s the diff between rainwater quality vs glacial and snowpack meltwater. Good question. Somewhat OT. Except that quality goes down when quantity goes down. I’m not glacier guy. I’m merely responding to someone’s question for info.

    rocks, you’re apology is welcome. But you came this close to being badly flamed for your stunt on unthreaded.

  29. Francois Ouellette
    Posted Jan 2, 2008 at 2:35 PM | Permalink

    I didi a Google Scholar search but couldn’t find much on global glaciers. There’s a good review by Roger Barry, “The status of research on glaciers and global glacier recession: a review”, Progress in Physical Geography, Vol. 30, No. 3, 285-306 (2006) but I don’t have access to that journal, only the abstract:

    Mountain glaciers are key indicators of climate change, although the climatic variables involved differ regionally and temporally. Nevertheless, there has been substantial glacier retreat since the Little Ice Age and this has accelerated over the last two to three decades. Documenting these changes is hampered by the paucity of observational data. This review outlines the measurements that are available, new techniques that incorporate remotely sensed data, and major findings around the world. The focus is on changes in glacier area, rather than estimates of mass balance and volume changes that address the role of glacier melt in global sea-level rise. The glacier observations needed for global climate monitoring are also outlined.

    There’s a UN report “Global outlook for snow and ice”, but it’s rather alarmist, and lacking detailed data. Many papers on regional glaciers.

    For those interested in the sociology of glaciers, I’ve also found this interesting piece, “the history of ice: how GLACIERS BECAME AN ENDANGERED SPECIES”, by Mark Carey. Abstract:

    In recent decades, glaciers have become both a key icon for global warming and a type of endangered species. But to understand why glaciers are so inexorably tied to global warming and why people lament the loss of ice, it is necessary to look beyond climate science and glacier melting—to turn additionally to culture, history, and power relations. Probing historical views of glaciers demonstrates that the recent emergence of an “endangered glacier” narrative stemmed from various glacier perspectives dating to the eighteenth and nineteenth centuries: glaciers as menace, scientific laboratories, sublime scenery, recreation sites, places to explore and conquer, and symbols of wilderness. By encompassing so many diverse meanings, glacier and global warming discourse can thus offer a platform to implement historical ideologies about nature, science, imperialism, race, recreation, wilderness, and global power dynamics.

  30. bender
    Posted Jan 2, 2008 at 2:39 PM | Permalink

    #67 Fair enough. But he referred to “popular myth” of Bow River flow dominated by glacial melt. Whereas I specifically linked previously to a data-rich paper that precisely quantified the flow, with the express purpose of making any such hand-waving unnecessary. In drought years in late summer with no glaciers and no snowpack, the Bow River is not going to flow clean, cold, fast, voluminous. It will be less than half. Possibly less still. I’ve seen the drought effect quantified (~50%) and the late summer effect quantified (~50%), but not both simultaneously. Flow will be less than half (maybe a quarter?). To most ranchers dependent on this water, I think they would say this impact is not negligible, thus satisfying Philip_B’s thirst for an impact example.

  31. Wansbeck
    Posted Jan 2, 2008 at 2:41 PM | Permalink

    A glacier can never be more than a reservoir, albeit a very big one.
    Maybe Libya has the answer. If a river dries up build a new one.

  32. Phil.
    Posted Jan 2, 2008 at 2:41 PM | Permalink

    Re #69

    Try this:

  33. SteveSadlov
    Posted Jan 2, 2008 at 2:45 PM | Permalink

    Take the Ganges or the Mekong. What proportion of each of these river’s flows is from glacial outflow and what proportion is from general run off? I would imagine that cloud forest climate zones would contribute the lion’s share. Case closed. Its Yunnan, not Tibet, that fills the Mekong.

  34. bender
    Posted Jan 2, 2008 at 2:46 PM | Permalink

    #71 Yes, they are reservoirs. That’s why their wastage will have an impact, contrary to what several commenters have either said or hinted at. Yes, new reservoirs can be built. At some cost.

  35. bender
    Posted Jan 2, 2008 at 2:48 PM | Permalink

    #73 Good question. That’s asking for three examples, now. And If you get a third, will you demand a fourth? What is the stopping criterion?

