Why Is Winter Snow Extent Interesting?

Guest post by Steven Goddard

Several people keep asking why am I focused on winter snow extent.  This seems fairly obvious, but I will review here:

  1. Snow falls in the winter, in places where it is cold.  Snow does not generally fall in the summer, because it is too warm.
  2. Winter snow extent is a good proxy for winter snowfall.  Snow has to fall before it can cover the ground.

So what about summer snow cover?  Summer snow cover declined significantly (from the 1970s ice age scare) during the 1980s, but minimums have not changed much since then.  As you can see in the graph below, the overall annual trend since 1989 has been slightly upwards.

click to enlarge

Data from Rutgers University Global Snow Lab

Note in the image above that there has been almost no change in the summer minimum snow extent since 1989, and that the winter maximums have increased significantly as seen below.

Summer snow cover is affected by many factors, but probably the most important one is soot, as Dr. Hansen has stated.

The effects of soot in changing the climate are more than most scientists acknowledge, two US researchers say. In the Proceedings of the National Academy of Sciences, they say reducing atmospheric soot levels could help to slow global warming relatively simply. They believe soot is twice as potent as carbon dioxide, a main greenhouse gas, in raising surface air temperatures. … The researchers are Dr James Hansen and Larissa Nazarenko, both of the Goddard Institute for Space Studies, part of the US space agency Nasa, and Columbia University Earth Institute.

http://news.bbc.co.uk/2/hi/science/nature/3333493.stm


The global warming debate has until now focused almost entirely on carbon dioxide and other greenhouse gas emissions, but scientists at the University of California – Irvine, suggest that a lesser-known problem – dirty snow – could explain the Arctic warming attributed to greenhouse gases….The effect is more conspicuous in Arctic areas, where Zender believes that more than 90 percent of the warming could be attributed to dirty snow.

http://www.scienceagogo.com/news/20070506202633data_trunc_sys.shtml

In summary, winter snowfall is increasing and currently at record levels, and summer snow extent is not changing much.  Earlier changes in summer snow extent were likely due primarily to soot – not CO2.

Why Is Winter Snow Extent Interesting?

Several people keep asking why am I focused on winter snow extent.  This seems fairly obvious, but I will review here:

1. Snow falls in the winter, in places where it is cold.  Snow does not generally fall in the summer, because it is too warm.

2. Winter snow extent is a good proxy for winter snowfall.  Snow has to fall before it can cover the ground.

So what about summer snow cover?  Summer snow cover declined significantly (from the 1970s ice age scare) during the 1980s, but minimums have not changed much since then.  As you can see in the graph below, the overall annual trend since 1989 has been slightly upwards.

Data from Rutgers University Global Snow Lab

Note in the image above that there has been almost no change in the summer minimum snow extent since 1989, and that the winter maximums have increased significantly as seen below.

Summer snow cover is affected by many factors, but probably the most important one is soot, as Dr. Hansen has stated.

The effects of soot in changing the climate are more than most scientists acknowledge, two US researchers say. In the Proceedings of the National Academy of Sciences, they say reducing atmospheric soot levels could help to slow global warming relatively simply. They believe soot is twice as potent as carbon dioxide, a main greenhouse gas, in raising surface air temperatures. … The researchers are Dr James Hansen and Larissa Nazarenko, both of the Goddard Institute for Space Studies, part of the US space agency Nasa, and Columbia University Earth Institute.

http://news.bbc.co.uk/2/hi/science/nature/3333493.stm


The global warming debate has until now focused almost entirely on carbon dioxide and other greenhouse gas emissions, but scientists at the University of California – Irvine, suggest that a lesser-known problem – dirty snow – could explain the Arctic warming attributed to greenhouse gases….The effect is more conspicuous in Arctic areas, where Zender believes that more than 90 percent of the warming could be attributed to dirty snow.

http://www.scienceagogo.com/news/20070506202633data_trunc_sys.shtml

In summary, winter snowfall is increasing and currently at record levels, and summer snow extent is not changing much.  Earlier changes in summer snow extent were likely due primarily to soot – not CO2.

About these ads

254 thoughts on “Why Is Winter Snow Extent Interesting?

  1. Quite possible, however temperatures seem to be falling on landmasses as well and so the soot problem might turn out to be a good thing.

  2. Just to be clear and avoid getting off on irrelevant tangents, snow cover did decrease from 1970 through 1989. Winter snow cover has returned to the unusually high levels of the 1970s. Summer snow cover has not returned to 1970 levels, but has not declined since 1989.

  3. It is interesting how Henson tries to do the global warming spin.If the first theory doesn’t work then change theory.Folks these people could care less about what is really happening .The planet is cooling .But this goes against thier theory and goverment control.

  4. Could we get a close up view of the summer snow extent, like we did for the winter to see what tend looks like over the same period?

  5. Soot – mostly man made. yet somewhat easy to cleanup. does not stay around for hundred years, washes out with the melting. tackle this first, then see where we are….IMHO

  6. Is the first graph the one you meant to accompany this post? It doesn’t match the description.

    PS. I just woke up, so maybe my brain ain’t workin’ yet.

  7. We may be reaching a tipping point for the next glacial period. Humans are surely to blame. I just haven’t quite figured out how yet.

  8. So there is more humidity in the Northern atmosphere during winters – so what?
    That is doing what in relation to climate change?

    It feels like reading the coffee pot – after having had one coffee, only.

  9. Zender had a little lamb,
    Its fleece was black as soot.
    And everywhere that Zender went,
    It’s sooty foot it put.

  10. Leif,

    Please feel free to do any statistical analysis you want, and share the results with us. If you want to repeat for the last 10 years that would also be appreciated.

  11. The first graph X-axis shows weeks since 1966. It is pretty easy to pick out winter maximums and summer minimums.

  12. Expanded winter snow-cover must increase the albedo-effect. I’d like to see a discussion by people wiser than myself about what the difference might be.

    I have some questions regarding the effect the bounced-back sunlight would have on the temperature of the atmosphere.

    I assume that when snow reflects sunlight the sunlight doesn’t change into infrared waves, but remains the same long(?)-wave radiation it was on the way down. Therefore it would not “excite” CO2 on its way out. Is this assumption correct?

    Does such long(?)-wave radiation warm the atmosphere at all, as it passes through?

    If it does, would not the albedo-effect increase the warming of the atmosphere, because the light would warm on its way in, and also on its way out?

    If this was true then, even as expanded snow-cover reflected a lot of heat back into outer space, we might see some slight warming of the troposphere. This might lead to the false impression things were getting warmer, even as we lost a lot of heat to outer space.

    Are Oxygen and Nitrogen warmed at all, when sunlight passes through them?

    If my ideas are way off base, I don’t mind being educated in the slightest. Thanks in advance.

  13. Steve Goddard (07:45:02) :
    Please feel free to do any statistical analysis you want, and share the results with us. If you want to repeat for the last 10 years that would also be appreciated.
    Come on. You have that data handy. And have software [Excel?] to put in the trend line. Just ask it for R^2 to gain some credibility.

  14. I notice that you only cite NH snow extent. What does SH look like?

    Does measuring snow extent give any indication of temperature trends in Arctic and Antarctic? Or is this more of a measure of mid-latitude variation?

    What would slope for winter snow cover look like without 2010, i.e., how much flatter might it be?

  15. “Christian A. Wittke (07:38:51) :

    So there is more humidity in the Northern atmosphere during winters – so what?
    That is doing what in relation to climate change?”

    The post doesn’t talk about “humidity in the Northern atmosphere”. How do you conclude that the humidity has risen? And while we’re at it, do you think that the mean humidity has risen? The peak humidity? Both? In which layers of the atmosphere? Any idea?

    Or have you been confused by The Time with their “In the meantime, warmer air could be supercharged with moisture and…”
    at

    http://www.time.com/time/health/article/0,8599,1962294,00.html?xid=rss-fullhealthsci-yahoo

    (which was pretty hilarious, i mean, SUPERCHARGED WITH MOISTURE. Like, more than 100% – which usually leads to precipitation, even when The Time doesn’t notice.)

  16. Leif Svalgaard (07:16:08) :

    I was going to write something about fish and barrels, but the analogy rather breaks down when the fish start shooting themselves.

  17. Snow cover is God’s gift to winter wheat fields. It is so many times better than mere rain fall.
    As we read about the permanent dustbowl and reality shows us great ground cover for wheat.
    Wheat can grow in the subsoil zones when covered by insulating snow.

  18. For coal-fired power stations, and perhaps to a lesser extent for other soot-generators, it should be cheaper to reduce the soot emissions than to extract and sequester the CO2 generated and emitted. Electrostatic precipitation has a head start, but CO2 separation and sequestration is only just out of the starting box in recent years. Hohlfield observed in 1824 that an electric discharge cleaned dust laden gases, and not too many years later, Lodge, and later Cottrell made and installed practical dust precipitators for smelter fumes. In 1935, Penney announced the development of the Precipitron, with a major reduction is size and voltage and an improvement in efficiency. This horse is already half way to the finish line.

    Surely we can work from this as a starting point, and get more bang for our warming reduction buck by reducing soot instead of CO2? For a start, simply slowing the exhaust gas velocity through larger precipitators would reduce soot emission. It should be a simple matter to calculate the improvement that can be achieved per dollar spent, even without new developments. Considering the millions of dollars being thrown at CO2 studies, let alone action, we could not only be reducing AGW right now, if in fact it exists; we could reduce real pollution and breathe cleaner air. What were those numbers for coal-fired plants in construction or planned, again?

  19. “Caleb (07:47:41) :
    I assume that when snow reflects sunlight the sunlight doesn’t change into infrared waves, but remains the same long(?)-wave radiation it was on the way down. Therefore it would not “excite” CO2 on its way out. Is this assumption correct?”

    Yes. Only you got the “long” wrong – the longwave part of the spectrum we usually talk about here is the Longwave Infrared that is absorbed by CO2 and to a larger extent by water vapour, > 1 mikrometer. Visible light has shorter wavelengths, around 350nm IIRC.

    “Does such long(?)-wave radiation warm the atmosphere at all, as it passes through?”

    No, not in a significant way, as it is not absorbed.

    “If it does, would not the albedo-effect increase the warming of the atmosphere, because the light would warm on its way in, and also on its way out?

    If this was true then, even as expanded snow-cover reflected a lot of heat back into outer space, we might see some slight warming of the troposphere. This might lead to the false impression things were getting warmer, even as we lost a lot of heat to outer space.

    Are Oxygen and Nitrogen warmed at all, when sunlight passes through them?”

    No interaction between visible light and O2 or N2, otherwise air wouldn’t be transparent for visible light. So visible light that is reflected e.g. by snow will go outwards. There might be refraction by water droplets, so clouds will diffuse visible light. In the absence of clouds, the light reflected by snow (or white roofs, or any part of the surface) will just radiate into space without warming anything significantly.

  20. Leif,

    R^2 correlation wouldn’t make sense on a X vs T plot. Its typically used to compare the degree of correlation of two separate variables (i.e. on a scatter plot).

  21. Norway has already given lots of money during many years to the Soviet’s to put filters on them large factories up in Nickel ….they never used it to filter the smoke anyway….the whole area was black for kilometers after kilometers.

    Dont know how it is at the moment…. but back then you could see when you crossed the border to Soviet-land.

    Free country; Clean.
    Communist land; Dirty.

  22. Richard,

    The analysis starts in 1989, because that is when the current upwards trend started. The 1970s was cold and had an unusually large amount of snow. Snow cover declined during the 1980s from the 1970s peak.

    Winter snow cover has now returned to the record high values.

  23. Trends aside for the moment, this past winter is reaching for the snowfalls not seen since the winter of 1977-78 (as has been well covered). What do this year and the winter of 1977-78 have in common? An El Nino combined with a near-neutral to negative PDO index. That notable winter was the last time this combination happened.

    Just as the causes of the PDO’s behavior is not well understood, I am of the opinion that it’s teleconnections are poorly understood, as well. (no surprise, it is a 60 year cycle and we have measured one half of one cycle via global coverage by satellite…and there is no guarantee that this particular cycle is representative of the norm as variability exists in the cycles themselves)

    PDO index by month: http://jisao.washington.edu/pdo/PDO.latest

    PDO index history plot here (only through Sept 2009): http://jisao.washington.edu/pdo/

    another plot…a little experimental: http://www.esr.org/~gunn/pdo/pdo_fig1.png

  24. Leif,

    While the trend is slightly upwards, there has been no statistically significant trend in weekly snow cover over the last 20 years. Winter maximums have increased significantly while summer minimums have stayed relatively constant.