  36. Francois Ouellette
    Posted Jan 2, 2008 at 2:56 PM | Permalink

    I’ve found this paper too: “Reconstructing the glacier contribution to sea-level rise back to 1850“, available on-line. It shows that glacier melting has been more or less constant since 1850, and concludes that there was more contribution to sea level rise from glaciers between 1850-1900 than from 1900-2000.

    More specific to this discussion is this Nature paper.

  37. Mike B
    Posted Jan 2, 2008 at 2:58 PM | Permalink


    #67 Fair enough. But he referred to “popular myth” of Bow River flow dominated by glacial melt. Whereas I specifically linked previously to a data-rich paper that precisely quantified the flow, with the express purpose of making any such hand-waving unnecessary. In drought years in late summer with no glaciers and no snowpack, the Bow River is not going to flow clean, cold, fast, voluminous. It will be less than half. Possibly less still. I’ve seen the drought effect quantified (~50%) and the late summer effect quantified (~50%), but not both simultaneously. Flow will be less than half (maybe a quarter?). To most ranchers dependent on this water, I think they would say this impact is not negligible, thus satisfying Philip_B’s thirst for an impact example.

    Sean quoted directly from the link he provided, namely:

    Contrary to popular myth, the majority of the Bow’s waters does NOT come from glacial melt. In fact, less than one percent of annual flows can be attributed to glaciers.

    Here is the link again:

    Maybe they’re wrong.

  38. bender
    Posted Jan 2, 2008 at 3:01 PM | Permalink

    #78 This is true. The majority of Bow flow does NOT come from glacial melt. It averages around 13%. My point was that during droughts and late summer – when you need it most – it is much higher, greater than 50%.

  39. Steve Moore
    Posted Jan 2, 2008 at 3:03 PM | Permalink

    Intesting paper on the Columbia River.
    Appears to be a worthwhile use of tree ring proxies.

    December 2004

    ABSTRACT:A network of 32 drought sensitive tree-ring chronologies
    is used to reconstruct mean water year flow on the Columbia
    River at The Dalles, Oregon, since 1750. The reconstruction
    explains 30 percent of the variability in mean water year (October
    to September) flow, with a large portion of unexplained variance
    caused by underestimates of the most severe low flow events.
    Residual statistics from the tree-ring reconstruction, as well as an
    identically specified instrumental reconstruction, exhibit positive
    trends over time. This finding suggests that the relationship
    between drought and streamflow has changed over time, supporting
    results from hydrologic models, which suggest that changes in
    land cover over the 20th Century have had measurable impacts on
    runoff production. Low pass filtering the flow record suggests that
    persistent low flows during the 1840s were probably the most
    severe of the past 250 years, but that flows during the 1930s were
    nearly as extreme. The period from 1950 to 1987 is anomalous in
    the context of this record for having no notable multiyear drought
    events. A comparison of the flow reconstruction to paleorecords of
    the Pacific Decadal Oscillation (PDO) and El Nifio/Southern Oscillation
    (ENSO) support a strong 20th Century link between large
    scale circulation and streamflow, but suggests that this link is very
    weak prior to 1900.

    Click to access pnw_2004_gedalof001.pdf

  40. bender
    Posted Jan 2, 2008 at 3:42 PM | Permalink

    Roger Dueck, Philip_B thanks you for this. Can you go through some of you links and post some data? For example on the status of the Columbia glacier, your url states:

    The glacier is not surveyed regularly, so it is not known if it is advancing (Ommanney, 2002).

  41. bender
    Posted Jan 2, 2008 at 3:50 PM | Permalink

    Drom Roger Dueck’s quoted url (which is from McMaster, not McGill):

    During the 20th century, the overall mass-balance trend for Athabasca has been strongly negative (Ommanney, 2002).

    I will see if I can find some more recent data.

  42. bender
    Posted Jan 2, 2008 at 3:55 PM | Permalink

    Peyto glacier. Recent work has shown that the net winter balance has been lower than average since 1976, and that this shift is commensurate with shifts in synoptic-scale winter climatology for the region.

  43. bender
    Posted Jan 2, 2008 at 4:06 PM | Permalink

    #87 I’m not jumping to any alarmist conclusions. I am citing other people’s work in response to someone’s request for evidence of an impact of glacier melt. I would be happy to read whatever information you’ve got available.

  44. WaterResEng
    Posted Jan 2, 2008 at 4:08 PM | Permalink

    Am I the only one that thinks this discussion is a tad silly?

    These cities, whose drinking water supplies are supposedly threatened by the melting of glaciers, only have a water supply BECAUSE the upstream glacier is melting!!!

  45. bender
    Posted Jan 2, 2008 at 4:13 PM | Permalink

    #40 is merely a repeat of #30. I agree it’s logical, and even said as much in unthreaded. When glaciers are growing, you are banking meltwater, even though flows are lower. When they are wasting, you are withdrawing from the bank. When the bank is gone, no more withdrawals. But why would you care about my opinion if you are that ready to dismiss it?

  46. bender
    Posted Jan 2, 2008 at 4:23 PM | Permalink

    #87 Read your link. It does not refute the data that has been presented on the Canadian alpine situation.

  47. Jack Linard
    Posted Jan 2, 2008 at 4:34 PM | Permalink

    As a long-time lurker and having 0ver 40 yrars experience in the development of hydroelectric projects in every continent exceot Antarctica, I feel compelled to comment on the inanities of this thread.

    When the subject turns to water, we in the hydro business have a saying: “I’m not a hydrologist, but …..”.

    Most of the comments on this thread should be prefaced with this.

  48. Nathan Schmidt
    Posted Jan 2, 2008 at 4:37 PM | Permalink

    #66 – Schindler’s paper needs a closer look.

    At first glance, I see Figure 2 showing changes in mean summer flows calculated for rivers that have had major dams constructed during the period of record (Peace River – Bennett Dam in 1967, Oldman River – Oldman River Dam in early 1990’s, South Saskatchewan River – Gardiner Dam/Lake Diefenbaker in 1967). These effects show up clearly in the summer hydrographs (snowmelt floods generally dominate on these rivers and this water is now retained for release later in the year). Schindler only notes dams as one factor, but they are the dominant factor on these rivers.

    The temperature analysis in Table 1 needs a closer look also. They claim negligible UHI effects but we all know the connection between “rural” airports and the temperature record. I suspect we’re dealing with spliced temperature records here (my scan of the Environment Canada database doesn’t show the long periods of record single stations as presented in Table 1; many moves are the norm). It’s interesting that they list no significant changes in temperature at Calgary, Banff and Fort Chip for the period 1970-2003.

    Then there’s the whole GCM thing – he’s using 5 to 8 degree celcius increases in temperature as a basis for calculations?

    That’s just my five-minute scan, but my advice is to treat that paper with a healthy degree of skepticism.

  49. Posted Jan 2, 2008 at 4:38 PM | Permalink

    “I’m not a hydrologist, but

    So, I haven’t commented. But Jack, could you suggest what does happen if glaciers melt? Or reform? (I have absolutely no idea. Either way, my drinking water comes from Lake Michigan which I suspect will not dry up before I die.)

  50. bender
    Posted Jan 2, 2008 at 4:56 PM | Permalink

    Why cancel? Steve M wanted some numbers. Let’s see some numbers.

  51. SteveSadlov
    Posted Jan 2, 2008 at 4:58 PM | Permalink

    Bender, you need to look at the complete basin level. So one bank account gets fully withdrawn, what about the other accounts and cash flows, so to speak. In those tropical basins where glacial run off is in some of the tributaries, wouldn’t you think that other run off related sources may actually increase, if the warmist visions of tropical increases in mid trop temperature and humidity came true?

  52. Nathan Schmidt
    Posted Jan 2, 2008 at 5:22 PM | Permalink

    #101 – Re: Athabasca River flows.

    It might help to have a look at the raw data. You can download the flow records from the Environment Canada web site at

    It’s unfortunate that flow records for the Athabasca below McMurray (Station 07DA001) only go back to 1957. Take a look at the data for the station upstream near Jasper (07AA002) and at the Town of Athabasca (07BE001), which include annual data from 1914 to 1930, and you’ll see little, if any, trend in mean annual flow. The same goes for mean summer flows (the ones Schindler uses) – have a look at the record at 07BE001, which includes complete data for May-August for 86 years (with a short gap in the middle). Little or no trend. It looks pretty convincing to me… I’ve got these graphed but I’m a novice at uploading such things here.