    It has been claimed repeatedly in the press that snow cover is declining, but the evidence does not support that over the last two decades. A Google search of snow cover decline turns up over one million hits.

    What is your point? Do you think that winter snow cover has not been increasing and is not at or near a record maximum?

  25. Matt (08:17:53) :
    R^2 correlation wouldn’t make sense on a X vs T plot. Its typically used to compare the degree of correlation of two separate variables (i.e. on a scatter plot).
    R^2 value does make sense in evaluating a trend. A trend is just that: a correlation of a variable with time. Remove the connecting lines [which are artificial anyway] and the graph is indeed just a scatter plot.

  26. Steve Goddard (08:30:34) :
    While the trend is slightly upwards, there has been no statistically significant trend in weekly snow cover over the last 20 years.
    What is your point? Do you think that winter snow cover has not been increasing and is not at or near a record maximum?

    My point is that it is a bit silly to open a post with a graph that has no significant trend with the words: “As you can see in the graph below, the overall annual trend since 1989 has been slightly upwards”
    And we still do have any explanation why the first 20+ years were left out. If you leave out all the data except the last two years, you’ll get quite a correlation. So what was R^2? You are beginning to sound like Mann who wouldn’t reveal his R^2 either :-)

  27. I am completely at a loss as to what some people are disputing.

    We have been told over and over again since Hansen spoke to Congress in 1989, that winter snow is declining and that winter storms are moving north.

    Clearly that is not happening, quite the opposite.

  28. Steve Goddard (08:18:41)

    There is nothing wrong with that. Provided you make it explicit what you have done, and acknowledge that by selecting the start point, you inflate the risk of a type I error. The post, as it currently reads, is deceptive. I am not implying that this is your intent. Plotting the entire dataset would be a good idea to clarify matters.

  29. I’ve made my own Excel analysis of the data. The R^2 for the graph above is 0.3, and the slope is 87,000 km^2 increase in Winter snow cover per year. But, this is very sensitive to the choice of starting year. 1989 is a cherry-picked minimum, and the current year’s data is not completely in yet. If you plot from the cherry-picked maximum of 1978-2009 you get a decrease of 8,700 km^2 per year and an R^2 of 0.0035.

    Using the Rutger’s data from 1974 onward, the slope of the trend is 19,000 km^2 increase per year, on a standard deviation of 1,400,000 km^2 per year, with an R^2 of 0.02, which is, for all practical purposes, flat.

  30. @ Tom P (08:01:37)

    Tom P, that doesn’t add anything to the conversation, and is actually insulting.

    Reminds of the comments when I used to read RealClimate.

  31. John Phillips [above] suggests we may be at the beginning of earth’s next glaciation of the northern hemisphere. I see that possibility too. It is overdue, geologically speaking. The short chart (last 20 years of winter snow extent) would fit a logarithmic increase, which is just what would be expected for the beginning of the creep of the ice sheets south. (Lookout Manhattan Island!)

    However, as I recall my geology lessons, the northern ice sheet should need to completely disappear so as to provide sufficient extra open water surface to provide sufficient moisture for the extensive increase in snow, to provide the extensive increase in snow cover, which keeps driving the temperature down with increased albedo and all the water phase change, keeping the ice sheets growing south until the earth cools too much to keep the precipitation cycle going.

    Of course, there were those making similar statements in the 1970s.

    Hopefully it will take centuries (perhaps millennia) to cool enough regardless.

    We still have engineers. I suppose I’m not worried.

  32. >> DirkH (08:15:05) :
    No interaction between visible light and O2 or N2, otherwise air wouldn’t be transparent for visible light. So visible light that is reflected e.g. by snow will go outwards. There might be refraction by water droplets, so clouds will diffuse visible light. In the absence of clouds, the light reflected by snow (or white roofs, or any part of the surface) will just radiate into space without warming anything significantly. <<

    Visible photons are not absorbed by N2 or O2, but they are scattered by collisions, moreso for shorter wavelengths (hence the blue sky). The scattering process has to add some momentum, and thus kinetic energy, to the air molecules.

    Hmm, as I write this I realize that the photon changes direction, but doesn't lose energy since the wavelength doesn't change. How is energy conserved when the air molecule gains kinetic energy but the photon energy remains the same?

  33. “Earlier changes in summer snow extent were likely due primarily to soot – not CO2.”

    If Zender is correct, yes. And what is the CAUSE of all that soot, accord to Zender? From the abstract:

    “Applying biomass burning BC emission inventories for a strong (1998) and weak (2001) boreal fire year, we estimate global annual mean BC/snow surface radiative forcing from all sources (fossil fuel, biofuel, and biomass burning) of +0.054 (0.007–0.13) and +0.049 (0.007–0.12) W m−2, respectively. Snow forcing from only fossil fuel + biofuel sources is +0.043 W m−2 (forcing from only fossil fuels is +0.033 W m−2), suggesting that the anthropogenic contribution to total forcing is at least 80%.”

    At least 80% of the soot is caused by BURNING FOSSIL AND BIOFUELS. So his theory is very interesting, and I hope it holds up, since it suggests the Arctic ice could rebound much more quickly if we took corrective action. But that corrctive action continues to be drastically reducing burning carbon compounds for energy.

  34. In the 1970′s when worries of “the coming ice age” were all the rage, they were watching the snow extent closely as a leading indicator of possible cooling. The concept was based on the albedo changes, and the snow’s effect on heat storage on the land mass areas.

    A reasonably good analogy would be to think of the land mass as a thermal capacitor, charged with thermal energy by an AC signal.

    A given parcel of land grows warmer through the summer, then cools during the winter, eventually reaching some thermal equilibrium, where each year on average its temperature oscillates around a uniform temperature. In the northern temperate region this temp is approximately the ground water temperature and in much of the U.S. is about 55 deg F. (12.8 deg C).

    The period of the global cooling concerns was also about the time the TAPPS study came out for Nuclear Winter, and one of the results of that were analysis of the effect of albedo changes on this stable thermal equilibrium temperature.

    If you get early snows in the fall, and late snows in the spring, the increased albedo at times when the sun’s elevation angle is high would have much more impact on the thermal energy balance than the same snow extent occurring in the dead of winter.

    A deep long lasting snow late in the spring would drastically slow the early spring heating (which would slow the entire summers heat storage). If you also had deep long lasting snow early in the fall, that would cut off solar heating of the ground early.

    The worst possible situation would be deep long lasting snowfalls early in the fall, and late in the spring and very little snow (thin snow cover) in the depths of the winter. This would minimize heat insulation effects of the snow in the peak cold of the winter while still maintaining high albedo.

    In the late 1970′s early 1980′s we had a nearly snow less winter here in Colorado and had major problems with deep freeze lines in the soil, causing all sorts of problems with frozen water lines etc. as the sharp cold temperatures in January and February froze the ground. Construction nearly ground to a halt over the winter as the ground was too hard to dig even with back hoes. A friend of mine spent that winter welding new teeth on backhoe buckets for a local construction company.

    Any way back to the snow extent. One of the theories that came out of all this examination of albedo and timing/extent of snow was the theory of the “snow blitz” trigger for entering an ice age. Basically they suggested that a couple short summers due to heavy snows at the right time, could basically cut off summer time heating enough to trigger a plunge in seasonal average temperatures and that global temperatures would follow.

    This pattern would have to repeat for several years running and in both hemispheres to trigger a global ice age. The snow blitz theory was one attempt to explain what triggering event pushed the climate over the edge into the actual ice age once the orbital cycles were in the proper phase to support long term glaciation. Some archelogical evidence showed that the changes in flora and fauna at the time of the shift into glacial dominance might happen very quickly (perhaps decades or even a few years). This led to hypothesis’s like the snow blitz hypothesis to explain the sudden change in prevailing weather, that lead to the change in the general climate. On the oceans the same theory applies to ice cover, as its extent late in the fall and late into the spring would strongly effect solar heating of the water beneath the sea ice do to absorption of solar energy. In that case the sea ice switches from acting as an insulator to keep the polar seas warm in the dead of the winter into sun shades reflecting away heating in the early and late part of the local summer seasons.

    http://www.masterresource.org/2009/09/the-global-cooling-scare-revisited/

    http://news.google.com/newspapers?nid=1346&dat=19750105&id=HI0sAAAAIBAJ&sjid=-_oDAAAAIBAJ&pg=7160,928089

    Larry

  35. I don’t think that a Trend is even necessary, the human eye can see that there is nothing realy going on with the Summer graph.

  36. Steve, points #1 and 2# are great, but you need a point #3:

    3. Because warmer air holds more moisture, it can be too cold for snow. Antarctica, very example, is in many places too cold for snow. Warmer conditions may increase snow, if they make it warmer but not too warm for snow.

  37. Steve,

    You got your data from Rutgers University Global Snow Lab.

    Have you taken the time to examine their snow cover anomaly chart? Its well done, and tells a bit of a different story:

    http://climate.rutgers.edu/snowcover/chart_anom.php?ui_set=0&ui_region=nhland&ui_month=1

    Similarly, just winter extent:

    http://climate.rutgers.edu/snowcover/chart_seasonal.php?ui_set=nhland&ui_season=1

    As Anthony is fond of saying, another entry for the “weather is not climate” department.

  38. @DirkH

    Did we read the same post?

    Snow is atmospheric humidity and temperatures below 32°F – or am I wrong?

    We have more snow in winters since 1989 now – on the Northern Hemisphere, that’s what the graph suggests and what it feels like – but only this winter in Scotland, not the last 10 or so years.

    Are you suggesting that we have more snow because we have the same amount of humidity in the atmosphere but the temperatures dropped considerably so what was rain is now snow? Is there any data on that?

    If the temperatures have not dropped and the humidity has not increased then where does the snow come from?

    From supercharging?

  39. Leif Svalgaard

    I grabbed the data rather crudely from the graph with an on-screen ruler, and looked at the curve in Excel.

    If we accept an approximation as a two factor scatter, then we get an R2 of 0.307.

    Taking the square root of this as r the correlation coefficient, we get 0.554.

    Applying an r of 0.554 to the correlation table, taking (also a crude approximation) the number of years minus 2 as the degrees of freedom, then the P value is between 0.01 (r=0.537) and 0.001 (r=0.652).

    If I’ve cocked this up, let the vultures descend!

  40. “John Innes (08:12:49) :

    For coal-fired power stations, and perhaps to a lesser extent for other soot-generators, it should be cheaper to reduce the soot emissions than to extract and sequester the CO2 generated and emitted. …

    Surely we can work from this as a starting point, and get more bang for our warming reduction buck by reducing soot instead of CO2? …”

    Well put. Let’s save people who are now inhaling enough soot to kill a half million people per year instead of worrying about what people 4 generations down the road will do to respond to gradual sea level rise of less than a foot per century.

  41. R^2 on the winter graph since 1989 is 0.298514013

    R^2 on the winter graph since 1999 is 0.312822303 Slope is 317,000 km2/year

    R^2 on the weekly graph since 1966 is 0.004697147

    R^2 on the weekly graph since 1989 is 5.25176E-05

  42. Zeke Hausfather (09:55:07) :
    Notice they always use Anomalies, but anomalies from what, there is no “normal” in earth’s climate. They always choose the “Base”. Try using the MWP as the base and everything changes. Try using the period when the vikings were in Greenland instead and see what you get.

  43. Leif,

    Couple points – 1. its very rare to use an R^2 value in a time series, unless you expect the relationship to be linear or monotonic. 2. Applying an R^2 to a linear fit of an oscillating system also makes no sense whatsoever! Its exactly things like this that cause skeptics to not be taken seriously – you fail in basic statistics, but you expect to be able to understand a complex climate system? Give me a break!

  44. Christian A. Wittke (10:01:43) :
    The world is pretty big, and the climate is wonderfully complex. Dirk H. knows his physics, and is gentle with folks that don’t know it as well.

    I tend to be a bit more blunt. Be careful using single point variables like atmospheric humidity and temperature to try to describe a giga-variabled system.

  45. A C Osborn,

    Setting a different period as the ‘base’ wouldn’t change the shape of the graph, only the zero-intercept of the data. The result is still the same.

  46. Steve,

    If you plot the average Dec-Feb snow extent against the PDO index, you get a very interesting correlation…

    Winter Snow Extent vs PDO

    As far as the r-squared goes… It’s a cyclical stochastic process. Unless you have a long enough time series, no reasonable trend line will yield a decent r-squared.