    Regional conditions in the 40’s (when monitoring data are scarce) were pretty dry, and in the early 70’s were very wet (large snowpack). The turn of the century was again on the dry side. We need to be cautious when dealing with periods of record which – at several decades – may appear to be long, but actually include wetter and/or drier than average conditions. It’s a luxury in Alberta to have records of 80+ years, but there are a few that exist. Schindler’s selection of a station which has no data for the first half of the 20th century is perhaps “convenient”.

  53. Jeremy Ayrton
    Posted Jan 2, 2008 at 5:27 PM | Permalink

    I think this thread has missed Steve M’s original point, which, if I may re-state in simple terms is – Glacier retreat is calimed to bad. It reduces the supply of water. However, counteracting this by stopping the retreat of glaciers must also be bad, because this must also reduce the water supply.

    This reasoning could be applied at whatever time in our history you care to choose. In other words, whatever change occurs, at whatever time, it was always a bad thing

    Gosh, how the heck have we managed to survive so long?

    We’re doomed.

  54. bender
    Posted Jan 2, 2008 at 5:28 PM | Permalink

    #113 Thank you for a data-driven argument. That is what Steve M wants. Hand-waving can not be audited.

  55. RobT
    Posted Jan 2, 2008 at 5:33 PM | Permalink

    As a long-time lurker, first-time poster, I dive in at this point because this topic does relate to my field of training (water resources & hydrology), and I happen to be a Canadian who longs to one day wet my fly line in the Bow.

    Far from being “data-rich” as bender suggests, I consider the Schindler and Donahue paper is rather data impoverished. The only data cited in relation to the Bow River is a 13-56% contribution from glacial wastage and a net decline of glacial area from 23 km2 to 16.88 km2 from 1951 – 1993. The 13-56% contribution to flow from glacial wastage is derived from a paper by Hopkinson & Young, Hydrological Processes, vol. 12, Issue 10. I am only able to retrieve the abstract, for which I quote below:

    The mean net glacier volume loss estimate of 934×106 m3 was divided into annual proportions of glacier wastage and storage using a local mass balance record collected at Peyto Glacier in the Mistaya Valley, contiguous to the Bow Basin. Unfortunately, the record began in 1966 and a hind-cast to 1952 (hydrological year) was necessary. Banff maximum summer temperature and Lake Louise snow course data were used as surrogates for summer and winter glacier mass balance, respectively. Monthly wastage proportions were estimated for 1967-1974 by using modelled values of glacial melt as a template. Glacier wastage inputs to and storage held back from the Bow River hydrograph at Banff were compared with known basin yields to assess the hydrological effects of glacier volume change. For 1952-1993, the average annual wastage/basin yield ratio was found to be around 1·8%. For the extremely low flow year of 1970 this ratio increased to 13%. The proportion of flow derived from glacier wastage in August of this year was estimated to be around 56%. Although the results tend to confirm the regulatory effect of glaciers on stream flow, it was found that in some years of low flow this situation has been aggravated by water being held in glacial storage.

    The abstract cites that that average flow contribution to the Bow from glacial wastage is actually only 1.8%, though it appears that there have been some exceptional years: 1970 – a ‘low flow year’; and presumably August of 1997 (“this year” likely refers to the year prior to publishing). The figure of 1.8% is remarkably close to the number that I estimated after reading Schindler (based upon glacial shrinkage of 23 km2 to 16.88 km2) and before reading the Hopkinson abstract.

    So, it appears that even the Schindler paper is not immune from alarmism presented in the absence of complete data. Isn’t the climate audit about carefully scrutinizing numbers like this?

  56. bender
    Posted Jan 2, 2008 at 5:47 PM | Permalink

    #119 What exactly does he have right? He’s got a few things wrong, which I noted.
    He’s also made the same error as Gore:

    The Columbia glacier is one of a number exiting the Columbia icefield and is actually advancing (source: Canadian Glacier Inventory Project, McGill University,

    Precisely which glaciers in the Columbia Icefield are gaining mass? And are you suggesting that advance of the Columbia glacier (in Alaska, not Alberta) necessarily implies mass gain?