  47. phlogiston (10:08:42) :
    if we accept an approximation as a two factor scatter, then we get an R2 of 0.307.
    A good way to think of R2 is that it is simply the fraction of the variation that is due to the trend [if the data points are independent - which they are not]. So, an R2 of 0.1 means that at least 90% of the variation is not due to the trend, and R2 of 0.3 means that at least 70% of the variation is not due to the trend. The ‘at least’ comes from the fact hat the autocorrelation at log 1 is not zero.

  48. Leif S.

    “As you can see in the graph below, the overall annual trend since 1989 has been slightly upwards.

    And what is R^2 for that graph?”

    Interesting. I don’t recall your EVER asking this for testing out the significance of temperature data. As I have been doing, using Student’s T, Mann Whitney, etc. to decide if certain observed changes (comparing year to year, decade to decade and century to century) are truely “statistically significant” considering the observed standard deviation. (Of course, these is the concept of normalacy and the validty of Student’s versus the non-parametric Mann Whitney test, my judgement is out on that. Because if you treat temperatures by “seasons” they are virtually “normal” in distribution. If you go over a chronological year, they appear as a bifurcated normal distribution.)

  49. Steve Goddard (10:20:59) :
    R^2 on the winter graph since 1989 is 0.298514013
    R^2 on the winter graph since 1999 is 0.312822303

    An interesting graph would be R^2 as a function of start year. And another one the same but without 2009-2010.

  50. Just so I cam see if I am following this, I’d appreciate if anyone would tell
    me if I have misunderstood Steve’s post.

    Steve Goddard (10:20:59) :
    R^2 on the winter graph since 1989 is 0.298514013
    ————Means snow cover has increased somewhat since 89.

    R^2 on the winter graph since 1999 is 0.312822303 Slope is 317,000 km2/year
    ———Means considerable increase sine 1999

    R^2 on the weekly graph since 1966 is 0.004697147
    ——–Means when averaged out since 1966, no big deal
    indicating (opinion) that maybe the climatologists have mistaken natural variability ( caused by xxxxx?) for one way climate change.

    R^2 on the weekly graph since 1989 is 5.25176E-05
    —–Showing that measured from 1989 there has been a significant increase in snow cover.

    Do I at least read this right ?

  51. I don’t see the growth represented on second graph on the first one, it looks pretty much constant.
    Also, 2010 isn’t quite finished, it just started, so how was calculated 2010 point on second figure? it seems it will look pretty much the same as 2008.
    Any way is hard to look for differences on sinusoidal signals, I like better the anomalies graph as in http://moe.met.fsu.edu/snow/ though their record isn’t as long, it only goes down to year 2000

  52. Hmm. Living in a “snow” country as I do, I’m less than convinced that snow extent is anything more than a weak proxy for flow of cP and Arctic air masses. Its certainly not a prioria linear proxy for temperature nor for global cooling. While we’ve observed snowfall along a lower latitudinal limit this year than in the previous few years, behind the snow line there are large areas with much lower than usual snowfall, and more moderate temps. At my place in S. Ont., for example, we’ve way below normal snowfall and temps have stayed moderate all winter (but not warm, i.e. +/- average without the usual large variation swings).

    I’m a little less sanguine about “records” as well. I seem to recall the Washington record this winter was ~ 1/2 in. more snow than the previous record. 1/2 in? A 1/2in of snow measure is meaningless in a pile several inches thick, as anyone who lives with snow will tell you. This is an anecdotal record, not a scientific one.

  53. vukcevic (07:36:13) :
    Keeps deep frost away from the ground surface. Any farmer will know benefits of the winter snow cover.

    Henry chance (08:05:49) :
    Snow cover is God’s gift to winter wheat fields. It is so many times better than mere rain fall.

    The only advantage to snow versus rain is in cases where slow snowmelt occurs leading to increased percolation resulting in reduced runoff in spring (March, April, May).

    Frost actually mellows the ground due to frost heave and aids in spring sowing. It also kills insect eggs and larvae in the soil, and some weeds and weed seeds.

    A persistent, deep snow cover winterkills winter annuals (wheat, rapeseed, winter peas, etc.).

    So no, we don’t want increased long lived snow cover in the US at least.

  54. Robert (09:24:32) : “At least 80% of the soot is caused by BURNING FOSSIL AND BIOFUELS. So his theory is very interesting, and I hope it holds up, since it suggests the Arctic ice could rebound much more quickly if we took corrective action. But that corrctive action continues to be drastically reducing burning carbon compounds for energy.”

    …by inefficient soot-producing methods and instead burning them in modern efficient power plants, and switching from wood/grass/dung cooking fires to natural gas, and producing abundant energy from nuclear power, etc.

    Fixed that for you, Robert. No thanks necessary.

  55. Only thing that makes any sense here is:
    delta area/delta time
    or to make more sense trend expressed as % pa, which is roughly 0.22% pa
    (no calculator or excel needed).

  56. Matt (10:36:58) :
    Applying an R^2 to a linear fit of an oscillating system also makes no sense whatsoever!
    If you consider the system to be an oscillation riding on a linear trend, then it does make sense to compute R^2 as a measure of the significance of the trend. In interpreting R^2 you must then take into account the non-zero autocorrelation due to the oscillation. There are standard techniques to do all this. One way of looking at this is to average over each cycle and plot the cycle averages as a function of time. This makes perfect sense. R^2 will still show how much of the variation is due to the trend. Here is a little experiment: http://www.leif.org/research/Cycles-with-Trends.png
    I generated 17 perfect sine waves [black curves] and ran a linear regression. The result was y = 0.00 x + 0 with R^2 = 0 as expected. No trend. Then I added a linear trend of 0.01 per step in x [red curve], result of linear regression: y = 0.01 x + 0 with R^2 = 0.1525. Perfectly reasonable as the trend explains 15.25% of the variation. Then I added a trend of 0.02 [green curve], result y = 0.02 + 0 with R^2 = 0.4185, i.e. the trend [which is recovered perfectly from the analysis] explains 41.85% of the variation, etc.

    No problem here whatsoever. No failing of statistics. For the green curve, t is 27.3 and p = 4*10^(-124). The significance is enormous for this case [I had 1037 points]. With fewer data points the significance drops dramatically: with 23 points t=7.92 and p= 0.00063, but then there is no noise, etc.
    R^2 is a good measure of how much of the variation is due to a linear trend, no matter what the data looks like.

  57. The global warming debate has until now focused almost entirely on carbon dioxide and other greenhouse gas emissions
    What “global warming debate”? There isn’t any. It has been debunked. It’s over!

  58. Steve Goddard (10:20:59) :

    A better R^2 on your second graph of winter snowfall would be to add every point with preceding point and plot as two-year accumulated snowfall. That R^2 should show much better correlation and is still proper statistics if you state the results properly. See stat book for proper title, seems would be “The Two Year Accumulated Winter Snowfall”. Could do it for three-year of more. Might uncover a hidden high-freq component.

  59. Sorry folks, I’m normally an avid fan of this superb website. But this discussion strand is futile. We sneer at the crazy warming gang when they extrapolate a small dataset to the point of absurdity; please let’s not do the same.
    It has snowed a little, so what?. Two swallows do not a summer make.

  60. Steve Goddard (10:20:59) :

    > R^2 on the winter graph since 1989 is 0.298514013

    9 significant figures? Careful – Leif will start teaching you about significant digits and error bars.

  61. Even intelligent, educated, and well-meaning people can convince themselves of all kinds of wrong things when analyzing large volumes of data from multiple experiments. It takes extraordinary discipline to keep oneself from being led astray by phantoms.

    What has been demonstrated here is that the northern Winter snow extent has very large year-to-year variability, and can not be used to support a warming trend, or a cooling trend, during all or part of the last 40 years. That is itself a useful finding.

    Some of the remarks made here regarding statistics have been mistaken. R-squared, in particular, is perfectly valid as a test of the assertion that time is responsible for some of the variation. But for year-on-year variations, it is necessary to annualize it first, either with seasonal averages, as above, or some other filter that removes the yearly cycle. Applying it to the raw weekly Rutgers snow cover data for the last 40 years makes no sense, but to a Dec-Feb annual average, or a 52-week average, makes perfect sense.

  62. @ Sordnay (11:05:01) :

    I’m trying to figure out how that is representative of reality. On Feb 12 we had one of the largest winter storms in the southeast in a long time, and it barely registers on that chart, if at all? I mean heck, you can look here:

    http://ge.ssec.wisc.edu/modis-today/index.php?satellite=t1&product=true_color&date=2010_02_13_044&overlay_sector=false&overlay_state=true&overlay_coastline=true&sector=USA7&resolution=2000m

    And see snow cover (howbeit shortlasting) all the way to north florida, but somehow that doesn’t even register on the chart. I wonder how they calculate ‘added snow cover’ because that chart does not seem indicative of what is on the ground right now in many parts of the South East. Does that cover has to remain on the ground for a week for it to register? Doesn’t seem to jive with the reality to me, at least.

  63. Sorry I just noticed green is ‘climatology from 95-09′ so it does make a little more sense on the graph, even if the blue/red chart doesn’t

  64. @ Brent Hargreaves (12:04:02) :
    It ‘snowed a bit?’ This is the second highest extent in the northern hemisphere on record.

  65. Caleb,

    Snow is not a perfect reflector. In fact, because of it’s structure, fresh snow acts as a black body. The high energy shortwave radiation from the sun rapidly melts the surface which freezes to form an ice crust that is more of a reflector. But even ice is transparent like glass and the fraction of the energy reflected back to space depends on the angle of incidence. It does radiate low energy longwaves as a function of surface temperature. Much of the outbound longwave radiation from the top of the atmosphere measured by satalites is frozen cloud tops.

  66. Quote: Steve Goddard (08:42:01) :

    1. “I am completely at a loss as to what some people are disputing.”

    Reply: Someone – skilled in bamboozle propaganda techniques – is purposely directing attention away from undisputed facts that would embarrass NASA and/or Hansen.

    2. “We have been told over and over again since Hansen spoke to Congress in 1989, that winter snow is declining and that winter storms are moving north.”

    Reply: That is exactly what the propaganda artist seeks to hide.

    3. “Clearly that is not happening, quite the opposite.”

    You are exactly right.

    Since that fact can’t be disputed, a talented propaganda artist might direct attention to R^2, 4/3(pi)R^2, PV = nRT, or anything else to bamboozle the audience.

    Thanks, Steve Goddard, for this post.

    A least one of your critics is more talented in propaganda than in science.

    With kind regards,
    Oliver K. Manuel

  67. I’m sorry Steve Goddard, but I think that the conclusion is largely cherry-picked from the time series and the trend is likely to be spurious. Yes 2010 appears to be a large extension of winter snowfall, but its one season.

    As Leif points out, the low R2 of the points says that most of the apparent trend is scatter and not significant.

    I think we have to wait a lot longer to see proof that the decline of NH snow cover since the 1960s has been broken.

  68. 1) The data does not go back that far. Like much of the over-hyped climate and weather records, it’s just no enough data to get excited about.

    2) There is no increase in snowcover (predicted by AGW proponents) .

    3) There is no decrease in snowcover (also predicted by AGW proponents).

  69. Question

    Why is the arctic ice extent so low for the time of year despite recent cool weather? Despite cool temperatures over most of the Arctic Ocean in January, Arctic sea ice extent continued to track below normal. By the end of January, ice extent dropped below the extent observed in January 2007 etc, etc?

    http://uk.answers.yahoo.com/question/index?qid=20100203092427AAzTQYz

    Looks like some-one needs to ask Yahoo UK a supplementary question:

    If there is global warming why has Arctic Ice Extent risen above that of 2005-6 and 2006-7?

  70. Steve Goddard:

    Is the Arctic Sea Ice anomaly graph at the following link actually saying the same thing that you’re saying about the snow? -

    If so, or not, could you help me understand how your graph and the one at this link fit together. TIA

  71. If I was to look at the top chart as plotting amplitude, while a sinusoidal signal was controlling the gain, I would wonder what all the distortion at the top of the range is about. Is the amplifier maxed out? Otherwise, as charted, I don’t see either way what all the noise is about.

    (Pun originally unintended, but I saw it and left it in. If people are willing to let the IPCC get away with it, well then…)

  72. For an historical last snow remaining on a 7,000′ peak 1902-1981:

    The Young family kept this record. It corresponds roughly to the US snow extent minimums where they overlap in time.
    Are there other records of this type in the US?
    A proxy could be built.