    And Jack Linard wants to know: which one of you is the hydrologist? The glaciologist?

  57. Jeremy Ayrton
    Posted Jan 2, 2008 at 5:48 PM | Permalink

    Bender, very briefly OT – (in answer to your #118. We haven’t survived for so long by adaption alone. I suggest we also learned how to anticipate. This is perhaps what our Warming friends are advocating now.

    Although, I have to say that we tend to anticipate one problem only to get bitten on the bum by another (unforseen) one…

  58. RobT
    Posted Jan 2, 2008 at 5:51 PM | Permalink

    Just to clarify my previous post, 13-56% does not represent the actual potential decline in river flow should the glaciers suddenly disappear. This is the proportion of flow derived from glacial wastage during low flow years, that is, during years in which precipitation derived-flow is low. During these periods, the flow from glacial wastage obviously occupies a greater proportion of the total flow. The actual volume of glacial flow would likely vary little from year to year.

    If we take the average flow contribution of 1.8%, this is statistically insignificant and within the range of error that water resource planners will apply in their methods. If this is lost, then the impacts on water availability in the Bow catchment in most years will be negligible, though it may exacerbate problems during periods of drought.

    To bring this topic back on thread, the Schindler paper cannot therefore be used to argue that the decline of glacial meltwater will have a persistent adverse impact downstream.

  59. Posted Jan 2, 2008 at 6:26 PM | Permalink

    Ok, so very few points have been made here other than the fact that melting glaciers provide water, retreating glaciers store ice but reduce water availability, and that the trend in melt/advance oscillation is glacier-specific. I assume that we are only concerned with mountain glaciers in this thread as the arctic and antarctic ice changes are not nearly as relevant (if at all).

    Here are the salient issues:
    -If one were to create a weighted average of mountain glacier ice deltas, would one see a trend towards glacial melting across the globe?

    -Is it possible to explain why certain glaciers have trended towards melting and others have not? For example, if we had precipitation and surface temperature information near or at each mountain glacier, would this be sufficient to explain each glacial trend?

    -Is it true, as bender seems to be asserting in #92, that once a glacier is gone that it cannot come back?:
    “When glaciers are growing, you are banking meltwater, even though flows are lower. When they are wasting, you are withdrawing from the bank. When the bank is gone, no more withdrawals”

    To address the last bullet: I understand that this is true in areas of low precipitation, and is certainly a concern for people relying on a finite declining resource of water to support them. These people (say, the population of La Paz as cited above) are pretty much screwed, as there is no(t enough) precipitation to replenish the glacier regardless of temperatures, and if temperatures were to cool their city would be waterless just the same as if the glacier had expired from warm temperatures. So, while this is a major concern in general, the end result is not largely GW-related, if at all.

    As for areas that receive precipitation, the same point applies. If the net glacial mass is decreasing over time, the local populace is living on borrowed water that they will have to come up with in some other way. Eventually, the “bank account” will be empty, but you’ll still have a salary. Bender’s point that “this is a problem” is, again, true, but it absolutely has to do with our willingness to use less water if a glacier cools and to create steady supplies of water from precipication should the glacier disappear.

    Glacial melt clearly reduces the variance in annual water availability as compared to precipitation sources alone, which is really the crux of the issue here. Areas that rely on a dependable source of water from the glaciers as drought insurance (not sure if any actually exist?) or to supplement their total water requirements (as in low-precip areas) will have some level of problems when/if glaciers disappear.

  60. bender
    Posted Jan 2, 2008 at 7:00 PM | Permalink

    Now which [paraphrased] argument is most difficult to refute:

    The audit:

    “I understand your concern about water; water management is a universal problem. I took a close look at one of the more alarmist papers, Schindler & Donohue (2007), and I find their numbers to be exaggerated, first because of the GCM scenarios they use, and second, because of their cherry-picked stream flow statistics. Although I agree with the argument that there will be a statistically significant impact of hypothetical climate warming on glaciers, snowpacks, and streamflow, the likely impact is small enough that the risk could be mitigated against using moderately inexpensive technologies such as …”

    Or the hand-waving POV:

    “Don’t tell me glaciers are melting. Some are and some aren’t.”

    “Name me one place where melting glanciers could have an impact.”

  61. Philip_B
    Posted Jan 2, 2008 at 10:13 PM | Permalink

    From my link in #118 which seems to answer the question quite well.

    The first results show a similar pattern (in Central Asia) as observed in the Alps in
    previous studies: under a present-day glacier extent the water yield rises drastically
    during the melting period, after a reduction of the glacierized area by 50 % the flood
    peaks are reduced to presently observed values, and after a complete disappearance of
    the glaciers the calculated runoff indicates a strong reduction of runoff during summer
    months, which would have a significant effect on the availability of water resources in
    these dry regions of Asia.

    Although, as I remarked above much of Central Asia has a summer rainfall peak and the ‘strong reduction’ in runoff would be against a background of generally high summer river flows.

    Steve: Any idea of the percentages involved?

  62. Doug M
    Posted Jan 2, 2008 at 10:20 PM | Permalink

    Susann #18. A fine example of cherry picking and failure to consider all factors you have given us. Th Canadian data from 1918 to 2006 are available at The average water level for 2006 was 176.02 meters above their reference point. The water levels were lower than this some years in the 20’s and 30’s. In ’73 the water level was an unusually high at over 177 meters. 177 meters was also exceeded in ’85 & ’86.
    The US Coast Guard had a different set of figures on line this summer. I could not locate them today. They showed a significant drop in water levels occurring in the sixties, with only the mid eighties exceeding the normal level of 581 feet above sea level. The sixties was when the St. Clair river was dredged to a depth of 26 ft. to allow deeper draft ocean going ships to navigate the Great Lakes. As part of the project weirs were supposed to be installed to reduce the increased outflow from Lake Huron. The weirs were never put in place. Since that time the governments of US and Canada have consistantly denied that the dredging could have any effect on the water level. But why were the weirs included in the design?

  63. Peter D. Tillman
    Posted Jan 2, 2008 at 10:21 PM | Permalink

    Re: changing post numbers

    Just a reminder to posters, don’t just quote the post # in your reply — it’s likely to change. Best to head your post with the OP’s name, topic, maybe a brief quote of what you’re replying to. See

    Otherwise, it’s an unholy mess to try to follow — like this thread!

    Try to help other readers, please.

    What a pity that blog software writers ignore 30 years of Usenet experience in dealing with these issues…

    Thanx in advance, Pete Tillman

  64. Steve McIntyre
    Posted Jan 2, 2008 at 11:31 PM | Permalink

    I’ve removed a lot of posts that seemed OT to me. Apologies if anyone feels abused.

    The general point seems simple unto being trivial. Wastage increases runoff.

    But what proportion of runoff in river basins of interest is contributed by “mining” of meltwater from receding glaciers relative to annual precipitation? 1% 10% 50% ??? Let’s try to collate some facts before speculating.

    As to people providing references: it would be far more helpful if you cited figures from the refs, rather than saying the refs look interesting.

  65. deadwood
    Posted Jan 3, 2008 at 12:24 AM | Permalink

    Areas with glaciers generally have sufficient precipitation to sustain human needs so long as some alternate form of storage is used, assuming the glaciers melt away. This requires only planning and funding for either surface or aquifer storage projects.

    India’s and Pakistan’s economies seem to be doing pretty good now. If there is a real threat of losing the storage from the Himalayan glaciers, I would expect their government to do something about it locally.

    America built most of its its surface storage during the great depression when it was practically broke, so I don’t buy the concept that India or Pakistan can’t afford to do it now.

    I can’t think of any other major population as being dependent on glacier melt other than India or Pakistan. Anyone else know of any?

  66. Posted Jan 3, 2008 at 1:55 AM | Permalink

    To bring this topic back on thread, the Schindler paper cannot therefore be used to argue that the decline of glacial meltwater will have a persistent adverse impact downstream.

    how do you come to this wild conclusion?
    a 50% loss of water in drough years is not an argument, supporting “a persistent adverse impact downstream”???

    in a raising temperature scenario, as we are experiencing it at the moment?