  73. Brent Hargreaves: “It has snowed a little, so what?. ”

    The issue is that the Team is claiming it has snowed a lot less. Thats a lie.

  74. Total amateur here. The last time I took statistics, we didn’t have Excel to do this for you…hell, we didn’t even have graphing calculators. So be gentle. :)

    I’m afraid to post this, but I did the entire thing, including the first 10 weeks of 2010, and I got a flat trend line with an r^2 of 1.0. That’s 43 years of data.

    I didn’t want to, but I averaged the yearly values–all I have is Excel and it won’t let you exceed 255 lines of data, and this has almost 2300. And I was in a hurry–I’ll double check it again later. If anyone knows how to increase the number of points Excel will let you work with, please let me know.

    There are spikes in 1966 (partial year), 1969, and 1971. After that, it drops sharply to 1974, comes up in 1976, and then stays relatively flat till this year, but it’s only February so we’ll just have to see how that plays out. The minimum is in 1990.

  75. Brent-

    I agree. This is kind of a joke. Going back to 1989? Give me a graph that goes back 200 years or more and we’ll start to talk. I’m 100% anti AGW and man induced warming, but we need more then this.

    We will get another ice age guys, it will happen, just give it time. Anyone who thinks climate will just stay in one state is nuts…maybe for a human life time span it may seem that way, but unless you get to live through the “flip” its hard to grasp how big the changes can be. For my kid’s future, i hope we blowtoch like no tomorrow and all the ice melts and the temps soar :)

    I was always under the assumption winter snowcover was good for crops, but i could be wrong. Without snowcover, areas of Wisconsin have had frost depths of over 50 inches. With snowcover, my frost depth right now is 8 inches or less and its been COLD, especially at night. I know a lot of snow sucks in spring because it retards warming (albedo) and also is the cause of flooding, but a lot of that is just stupid people building in areas they shouldn’t.

    Ice ages? I thought the assumption was that if winter was WARMER and summer was COLDER then you could/would build an ice sheet. The warmer winter would allow more snowfall and the colder summer would allow less melting, rinse and repeat.

  76. I’d also point out that with an ever growing percentage of thermometers at Airports (where snow is typically cleared away so that the airport can function) there is a bias in the temperature record due to this. While there may not be many take off / landing events during snow season, especially at smaller airports, the tarmac never leaves….

    So “snow extent” tells you more about where snow has actually stuck to the ground (and implicitly how warm that ground is – the melt line of 32 F / 0 C) while the temperatures may just reflect the temperatures of a black asphalt patch in the sun on a clear day… with 2 feet of snow in the woods a few miles away…

    Put more bluntly: If a fine dusting of soot can melt snow, just think what acres of tarmac in the sun can do to a thermometer…

  77. Leif,

    What equation are you using to calculate R^2 in your example? I have a feeling something is wrong, since the R^2 should be very similar in all three of those cases. A good illustration of what R^2 shows can be found here http://mathworld.wolfram.com/CorrelationCoefficient.html As R^2 -> 1, the points being fit lie closer and closer to the fit line.

    It sounds to me like you’re just applying statistical measures with no understanding of their significance or purpose…

  78. JonesII (12:28:07) :
    …not to mention that Birkeland current in the Andromeda Nasa picture.
    There is no Birkeland current visible in that picture. C.P.Snow was lamenting the schism between ‘the two cultures’. Other people have proposed that a ‘third culture’ exists http://www.edge.org/3rd_culture/
    Your post and others of the same ilk show that there is a fourth culture out there of science illiterates. This is a dangerous situation in an age where we depend more and more on science and the scientific approach. This is a new Dark Age arising.

  79. With snow it seems like you have two things at war. Snow does increase albedo tending to perpetuate cold conditions, but on the other hand it also insulates the ground preventing deep penetrating cold. I have to think these kinds of “warring” things tend to make climate an inherently unstable weatherlike kind of thing.

    Being a farmer I really do think we could use a little global warming to guarantee our crops against fluky one or two hundred year frosts. A frost that might come about if you had an early winter hard freeze with no snow cover followed by a deep extensive snow. I can’t, for sure, say that this would translate into a June frost but it certainly would make it more difficult to get the crop planted in a timely fashion. By the way, there is still a good deal (maybe 5%) of corn standing unharvested in Northern Illinois due to last years cool summer, difficult fall harvest, and snowy winter conditions. That much corn unharvested is UNPRECEDENTED – not to coin a phrase.

  80. John A (12:21:26) :

    “I’m sorry Steve Goddard, but I think that the conclusion is largely cherry-picked from the time series and the trend is likely to be spurious. Yes 2010 appears to be a large extension of winter snowfall, but its one season.

    As Leif points out, the low R2 of the points says that most of the apparent trend is scatter and not significant. …”

  81. I think I have an alternate explanation. Some time ago I offered to co-integrate my lifetime data set series with my WOC (woman of choice). The WOC lifetime data set series has now been co-integrated with my own for 30 years. As a direct consequence of co-integration, several children (C) were formed. The C elements as stand alone were highly unstable from the start, but over time began forming unstable relationships with Others (O) of opposite polarity. These were also unstable for the most part with frequent discarding of one O in favour of another. Eventually however some of the CO relationships stabilized, producing an additional O with close bonds to both the C and the O. The unit as a whole is referred to as CO2. At time of writing, no CO3 has yet emerged.

    I graphed Ti (total income) against Te (total expenses) to get Tr (total retained) and determined an inverse correlation with Tr to amount of C supported. The trend tailed off after CO stabilization, but with a HUGE downward inflection at WC (wedding ceremony). I conclude therefore, that Tr is inversely proportional to frequency of C, but that increases in CO2 instigate a major negative short term fluctuation with no long term trends following.

    However, as to the current snowfall and cold weather, I must advise that my WOC recently requested an analysis regarding “dress” and “perception of fat”. My analysis was poorly presented, resulting in outright rejection and a pronounced cooling trend in the immediate vicinity. Until reading this post I had no idea the effects were so wide spread. I have ordered large quantities of S (sparklies) and F (flowers) and having oberved the cyclical nature of this particular type of cooling trend in the past, am predicting a warming trend in the coming weeks for the northern hemisphere at a minimum.

  82. Matt (12:52:57) :
    I have a feeling something is wrong, since the R^2 should be very similar in all three of those cases.
    Certainly not. Try to do as I did. Generate y = sin(x) + a*x
    and use your favorite formula to calculate R^2.

    The one you cite will do fine:
    A good illustration of what R^2 shows can be found here…

    It sounds to me like you’re just applying statistical measures with no understanding of their significance or purpose…
    Revisit this statement after you have completed the above assignment.

  83. An important thing to note is that the US started installing scrubbers in coal plant smokestacks in the early 70′s, and 86 was around the time that China started expanding their energy infrastructure significantly, all coal plants with zero scrubbers…

  84. Has anybody bothered to see what the published literature actually says about snowfall predictions?

    I haven’t had time to get into this topic, but the one paper I have on hand is observational, not modeling. Changnon, Changnon and Karl, 2006.

    They observed maybe an *increase* in snowstorms vs time in the US, over the entire last century, but the data is really really noisy.

    Broken down by region, it’s even noisier.

    This backs up what Leif is insisting on, when looking for the strength of the correlation. Snowfall data is very noisy, and you can fool yourself into seeing a trend where there’s just noise.

    In that context, I’m not sure if anything would surprise me, out of a single winter. Snowfall is truly weather.

  85. Tom_R (09:12:23) :
    Hmm, as I write this I realize that the photon changes direction, but doesn’t lose energy since the wavelength doesn’t change. How is energy conserved when the air molecule gains kinetic energy but the photon energy remains the same?>

    My recollection is that a photon has no mass so can’t transfer any kinetic energy?

  86. Leif Svalgaard (11:39:31) :

    [...]

    R^2 is a good measure of how much of the variation is due to a linear trend, no matter what the data looks like.

    Wouldn’t that R^2 only be meaningful if you had a long enough time series to cover several complete cycles?

    The Rutgers snow extent data only goes back to 1966. If the climate follows a ~60-yr cycle tied to the PDO and/or ENSO, 1966-2009 wouldn’t even cover one full cycle. Wouldn’t a linear or secular trend derived from such a short time series be spurious?

    If I draw a linear trend-line through less than one full cycle of a sin wave with no underlying secular trend, I get an R^2 of 0.88…

    Sin Wave 1906-1913

  87. U know what i am afraid of?

    That they will push full throtle on the weather machine HAARP and give us hellish high temps in the summer of 2010 on some big locations..

    They use the olympics too for convincing us that global warming is real.. If the olympics cant continue then there must be global warming!!

    Or is mother Earth more powerfull? Which i hope..

  88. So the logic is that with a slight larger area covered in the summer and more so in the winter is due to soot for the summer cover, but only for the summer cover and only back in the day when the summer cover wasn’t as wide as today.

    Come again?

  89. There is a total media blackout on the snowfall in America in the Netherlands.

    They mentioned it 2 or 3 times in the news and then went on to the order of the new world..

    War, theft, murder, lies and cheating.. Thats all we ever get to see on the television.. Since i started to see through the lies of commercials and stuff, i seem to want to wake more people up.. Show them that the global warming scare is one big conspiracy and helps only 2 goals.. Estabecialing a NWO government and filling in the gap between truth and fiction! Truth: CO2 is not dangerous and needed, fiction: CO2 is a polution and very bad for our planet..

    But u know what the answer is of most people i speak to?

    I cannot change anyting about it, it doesnt help me if i know, so i just sit here and do nothing.. :-(

    They wanna help the planet, but not by believing in “conspiracies”.. The movies told them so…

  90. Steve Goddard (08:42:01) :

    “We have been told over and over again since Hansen spoke to Congress in 1989, that winter snow is declining”

    Please provide a link to this statement or publication from Hansen. I’m not familiar with it.

  91. Leif: “This is a new Dark Age arising”

    Do you mean the one where most people no longer trust scientists because “Climate Scientists” are no more trustworthy than MSM “Journalists”.

    Who caused that? Which profession kept their mouths shut? Who were too afraid to speak out because they kept getting called deniers?

    Peer Review is now just another code word for either:

    1) Cronies slapping you on the back

    or

    2) Enemies delibererately shutting out anyone threatening their flow of cash.

  92. HankHenry;
    Snow does increase albedo tending to perpetuate cold conditions, but on the other hand it also insulates the ground preventing deep penetrating cold>

    Grew up on the farm Hank and your comment gave me an aha! moment. I remember all those years that there wasn’t much snow and so it melted pretty quick and everyone got hopefull for an early planting season. But then it seemed that spring just couldn’t quite get started and winter would just hang on by its finger nails for the longest time. I’m thinking your post is why, but it would be interesting to see if memory is playing tricks on me and compare snow extent to full seeding to see if the correlation is real.

  93. Robert (09:24:32) : But that corrctive action continues to be drastically reducing burning carbon compounds for energy.

    Ah, the smell of agenda driven conclusions….

    How about we just put soot trap oxidizers on the billions of dollars of power plants and vehicles we use? Fast, fairly easy, and much much cheaper… You know, like we’ve already done for Diesels in the USA and Europe?

    So the more accurate conclusion would be:

    Have the 3rd worlders stop burning down the rain forests, and bring them to modernity so they can use natural gas to cook instead of trees (see the travesty in Madagascar and the denuded landscape in Haiti as examples of doing it wrong. See “Gobar Gas” in India as an example of how to do it right.)

    Get the “newly developing economies” to put trap oxidizers and precipitators on factory smokestacks instead of running wide open.

    Problem solved. Everyone living happier, healthier, and more PRODUCTIVE lives with wealth creation and modernity.

    The alternative path (energy starvation) ends in economic collapse and subsistence living. That ( again, see Haiti and Madagascar – and Sudan and… ) leads to a denuded landscape as the forests are pillaged and burned to survive.

    The USA has MORE forest / woodlands cover now than a generation or two ago, not less. We don’t need to burn trees any more and we produce more food than we can eat (and sell) on the farm land we do have, so more has been allowed to revert to scrub and forest. The direct result of using fossil fuels instead of biomass as done in traditional economies.

    So before you run off to cause a global economic collapse, throw away a few $Trillion of total capital stock, condemn a few Billion people to death and starvation (NOT an overstatement. Shut down the tractors and trucks and see what happens…) and generally force everyone who survives to take an axe to the nearest trees to stay warm and cook: Think about the lowly ‘trap oxidizer’, the electrostatic precipitator, and the natural gas stove… “KISS” principle, please.

  94. Steve Goddard

    You really stirred up a few folks.
    When I get back from Colorado in September, I’ll be sure to let you know if my 20 plus years of observation of the snow-pack Around Monarch Pass fits with your increasing snow in NH in the winter. I really hope you are right.
    What a great site. I’m sure everybody agrees except maybe Tom P.

  95. If I said that it has warmed up since 6 AM, some people would complain that I am cherry picking. “Why didn’t I start my measurements at 3 PM yesterday?”

    Winter snow extent (like everything in the climate) is cyclical. Prior to 1989 it was going down, and since 1989 it has been going up. It makes no sense to apply linear statistical techniques starting at a date on the other leg of the cycle. If snow extent was actually linear with time, we would either have no snow or complete snow cover of the entire planet. Obviously neither of those is the case.

    The trend for the last twenty years is obvious. Don’t get confused by charlatans.

  96. Dr. Tim Ball was interviewed on the Tom Sullivan Show yesterday (16 Feb 2010) in regard to Tim’s views on global warming/climate change.

    Dr. Ball proposed the anology of the red noise of an individual in a large stadium and the roar (white noise) heard outside the stadium with the IPCC’s conclusion that the individual (CO2) red noise is responsible for the white noise (global warming/climate change).
    ————————————————————-

    Sunspots and Cosmic Radiation
    Governments Plan for Warming Based On Corrupt IPCC Science
    By Dr. Tim Ball Thursday, February 18, 2010

    http://canadafreepress.com/index.php/article/20150

    Do Nothing Than The Wrong Thing
    The IPCC was deliberately focused on establishing human causes of climate change so it failed to consider major natural causes. Chief among these were changes in the Sun. It was also deliberately directed to proving CO2 and especially human CO2 was the cause of warming and climate change. This became the basis of all government’s energy and environmental policies, which assume warming is the only possibility. A real scientific investigation would establish natural climate variability and mechanisms first, which is further proof of the political exploitation of the issue. It is better to do nothing than do the wrong thing for the wrong reason.

  97. Someone here suggested that they won’t trust a trend that is less than 200 years.

    If snow extent increased at the current 10 year rate of 200km2/year for 200 years, that would cover all of the currently bare ares Asia, Europe and North America. i.e, an ice age.

    Climate is cyclical – over the last two hundred years, snow cover has gone through many phases where it has increased and many where it has shrunk. People who insist on doing a a linear fit or linear analysis across different legs of a cycle, are just not thinking clearly.

  98. Steve Goddard;
    The trend for the last twenty years is obvious. Don’t get confused by charlatans.>

    I’m a confirmed skeptic Steve, and I don’t dispute your data for a moment. That said, there is about 10 years of data on AMSU-A. Now 20 years is not enough data for a measure of climate, so the last 10 is even worse…. But, that data still shows a slight upward temperature trend over the last 10 years at sea surface, near surface, and lower atmosphere while snow extent is also rising. How do you reconcile these?

  99. @ Fred H. Haynie (12:19:57) :

    ‘Snow is not a perfect reflector. In fact, because of it’s structure, fresh snow acts as a black body. The high energy shortwave radiation from the sun rapidly melts the surface which freezes to form an ice crust that is more of a reflector. But even ice is transparent like glass and the fraction of the energy reflected back to space depends on the angle of incidence. It does radiate low energy longwaves as a function of surface temperature. Much of the outbound longwave radiation from the top of the atmosphere measured by satalites is frozen cloud tops.’

    So there should be two factors working against the equator-ward extension of snow cover: the increasing effective radiation intensity (as the angle of incidence of the incoming radiation becomes progressively normal to the ground surface), and the increasing absorbtion of that radiation by the fallen snow/ice crust (as reflection decreases with increasing the angle of incidence). Though it will be more complex than this of course, due to atmospheric scattering of the incoming radiation.

  100. Steven Goddard:

    I think your article was great. Maybe a needed a tidbit deeper analysis and style according to some. Here’s a statistical curiosity.

    Using your weekly global data, just analyze the period from first week of 1989 forward. The linear trend is basically flat. Assuming y is in sq.km, the upward slope is a positive 39 sq.km per week over that 21 year period. It is then factual that no decrease has occurred since 1988 and all of the decrease in snowfall occurred prior to 1989. During 1967 to 1988 the decrease was a negative 620 sq.km per week over that period. Wow.

    That’s new news to me anyway, that the decrease in snowfall many tout and tie to AGW stopped long, long ago and has been a falsehood for two decades now.

    (If you want good R^2 for some, annualize the data first. You get the same trend line, but many magnitudes of better R^2. However, I won’t go there, R^2 is meaningless in this context unless the data was supposed to be a dead straight line and you want to guage how close the data is to being linear in fact, then it could have meaning.)

  101. “The 28 year trend in snow extent derived from visible and passive microwave satellite data indicates an annual decrease of approximately 1 to 3 percent per decade with greater deceases of approximately 3 to 5 percent during spring and summer. Precipitation in regions of seasonal snow cover appears to be constant or increasing slightly in some locations over the same time period, which suggests that diminishing snow cover is the result of increasing temperatures. One region where the snow appears to be diminishing rapidly is the Western United States, especially in spring when the duration of snow cover has been decreasing by 2-3 days per decade (see blue-colored areas in the Spring Duration figure). This satellite-derived trend agrees with direct measurement of snow depth and extent on the ground.”

    Mote, P.W., A.F. Hamlet, M.P. Clark, and D.P. Lettenmaier. 2005. Declining Mountain Snowpack in Western North America. Bulletin of the American Meteorological Society 86: 39-49.

    Maybe the last four years have changed the trend, but it seems like summer snow continues to decline.

    “The alternative path (energy starvation) ends in economic collapse and subsistence living.”

    Wrong on two counts: reducing CO2 emissions and other methods to reduce AGW are not an either/or proposition, and reducing emissions has never been shown to imply “energy starvation” or any of the other terrible consequences anti-mitigation alarmists posit without offering evidence.

  102. When I go to Rutger’s Global Snow Lab (which is an awesome site, by the way) the graphs it generates for me look very different than the ones Mr Goddard is using. Take a look for yourself:

    Winter: http://climate.rutgers.edu/snowcover/chart_seasonal.php?ui_set=nhland&ui_season=1

    I don’t see a trend. If it’s there it’s subtle.

    Fall: http://climate.rutgers.edu/snowcover/chart_seasonal.php?ui_set=nhland&ui_season=4

    A lot of variability. No trend I can see.

    Spring: http://climate.rutgers.edu/snowcover/chart_seasonal.php?ui_set=nhland&ui_season=2

    Now there’s a trend. Snow extent is way down.

    Summer: http://climate.rutgers.edu/snowcover/chart_seasonal.php?ui_set=nhland&ui_season=3

    Again, dramatically lower. 2008 is a record low; 2009 is the second-lowest.

    The overall trend seems to be the reverse of what Mr Goddard is describing.

  103. davidmhoffer,

    My point is that winter snow cover has been extending towards lower latitudes over the last twenty years, which goes against the endless claims that snow cover is decreasing and storms are moving north.

    I’m not definitely not trying to predict a long term climatological trend into the future.

  104. Steve Goddard (16:09:58) :

    The point is that since we have a longer data set, we have some idea of what the variability in that data set is. Given that knowledge, your trend is not distinct from the variability of the data. If you’re OK with that, then fine.

    By the way, from the IPCC report,

    [SCA = snow covered area]

    “From 1915 to 2004, North American SCA increased in November, December and January owing to increases in precipitation (Section 3.3.2; Groisman et al., 2004). Decreases in snow cover are mainly confined to the latter half of the 20th century, and are most apparent in the spring period over western North America (Groisman et al., 2004). Shifts towards earlier melt by about eight days since the mid-1960s were also observed in northern Alaska (Stone et al., 2002).”

    So decreases in wintertime snow extent are maybe not expected in North America, after all. You might have checked that first.

  105. Leif: “This is a new Dark Age arising”:

    I am astonished you should come to this conclusion based on the supposed (or even real) ignorance of participants at WUWT. Unless people here were 100% science literate, apparently we count as a rising hoard of barbarians. Yet scientists themselves display gross ignorance across disciplines – how’s your microbiology and understanding of PCR and the methods of genetic modification, for example? One does not expect specialists in one scientific feel to be able to address the esoteric and technical aspects of experts in another field. C.P. Snow lived too early – there’s a gulf between scientific disciplines let alone between the science community and non-scientists.

    And yet here we find a blog which is engaging non-scientists in some of the most pressing questions created by scientists – and with good reason. The AGW CO2 theory has needed a good dose of skepticism from everyone, because, as has been well documented since Climategate, and was evident even before this — with examples of scientists losing their positions over failing to toe the AGW line — the ‘science’ has been dangerously tainted by the political advocacy of some of its most influential leaders. They have been allowing their science to be used to justify one of the most egregious attempts at grabbing and redistributing wealth from citizens bamboozled by expertise.

    I’m not a scientist, I’m a historian of science, and I felt concern when the ‘scientific barbarians’ like Mann decided that they could massage a well documented historical phenomenon – the MWP – out of existence by superimposing various scientific proxies on top of the written record. Barbarism extends in many different directions, and one manifestation is to forget or deliberately obliterate the past for political or propagandistic purposes– and there is no denying that science was used in a propagandistic fashion by Mann. No matter how energetically Gavin is currently trying to rewrite the history of AGW propaganda on the MWP (and his own past proclamations), the fact remains that for nearly 10 years an important historical era and the quiet and professional work of many historians was being ignored and even ridiculed for ‘scientific’ purposes. But historians are evidently among the barbarians whose rising you fear, since we do not share all of your professional understanding. Insulted? You bet!

    Incidentally, I am encouraged by seeing here at WUWT and elsewhere the greatest signs of amateur interest in science probably science the end of the era of amateur science at the turn of the last century. And bloggers here are generally justified in being sceptical in tone due to the arrogance and condescension with which their concerns have been treated by Mann, Jones, the IPCC et al in the past. Not a good idea to pour contempt on those you are perhaps trying to impress.

  106. Sorry -several mistakes in last post: I meant to say:

    Incidentally, I am encouraged by seeing here at WUWT and elsewhere the greatest signs of amateur interest in science probably SINCE the end of the era of amateur science at the turn of the last century.

    “science” was on my mind!

    REPLY: FYI, your last comment ended up in spam and was accidentally deleted with a bunch of viagra, I saw it sail by. You’ll have to try again, sorry,m Anthony

  107. Leif Svalgaard (12:57:36) :
    “…This is a new Dark Age arising.”

    Leif – you may be more correct than you realise. The Dark Age at around about 450-700AD corresponds to “Bond Event 1″, the last Bond event, when apparently climate related migrations helped cause the fall of the western Roman empire. Bond estimates 1470 +/- 500 years between these cycles, which means we’re about due. (ref: Science v278, p1257)

    Re the maximum NH snow extent, one possibly interesting correlation is land temperature vs snow extent (can’t be UAH series since data range needs to be over at least one PDO cycle). Everyone has been arguing about R^2, it might be interesting to see how correlated these variables are. If R^2 is high then we have ourselves a nice little feedback of humidity->snow->albedo correcting for increased solar W/m2 etc.

    On the other hand I don’t see how the increasing Antarctic sea ice extent fits with this since sea ice doesn’t equate to snow.

  108. Steve Goddard (16:27:25) :

    “People who insist on doing a a linear fit or linear analysis across different legs of a cycle, are just not thinking clearly.”

    You don’t have to draw the line through the entire data set. But if you pick a subsection, then you have to pass a higher statistical bar to reach significance. Otherwise, you cannot say your trend is different from the noise.

    I could note that it warmed up a lot from 1996 to 1998. It’d definitely be a true statement. But I would not call it a trend.

    By the way, I’m still looking for any signs that Hansen or anybody else was proclaiming that winter snow extent was decreasing over the US. Still haven’t found it. Do you have anything?

  109. Larry Kirk (16:41:29) :

    Could you clarify a bit on that thought of snow’s ability to reflect. Does that have to do with the fact that the thinner a surface is (low optical path length) the less emissivity it has? Read somewhere that emissivity is always stated for a surface of infinite thickness.

    Is that the releated factor of snow crystals you speak of or something entirely different?

  110. Fred H. Haynie (12:19:57) said “Much of the outbound longwave radiation from the top of the atmosphere measured by satalites is frozen cloud tops.”

    I agree with everything else you said, but not this. Muchof the longwave measured by the satellites is radiated from low (unfrozen) cloud tops and clear sky. The higher and colder the cloud tops, the less longwave that they radiate. That’s why the IR satellite picture is in false color (e.g. http://www.ssec.wisc.edu/data/east/latest_eastir.jpg) showing bright white for cold cloud tops and darker colors for low clouds and no clouds. If they used real color, the high clouds of (e.g. thunderstorms and hurricanes) would not show up and the warm ground would, and the picture would be useless.

    This is another reason why global warming is so heavily dependent on clouds, not snow on the ground, or CO2 in the air (esp in cloudy skies). The low clouds from GCR are hypothesized to radiate more IR to space and cause global cooling. Conversely high clouds radiate less and cause global warming.

  111. Please find below an interesting (to me, if no-one else) breakdown of the growth in snow extent for each week of the year, using the full raw data set from the Rutgers snow lab.

    This is hugely amateur effort, hacking around with Excel and probably statistically worthless, but it says to me that the eye is drawn to peaks in graphs, and not slopes.

    I basically pivotted the data to plot the extent as a function of year for each week, giving me 53 series of data. I then used Excel’s linear trendline function to give me a gradient for each series (ie each week), and an R^2. The results are below.

    What is telling is that there is a positive gradient only for weeks 1, 3, 4, 47, 48, 50, 51 and 52. Negative gradients (suggesting a reduction in extent over time) are found for every other week, and significantly steeper than positive gradients. Plot the “Increase per year” / “Week” data below to see what I mean.

    My “analysis” (worth what you paid for it) is that the snow extent is falling for most weeks of the year. I have zero idea what implications that has, but I think it contradicts Steve’s interpretation that extent is increasing.

    Anyway, source data, methods and results freely disclosed for peer review – I’d be interested in your comments.

    Regards,

    Dermot

    Week,Increase per year,R^2
    1,10840,0.00570
    2,-15046,0.00790
    3,5791.1,0.00140
    4,5937.1,0.00140
    5,-6092.6,0.00220
    6,-37411,0.04000
    7,-24899,0.01590
    8,-24303,0.02450
    9,-35498,0.04950
    10,-72719,0.16240
    11,-84099,0.19790
    12,-66987,0.15870
    13,-50607,0.06910
    14,-56710,0.09580
    15,-53113,0.10290
    16,-79549,0.19180
    17,-60042,0.11990
    18,-75881,0.24170
    19,-75959,0.21560
    20,-48842,0.09170
    21,-87189,0.24530
    22,-77784,0.16670
    23,-119080,0.33880
    24,-106579,0.34560
    25,-94858,0.30870
    26,-86636,0.32810
    27,-100677,0.51870
    28,-75222,0.47350
    29,-68542,0.48230
    30,-61234,0.47860
    31,-52238,0.39010
    32,-35916,0.38450
    33,-33119,0.27900
    34,-27393,0.19950
    35,-24235,0.14650
    36,-9935.9,0.03700
    37,-4007.6,0.00240
    38,-1610,0.00010
    39,-18059,0.00830
    40,-35353,0.02590
    41,-48776,0.03750
    42,-51209,0.02940
    43,-28394,0.00950
    44,-24769,0.00900
    45,-16896,0.00530
    46,-16282,0.00600
    47,25341,0.01990
    48,951.56,0.00002
    49,-2583.7,0.00020
    50,24562,0.01700
    51,39445,0.04160
    52,31345,0.03380

    53? hmmm…

  112. From the Rutgers website . . .

    “Annual snow cover extent (SCE) over Northern Hemisphere lands averaged 24.4 million square kilometers in 2008. This is 1.1 million sq. km less than the 39-year average and ranks 2008 as having the 4th least extensive cover on record…”

  113. “over the last two hundred years, snow cover has gone through many phases where it has increased and many where it has shrunk. People who insist on doing a a linear fit or linear analysis across different legs of a cycle, are just not thinking clearly.”

    Does the analysis of this data give us insight into any possible causes for these cycles? That seems to be the truly important thing here- I’m sorry if someone has already offered a possible explanation, I must have missed it.

    So yeah, why the big dip between 1967 to 1988, as Wayne describes above for instance? Or from another perspective, by looking at the whole data, I can kind of make out a cycle of about a decade in length, a rise for 5 years, then a dip for 5 years. I’ve no idea if this is significant of anything or not. Given the limited data, we obviously can’t say much about any larger cycles that may be occuring.

    As you say, we need to get a handle on these cycles. That would presumably come first- before we start beating each other over the head over whether what we are seeing is significant in terms of (A)GW, or not.

  114. Lief I really do not see what your point is. And I do not think you understand the meaning of R^2 for the trendline in this case. Basically it would mean how well does your trendline fit the data and you can see visually the trendline is flat. Thats why you graph because the eye can tell you what is probably correct.

    I let you off when you made a blunder re black bodies, also about CO2 freezing out of the atmosphere. But you are getting insufferable. If you do not have a point keep quiet, just let it go.

  115. davidmhoffer (13:24:07) :
    My recollection is that a photon has no mass so can’t transfer any kinetic energy?
    A photon has momentum so can give you a kick = increase your kinetic energy.

    David Middleton (13:28:38) :
    If I draw a linear trend-line through less than one full cycle of a sin wave with no underlying secular trend, I get an R^2 of 0.88…
    Indeed, and you also get a spurious linear trend. However, if people start out by claiming there is a trend, then they have forfeited the wave.

    Bruce (15:51:16) :
    Leif: “This is a new Dark Age arising”
    Do you mean the one where most people no longer trust scientists

    No, I do not mean that. I mean that people believe in pseudo-science, which can well be peddled by some scientists [iron suns, plasma universes, intelligent design, astrology, young earth, etc]

    vigilantfish (17:19:22) :
    Not a good idea to pour contempt on those you are perhaps trying to impress.
    This is not about ‘impressing’ anybody. The Dark Age crowd are not trying to learn anything or skeptical, they are pushing [hard] an anti-scientific view of the world and are learning-resistant.

  116. While we’ve seen some record snow this winter, we’ve also seen record tropospheric temperatures. Februrary continues on the same track following in the footsteps of a record January. See:

    http://discover.itsc.uah.edu/amsutemps/

    For the latest. Note especially sea surface (which is NOT ocean temps) up to 25,000 feet or so. Some of us have pointed out that warm temps and more snow go hand in hand in the winter, just as the very cold Antarctic region is also one of the driest regions in terms of precip. I maintain that yes…this has been a snowy winter, and the warmer tropospheric and ocean temps are to thank.

  117. I’m arguing with Tamino about these same issues at his blog (and he’s being a little rude to be perfectly honest).

    It is not cherry picking to say that since 1989, snow extent has been increasing. From the graphs at Tamino’s blog, it appears snow extent decreased from 1970 to the mid-90′s and then began increasing until the present. Fitting a single line to all the data does not help anyone understand the variation of snow extent. In fact, centered around the mid-90′s, 15 of 16 snow extents were below the forty year trendline. A second-order polynomial fit would likely result in a higher R^2 value.

  118. Leif Svalgaard (18:29:51) :
    davidmhoffer (13:24:07) :
    My recollection is that a photon has no mass so can’t transfer any kinetic energy?
    A photon has momentum so can give you a kick = increase your kinetic energy>

    no Leif… you have photons confused withy photon TORPEDOES. They haven’t been invented yet.

    Steve Goddard (17:12:30) :
    I’m not definitely not trying to predict a long term climatological trend into the future>

    Didn’t mean to imply that you were. Was just wondering what you (or anyone else) thought as to cause/effect re warmer temps and snow extent increasing and sliding toward equator.

  119. Fellow thinkers,

    We must be careful not to fall into the trap of using inadequate statistics derived from incomplete data, ignoring other inadequate statistics derived from incomplete data that may (or may not) have a bearing on the subject, because of preconceived ideas.

    I quote an apocryphal story, told to me by my father, told to him by his father who got it from his brother who was in the B.E.F. 1914/15.

    When the British army first started issuing tin hats (all right chromium/molybdenum steel helmets) in 1915 there was resistance by the higher authorities on the grounds that “Real soldiers have never worn helmets, in their experience”. This ignored the historical evidence and even that some cavalry units (from which many of the senior officers were drawn) wore helmets. Shortly afterwards hospital returns showed a large increase in head wounds being treated. This was cited as evidence that helmets were not only not needed but were dangerous. It was proposed that steel helmets be withdrawn! Further investigation showed that the number of deaths from head wounds had decreased more than the increase in number of treated head wounds, and of those treated the head wounds were in general less severe than before the introduction of the helmets.

    Fortunately for many concerned the steel helmets continued to be issued and saved many lives.

    The obtaining off complete evidence, and not partial evidence in isolation, negated possible catastrophic effects. Now where have I heard something similar? Perhaps the proposal to destroy the worlds economy for a religious and preconceived idea.

    p.s. It was indeed fortunate for me, as my grandfather was invalided home in 1917 with a head wound, finally released from hospital in 1919, although much of the time was due to re-constructive and plastic surgery.

  120. Hi Steve,
    Perhaps I missed something, but you never did get around to explaining why you concentrate on winter snow? I’m at a loss. Is increased precipitation telling us something exciting about climate?

    Thanks for the Northern Hemisphere annual snow plot. Wow! That’s quite the upward trend you’ve got rocketing away there. Woosh. Although to be fair, the sideways trend is rather more evident. Prey tell though: what is the margin of error on that thing?

    And remind me, why did you pick 1989 is a start point? What was wrong with all the other years? What have you got against 1992 for example? Or 1985?

  121. Richard (18:19:39) :
    I let you off when you made a blunder re black bodies, also about CO2 freezing out of the atmosphere.
    It is nice that you let me off all my blunders. The one about black bodies I don’t recall. Refresh my memory [use email if needed].

    But you are getting insufferable. If you do not have a point keep quiet, just let it go.
    Well, I think I do have a point, namely that one does not start out a topic by claiming there is a clear trend. when there isn’t. R^2 is a perfectly valid method to gauge this as I have tried to demonstrate. Now, I’m willing to ‘let it go’ as it is clear to everybody [without my insistence] that there is no significant trend in the first Figure and that by judicious cherry picking one can claim a [barely] significant increase.

  122. [snip - nice try "thefordprefect" - you are no longer welcome here when you do shape shifting, pick a name preferably your real one, and stick with it, otherwise bugger off]

  123. Steve – Zender did NOT claim that 80% of the soot was due to burning of fossil and biofuels. Read it more carefully

    “Applying biomass burning BC emission inventories for a strong (1998) and weak (2001) boreal fire year, we estimate global annual mean BC/snow surface radiative forcing from all sources (fossil fuel, biofuel, and biomass burning) of +0.054 (0.007–0.13) and +0.049 (0.007–0.12) W m−2, respectively. Snow forcing from only fossil fuel + biofuel sources is +0.043 W m−2 (forcing from only fossil fuels is +0.033 W m−2), suggesting that the anthropogenic contribution to total forcing is at least 80%.”

    If I read this right this is saying that if you could somehow eliminate the soot due to biomass burning (forest fires and the like), then the remaining carbon generated by fuel burning would cause 80% of the warming. But that isn’t the same as saying that 80% of the soot is due to these sources as the warming effect is non-linear. In other words snow only has to be slightly dirty to cause significant warning and making it twice as dirty probably wont double the effect.

    The non-linearity makes the way in which the results are stated deceptive. I suspect that if you analysed the warming contribution of soot from biomass burning in the same way you would find that it too could be said to be `responsible for 80% of the warming’ – maybe more! If so then that would mean cutting all the soot contribution from fuels would only diminish the warming by 20%.

    He has presented his results in a rather strange way which I don’t think is very informative. What you would want to know is the marginal impact of mans contribution to the underlying situation, and that isn’t the number he’s chosen to give us.

  124. Dude (18:46:18) :
    It is not cherry picking to say that since 1989, snow extent has been increasing.
    It is clear that for a climate trend, one or even two years more or less shouldn’t matter. So what is missing is a sensitivity analysis. What difference in R^2 [which is a convenient and valid measure] does it make to start in 1988, 1989, or 1990 and to end in 2009, 2008, or 2006. For a robust signal these small changes shouldn’t matter much. Yet no such analysis was made. The resistance one feels about this may be telling. Of course, I could make that analysis myself, but it really behooves Steve to make it.

  125. Steve Goddard (18:46:03) :

    Ah, you found something.

    So they say the models do a lousy job with snow cover, they don’t show any (model) trend over the 20th C, but a decreasing trend over the 21st C.

    You should also discuss this one (Brown, 2009), which looks to be a newer and more detailed study.

    http://ams.allenpress.com/archive/1520-0442/22/8/pdf/i1520-0442-22-8-2124.pdf

    As well as the observations, as discussed in Changnon et al, and whatever all the IPCC AR4 had to say about the topic (both observations and projections).

    You wouldn’t want to be accused of cherry-picking by not giving the whole context of the literature, after all.

  126. Dude (18:46:18) :

    “A second-order polynomial fit would likely result in a higher R^2 value.”

    The game is not to increase your R^2 by increasing the order of the polynomial. That’s generally a horrible idea.

  127. Steve Goddard (18:46:03) :

    Why are all these models unable to show any increase in snow, decrease in temps, increase in rainfall, etc?
    I am wanting to call the computer banks they use Deep Gloom.

  128. davidmhoffer (19:25:41) :
    “A photon has momentum so can give you a kick = increase your kinetic energy”
    no Leif… you have photons confused withy photon TORPEDOES. They haven’t been invented yet.

    The energy and momentum of a photon are related by E = pc, where p is the magnitude of the momentum vector p. The momentum p points in the direction of the photon’s propagation and has a magnitude of p = h/L, where h is Planck’s constant and L is the wave length. So, photons do have momentum and can give you a kick and increase your kinetic energy.

  129. Richard (18:19:39) :
    And I do not think you understand the meaning of R^2 for the trendline in this case. Basically it would mean how well does your trendline fit the data

    No, that is not what R^2 tells you. It tells you how much of the variation of the data can be explained by a [linear] trend. So, if R^2 is 0.03, it means that 97% of the variation of the data is not due to the trend. By smoothing the data, you increase R^2, but you also decrease the significance as you have fewer independent data points. In the extreme you can form the average of the first half of the data and of the last half. R^2 for those two average points is 1, but the significance is 0.

  130. Carot Eater and Stephen Goddard,

    Please note that this June 2009 report indicates that the North East will experience Less Snow due to global warming.
    It seems to me that we can find any position we want in these junk government reports except Science:

    “June 2009
    Members of Congress:
    On behalf of the National Science and Technology Council, the U.S. Global Change Research Program is pleased to
    transmit to the President and the Congress this state of knowledge report: “Global Climate Change Impacts in the United
    States.” This report summarizes the science of climate change and the impacts of climate change on the United States,
    now and in the future.”
    Global Climate Change IMpacts page 107

    “Since 1970, the annual average temperature in the Northeast has
    increased by 2°F, with winter temperatures rising twice this much.150
    Warming has resulted in many other climate-related changes,
    including:
    More frequent days with t • emperatures above 90°F
    • A longer growing season
    • Increased heavy precipitation
    • Less winter precipitation falling as snow and more as rain
    • Reduced snowpack
    • Earlier breakup of winter ice on lakes and rivers
    • Earlier spring snowmelt resulting in earlier peak river flows
    • Rising sea surface temperatures and sea level”

    Also this seems to contradict recent claims by the NYT and others who claimed that scientists predicted the increased snow in the North East we recently experienced.
    Help!!!

  131. I fitted the equation to the data provided.
    snow=25,320 + 22,303*SIN[-5.29*2*PHI/(week of year)}
    then looked at residuals.
    If anyone can see a pattern in these residuals, then he or she has been smoking too much pot!

  132. It is quite clear that climate is neither linear nor a second order polynomial, given that the oceans are neither evaporated nor frozen after billions of years – which would be the necessary consequence.

    People can argue endlessly about lies, damn lies and statistics, but sometimes they just need to trust their own eyes.

    Starting in 2001, 8 out of 10 years are above 45 million km2. Prior to that, 8 out of 11 years were below 45 million km2. The current year is the second highest on record. Enough with the statistics games, please.

  133. I think the point of the topic was the lack of the CLAIMED negative trend no? I see no negative trend. I think I got the point. I think we could use more common sense in our scientists, maybe there should be a new collegiate prerequisites for BAs or at least PHDs as a minimum. CommonSense Logic101, shortly to be followed by contortionist logic1A, and for the undergraduate requirement…the forest through the trees1B!.

  134. Steve Goddard (20:47:37) :

    “just need to trust their own eyes.”

    For them to do that, you need to show the entire data set. You can start your trend at 1989 if you really want to, but show the entire data set so people can judge the thing in context – are these trends, or just noise?

    Just like with temperature: you could draw a trendline in global temperature up from 1996 to 1998, and then another one down from 1998 to 2000. They’d be pretty impressively steep lines, but would it mean anything?

  135. Ric Werme (12:12:08) :

    Steve Goddard (10:20:59) :

    > R^2 on the winter graph since 1989 is 0.298514013

    9 significant figures? Careful – Leif will start teaching you about significant digits and error bars.

    NO THAT IS MY JOB!

  136. carrot,

    We have been bombarded with disappearing winter snow stories for years, and if you have been paying attention you should know that.

    A CU professor predicted the demise of Colorado skiing in the middle of a record snow year two years ago. The Met Office predicted the demise of snow in England 10 years ago. RFK predicted the demise of snow in Washington DC two years ago. I just pointed you to an article which showed all IPCC GCMs predicting the demise of winter snow.

  137. carrot,

    The upwards trend started in 1989. Prior to that there was a downwards trend. If I am showing a graph of the most recent upwards trend, why would I include years that aren’t part of that trend?

    If I said “it warmed up 15 degrees between 6AM and 3PM” is that something you would dispute because I didn’t mention the downwards trend prior to 6AM?

    Obviously every upwards trend is preceded by a downwards trend. That is a fundamental principle of nature.

  138. Typo. Apologies to all. Should be:
    snow=25320 + 22303*SIN[-5.29 + 2*PHI*(week of year)/52]
    This is just a curve that symmetrically mimics the earth about the sun. You can get a better fit with extra frequencies, but that just makes it hard to understand and nothing meaningful changes.
    I don’t know how to paste the chart of residuals into these comments, but it is easy enough to generate in excel. If you look at the entire period from 1966 there is just no visual pattern of structual change. Sometimes the residuals are bigger sometimes smaller. On the winter side they are smallest around 1989, and largest around 1977-78 and the last couple of years.

    Some years there is more so, some less. The year to year variation is entirely within statistical (and visual) norms.

  139. Don Shaw (20:27:13) :

    Good find. Keep digging, people. Context is important here: was everybody saying the same thing? Or are the predictions a contradictory and confused mixture? One complication:

    “• Increased heavy precipitation
    • Less winter precipitation falling as snow and more as rain”

    If you combine these two, you could still on net get more snow…

    I honestly had no idea what the predictions/statements were on this matter, but I can easily imagine its something the climate models would have trouble with. It’s known that regional precipitation projections need some improvement.

    “Also this seems to contradict recent claims by the NYT and others who claimed that scientists predicted the increased snow in the North East we recently experienced.”

    First, it’s simply incorrect for anybody to say they used understanding of climate to predict one freak winter. Weather is weather, climate is climate. If they said, “x is consistent with expectations”, that’s fine. Did the articles you read cite any papers? That could add to the pile. I think the Changnon paper shows some sort of increased snowfall trend for the US northeast, though I don’t know if it was significant.

    That’s the other problem here – One freak snow season doesn’t prove or disprove any climate projection. And given the high variability in snow data, Goddard’s trend will continue to be questioned.

  140. Well, because I love numbers, I took a look at the snow area figures. I subtracted the weekly average from weekly data to give anomalies. I expressed them as a percentage of NH average snow cover. Here is that graph:

    Snow area

    A few comments, in no particular order:

    1. As someone mentioned above, the PDO is important in this graph. The PDO went from cool to warm in about 1976, as can be seen in the snow area anomaly.

    2. From about 1976 to about 1990, snow area was decreasing slightly.

    3. Since 1990, it has increased slightly.

    4. Changes over the entire period are about ±5%. By eyeball, none of them look statistically significant.

    5. We’re about where we were in 1982.

    6. There has been basically no change since 1995.

    Now, perhaps there is a signal of a dangerously warming earth in there, but I don’t see it. And perhaps there is a signal of increasing soot melting the snow in there, but I don’t see it.

    The change in albedo due to decreased snow coverage is supposed to be a strong positive feedback in the global thermal balance … I don’t see that either.

    My conclusion? Either:

    a) The land temperature records are badly corrupted, by some unknown combination of UHI, bad station siting, landuse/landcover changes, and nefarious meddling, or

    b) Slight changes in temperature don’t change the snow area much. Increasing warmth theoretically increases both snow melt and precipitation, so they may basically offset each other, or

    c) Both.

    w.

  141. Yeah Willis! Cutting to the chase as always. Now we have some clarity on the question of the shrinking or swelling winter snow extent, no so-called ‘cherry picking’ of dates involved.

  142. “People can argue endlessly about lies, damn lies and statistics, but sometimes they just need to trust their own eyes.

    Starting in 2001, 8 out of 10 years are above 45 million km2. Prior to that, 8 out of 11 years were below 45 million km2. The current year is the second highest on record. Enough with the statistics games, please.”

    Steve, the site that is your source for this information directly contradicts some of the things you are taking from their data. There is a strong negative trend in both summer and spring snow extent. In 2008, they report, the snow extent was: “1.1 million sq. km less than the 39-year average and ranks 2008 as having the 4th least extensive cover on record…””

    It seems as though you are having trouble taking your own advice, and looking at what it right in front of your eyes.

  143. One thing that no one has mentioned would be to compute the area under the curves from minimum to minimum for each year. The time x area should give an interesting look at how much the albedo changes. Any trend in peak levels or minimum levels strike me as being less valuable than the product of the time and the total snow surface available to reflect light.

    A cycle with a tall but relatively sharp peak might have less actual effective change in albedo than a shorter but fatter cycle, as the latter would have snow present for a longer period of time.

    If you really wanted to get fancy you could compute the days of sunlight for each day of the cycle and also compensate for the changing length of days. A given square area of snow during late December would reflect significantly less energy than the same area during March or April due to the longer day.

    Larry

  144. Robert,

    Do snow extent maximums occur in the summer, or in the winter?

    The graph is clearly marked as “Winter Snow Extent.”

    Did you bother to read the article before posting?

  145. davidmhoffer (13:24:07) :

    “Tom_R (09:12:23) :
    Hmm, as I write this I realize that the photon changes direction, but doesn’t lose energy since the wavelength doesn’t change. How is energy conserved when the air molecule gains kinetic energy but the photon energy remains the same?>

    My recollection is that a photon has no mass so can’t transfer any kinetic energy?”
    “Leif Svalgaard (18:29:51)A photon has momentum so can give you a kick = increase your kinetic energy.”

    My understanding is that the CO2 molecule absorbs the photon, thus increasing the molecule’s internal (vibrational?) energy and then re-emits a photon of the same energy and thus wavelength but in a random direction. Otherwise we have a perpetual motion machine. Have I got this wrong?

  146. carrot eater (13:23:10) :

    It’s pretty clear the problems with relying only on published works.

    Raw data is better.

    But you want to distract from that. And true to that form, you used the word maybe. Maybe, could, if, the infamous words of a troll.

    “you can fool yourself”

    I think yes, you could fool yourself more easily than most other people carrot.

    And I see you attacked Steven Goddard. You people never quit.

  147. carrot eater (15:36:52) :

    Steve Goddard (08:42:01) :

    “We have been told over and over again since Hansen spoke to Congress in 1989, that winter snow is declining”

    Please provide a link to this statement or publication from Hansen. I’m not familiar with it.

    Some schools offer courses in reading comprehension.

  148. Willis,

    Thanks for the analysis. Soot would only affect late spring/summer extent, because it is covered by fresh snow the rest of the year. It requires an extended period of time with no snowfall to accumulate. And forest fires occur in the summer.

  149. carrot eater (17:24:07) :

    By the way, I’m still looking for any signs that Hansen or anybody else was proclaiming that winter snow extent was decreasing over the US. Still haven’t found it

    Keep looking for that Hansen quote, keep looking. Knock yourself out carrot.

    Maybe you could link your search to global warming somehow and get some funding for it: you know, join that party! Or are you already in it?

  150. Steve Goddard (21:46:26) :

    Not really. Still just looks like noise to me. I like Willis’s smooth better.

    Steve Goddard (21:43:56) :

    Summer is interesting in that it’s more simple. There’s a more obvious trend, and it’s more obvious why there should be a trend, with warming. It isn’t obvious to me which way the winter trend should even go with warming. It’d be complicated, with air circulation patterns affecting what you get. At least, that’s how I see it.

    Seriously, I hope you gather several sources for a write-up on what is known/predicted about snowfall/snowcover. A couple modeling papers, a couple observation papers, the report for the government above, the relevant sections of the IPCC – I don’t think anybody’s written something comprehensive along these lines recently, so it’d be a helpful contribution if you got a nice overview.

  151. carrot eater,

    There is nothing simple about summer snow melt. People who have studied the problem (like Hansen) thank that soot is a primary factor. Zander says up to 90%.

    I flew over Greenland during late summer 2007 (the big melt year) and the snow/ice along the western side was absolutely filthy. perhaps from forest fires in Canada.

  152. Wait a minute…. Is there even warming in the winter months, in this region? Monthly trends can be weird.

    Looking at GISS trend maps, for Goddard’s period of 1989-2010, January is getting colder in the eastern half of the US, but warmer in the Arctic to the North.

    The Dec-Feb winter trend, 1989-2009, is similar, but not as pronounced. The Atlantic ocean offshore is about trendless, with some spots either way.

    I have no idea what that means for winter snow, but it adds a complication.

  153. carrot eater (22:32:56) :

    A couple modeling papers

    ============================================

    Why?

  154. Stu (17:56:03) : Steven: Leif:

    Careful, if the science is correct, the GW factor need not come in to play. Proper science will handle the GW factor itself. Do the proper science anyway; what you learn may apply to something a month from now that would directly apply to GW factor.

    Time series can be problematic when re-grouped in time periods. Here’s an example. Just as Steven’s top chart’s data, I have another set of snowfall data. I graph it. It jumps up and down, down and up in big jumps all over the place. I get the R^2, its terrible, 0.03. Now if I allow myself to change the division between the years in fifteen different places no more than one week, I now have a perfectly straight line for a graph with an R^2 of one!

    The only thing that happened is I allowed the big snowstorm near year end to fall in this year instead of that year. The big storm on Dec 31 was moved to Jan 1 of the following year, not actually; actually the division was moved one week. Another year I moved the storm from Jan 2, to Dec 30. That tiny 7/365 (2%) shift in the time dimension took a graph from no correlation to perfect correlation. Does that make sense? By a 2% flex in time division placement, data is either laughed as meaningless or accepted as if written in granite.

    Even Feynman would shout, are you crazy, use you logic, reason, common sense for heavens sake. Statistics never takes the front seat. Statistics is a tool for ease and speed and to simplify complex questions but never overrides correct reason. PC’s came out, and in some respects, logic went out the window!

    Leif, maybe you are enough of a specialist in statistics to help me here. I hit this type of problem years ago but still have no answer of the proper way to handle this. This sometimes applies to temperature series of course.

    The bottom line, this type of time series can be very sensitive in the time dimension. See my point? How do you handle that proper in science and statistics? (My stat book seems to say nothing on it)

  155. carrot eater (22:32:56) :

    the relevant sections of the IPCC

    ================================================

    Have you heard about the problems with the IPCC report?

    Surely you must have.

  156. aMINO aCIDS iN mETEORITES (22:41:12) :

    It appears that Goddard is writing an article to the effect of “what was/is the conventional wisdom about winter snow trends, past/present/future”. To that end, one would have to see what the different models say, whether the models agree with each other and with past observation, and also whether the authors or IPCC authors express any confidence in those model projections.

  157. Steve Goddard (22:38:19) :

    Whether it’s more soot or more warmth, one would expect less snow either way. At least until soot decreases. To me, that makes it easier to figure out than winter, where I can’t reason which way the trend would be. That there isn’t uniform warming spatially in the winter just adds to that mess.

  158. why would it matter what the IPCC thinks?

    Have you been reading about the IPCC over the past few months?

  159. carrot, why don’t you try to write you own paper. Your line of what is expected is your own. You don’t seem to be following what Steve Goddard is saying. I don’t know if it is intentional, unintentional, or what.

    There may be other places for you to find people who you don’t have to try to steer toward what you want them to be like.

  160. “Maybe, could, if, the infamous words of a troll.”

    I don’t think you understand what a “troll” is. Somebody who despises words like “maybe” and “if” is a lot closer to trolldom then one who has mastered the use of those words.

  161. ” Steve Goddard (21:43:56) :

    Robert,

    Do snow extent maximums occur in the summer, or in the winter?

    The graph is clearly marked as “Winter Snow Extent.”

    Did you bother to read the article before posting?”

    Steve, did you forget that you claimed: “Summer snow cover declined significantly (from the 1970s ice age scare) during the 1980s, but minimums have not changed much since then. As you can see in the graph below, the overall annual trend since 1989 has been slightly upwards.”

    Maybe you should read the article you wrote.

  162. ” aMINO aCIDS iN mETEORITES (23:13:29) :

    carrot, why don’t you try to write you own paper.”

    Skeptics who question assertions are just so frustrating, aren’t they?

  163. aMINO aCIDS iN mETEORITES (23:13:29) :

    Of course, Goddard will write whatever he wants. I’m only suggesting a wide scope, as I think it’d be filling a need. He can ignore my opinion. But you might note that he said he was starting with a modeling paper, so I don’t think we differ that much on the idea of whether to look at a modeling paper….

    “why would it matter what the IPCC thinks?”

    When a lot of people have been saying “it’s snowing a lot this winter, so therefore the IPCC and all the warmists must be wrong”, you don’t think it worthwhile to actually see what the report or the literature says on the matter? Myself, I don’t think a single winter is meaningful either which way; weather is weather, and one year is not a trend, but Goddard sees a trend since 1989. So let’s see what comes of it.

  164. Steve,
    Thanks for the response. I read the article, but it only consists of a few sentences and a couple of plots. Perhaps it’s because I’m new here, but I’m not sure what you’re trying to tell readers.

    Simple (pertinent, I think) questions.
    What do you think snow cover tells us about climate?
    What’s the margin of error on the Northern Hemisphere Snow Cover Plot? How confident are you that you’ve got an upwards trend?
    Why do you think starting a plot from a minimum in the data is a good way to tell us something useful about trends?

  165. Carrot Eater said:

    “A second-order polynomial fit would likely result in a higher R^2 value.”

    The game is not to increase your R^2 by increasing the order of the polynomial. That’s generally a horrible idea.

    Here’s what I’m trying to say…you fit linear regessions to data you think is linear. Looking at winter snow extent, it’s pretty clear to me, it’s not linear…it’s got a cuve to it. Therefore, fitting a single line to all the data doesn’t make a lot of sense.

  166. Does anyone think there might be a problem with choosing your starting point because that’s where the trend started?

  167. Steve

    Your last sentence:

    “Earlier changes in summer snow extent were likely due primarily to soot – not CO2.”

    Is this Hansen’s statement or yours? It goes unquestioned. We dont need to be hypnotised by AGW bluster from Hansen or anyone else – ascribing snow trends or anything else in as complex a system as climate, to a single factor, e.g. soot, is certain to be false. Soot in particular is an irrelevant red herring, AGWers are grasping at straws.

  168. This conversation seems to have wandered all over the place, pretty aimlessly.

    I’m sitting here looking at a map of air mass movement from a 1983 textbook on maritime weather, and I see an intersection point over about Asheville NC, of the maritime polar(mP) air mass, the continental polar (cP) air mass, and the maritime tropical (mT) airmass, with a potential contribution from an arctic (A) air mass from the north. The intersection point is right where the Washington weather bomb got up a head of steam and charged forward, and the relevant air masses have all the necessary ingredients. The historic path of both the mT and cP air masses is right out over Delaware (don’t have a chart for eurasia, but would be surprised if there aren’t similar confluences).

    Now, it seems likely that the confluential interaction of these air masses was just so this season, producing the weather experienced this winter.

    Since there are previous records for similar events (of equal magnitude – I’ll re-iterate what I said earlier – a 1/2″ difference does not a “new” record make), it is reasonable, and probably even likely, that the snow extent is chaotically cyclic (chaotic, only because we haven’t yet ascertained what the respective periodicities are), and we are seeing a normal event that occurs whenever certain planetary drivers coincide (ie, el ninos, pdos, enso, solar etc.). Forget soot, CO2 and the IPCC and look to the macro events before you blow your minds on the micro political drivers.

  169. Another masterful analysis by Steve “Pixel Counter” Goddard. Using Excel and 21 years worth of data to contradict 100 years of science – and in 500 words or less….impressive!

    The “skeptic” community has made many valid observations and critiques of climate science, but allowing folks like Steve Goddard to speak for the community at large illegitimizes potentially useful insights contributed by others.

    My advice to everyone here: do not visit this website for science. Peer-reviewed literature is where this debate should be happening, not the blogosphere.

  170. >> Leif Svalgaard (20:02:12) :
    The energy and momentum of a photon are related by E = pc, where p is the magnitude of the momentum vector p. The momentum p points in the direction of the photon’s propagation and has a magnitude of p = h/L, where h is Planck’s constant and L is the wave length. So, photons do have momentum and can give you a kick and increase your kinetic energy. <<

    This is as I remember it from my basic physics. What I don't remember is what happens to the photon. If an air molecule gains kinetic energy from a collision, the photon must lose energy in the collision. Does the photon come off of the collision with a longer wavelength?

    Starting with a molecule at rest, and a photon scattered directly backwards, the transfer of momentum (dp) would be:

    dp = 2 * E/c [E = photon energy, c = speed of light]

    the final velocity of the molecule would be:

    v = 2 * E / mc [m = mass of air molecule]

    and the kinetic energy (dE) would be:

    dE = 2 * E * E / (mc2) [c2 = speed of light squared]

    A quick run through the physics tables gives a ballpark value for E/mc2 of about 10 to the -10, so the change in photon energy would be barely noticible.

  171. “Why Is Winter Snow Extent Interesting?”
    Guest post by Steven Goddard

    Well it seems we have several answers to the question and many questions about the answers and many questions that never got answers. I never knew that there was so much interest in winter snow extent graphs.

    I’m just a little confused though. It sure seemed that more are more interested in the math of the graphs and not the extent of the snow.

    Perhaps those who know about snow and ice and R2 (whatever it is) will put together another article someday that addresses the connection between annual snow and ice extents so the minutia mob stay in the closet.

    Somehow I doubt that Plato put up with “R2″ minutia and bickering. Wonder how he kept the conversation from such diversions and actually came up with a meaningful discussion and reasonable answer(s) to the question(s).

  172. Phil M. (05:44:51) :
    [..]
    My advice to everyone here: do not visit this website for science. Peer-reviewed literature is where this debate should be happening, not the blogosphere.

    Apparently you missed the “memo(s)”. There is no debate in “peer-reviewed” literature…

  173. Phil M,

    If you have a specific objection to anything in the article, please state it.

    Otherwise, don’t waste other people’s time with your pointless diatribe.

  174. Ronaldo (21:52:24) :
    My understanding is that the CO2 molecule absorbs the photon, thus increasing the molecule’s internal (vibrational?) energy and then re-emits a photon of the same energy and thus wavelength but in a random direction. Otherwise we have a perpetual motion machine. Have I got this wrong?>

    I bow to Leif and others on this issue who have posted various explanations. I can only advise that the explanations may not be complete as they do not support a practical method to build a photon torpedo. I’ve been told that photon torpedos are science fiction, to which I respond that there is just as much evidence for them as there is for perpetual motion. And AGW.

  175. BTW Phil M,

    NSIDC teaches pixel counting to their students at CU as a technique to measure Arctic ice. If you find that offensive to your sensitive disposition, please contact them with your complaint.

  176. Quote : Phil M. (05:44:51) :

    “My advice to everyone here: do not visit this website for science. Peer-reviewed literature is where this debate should be happening, not the blogosphere.”

    Like the IPCC?

  177. phlogiston ,

    The soot remark is my statement, not Hansen’s. I’m just back from driving on snow being melted by intentionally placed dark material. It is a very effective way to melt snow – orders of magnitude faster than it would have otherwise.

  178. A late response to Leif – posted last night but it went missing due to my own error:

    Leif:

    I went back over previous e-mails and realized belatedly that you were responding to a strain of ’strange’ science in some of the comments that I had not followed closely enough.

    However, if you care to look at various episodes in the history of science, such as the belief in phrenology, or the electricity of the body, or the inventor Tesla’s belief in the future of transmitting electrical power through the air or his determination to invent a perpetual motion machine, you will realize that this is nothing new. William Harvey, so famous for determining experimentally the circulation of the blood in the 17th century, also determined experimentally that females of the human and ovine species could become pregnant by gazing lustfully at the males