    The one comment that I would be inclined to make on this is that, if people are depending on water from glacier retreat in tropical and temperate settings, then it seems to me that their water supply would be equally diminished by glacier stabilization or advance, not just by glacier disappearance.

    this is obviously false. glacier stabilization and advance could both be accompanied by the same or even more melting water.
    it will only reduce melt water, if you assume a constant level of snowfall.

    it looks like for us, the glaciers provide a helpful feedback during times of temperature increase: they provide the steady water necessary in dry years and with warming temperature tend to provide MORE water.
    right up to the moment, when they are (sort of) gone…

    it is a very good example of a climate change problem. it illustrates how nature compensates and delays the change. to judge the effect, we need to factor an increased vulnerability. people depend on the water for irrigation and electricity, which will lead to conflicts. people adjust to a slow change and will have difficulties to handle a fast one. (multi year drought, with no glacier counterbalance)

  67. Philip_B
    Posted Jan 3, 2008 at 6:05 AM | Permalink

    Re: the question on status of the Canadian Columbia Glacier.

    The (Columbia) glacier retreated 394m between 1724 and 1924 and advanced 1km between 1966-1977. After 1975, the glacier advanced further to completely fill the large proglacial lake, a distance of approximately 800m (Ommanney, 2002).

    The size of the advance since the 60s is surprisingly large, especially relative to the size of the retreat from the LIA. Some glaciers are known to periodically surge. Maybe this is one of them. As noted earlier there is no data on ice mass for this glacier.


  68. bender
    Posted Jan 3, 2008 at 8:54 AM | Permalink

    #69 Numbers, please. 56%+ reduction in streamflow during the hottest, driest, riskiest months of crop growth is a problem. Especially if it coincides with increased evapotranspiration in an area such as the great plains which is already drought-prone. If it happens, it will be a big (regional-scale) economic problem. The risk bulit into this situation is a no-brainer. And that is from a non-alarmist. Think about that.

  69. bender
    Posted Jan 3, 2008 at 9:00 AM | Permalink

    #71 Philip_B
    Advance is not gain in mass. Advance means increased flow downslope, which, as you well, know is expected to happen as the ice underneath liquefies. Are you intentionally torquing the interpretation, or do you really not understand the language and mechanics of glacier flow?

  70. Steve McIntyre
    Posted Jan 3, 2008 at 9:36 AM | Permalink

    bender and others, I’m struggling to limit this thread to things climatic.

    conceptually, I can picture circumstances where glacier melt is substantial proportion of runoff and circs where it isn’t. The Ottawa River of 10000 BP was Amazon-sized due to the recession of the Laurentide glaciation. The modern Mississippi has negligible contributions from glacier recession.

    On a worldwide basis, I’m satisfied there is actual glacier recession in the 20th century and it’s idle to argue that some glaciers are advancing and others are retreating as some kind of supposed rebuttal of this.

    I presume that everyone agrees that during a period of glacier recession, runoff will be larger than the annual precipitation by some amount. The practical question is whether the percentage is trivial or material.

    bender argues that contribution from receding glaciers in the Canadian prairies is non-negligible. If so, in practical terms, all this suggests to me is that this particular area is experiencing a non-recurring and non-sustainable supply of water – sort of like mining an aquifer.

    As far as I can tell, this may well be a real long-term problem, but one that would exist anyway. If global warming stopped and the glaciers stopped receding, the runoff would be reduced.

    My guess is that the percentage contribution of receding glaciers in India is negligible and that most of the runoff is from current precipitation. However, if this is incorrect and some substantial proportion is from “mining” the meltwater, then isn’t this a potential problem whenever the recession stops for whatever reason?

  71. bender
    Posted Jan 3, 2008 at 10:36 AM | Permalink

    My advice is to close the thread as it has served its limited purpose. ,,, Instead of discussing cliamte science we meandered into areas that have little to do with climate change and everything to do with adaptation to accommodate overdevelopment.

One Trackback

  1. […] Climate Audit: Water Availability […]

%d bloggers like this: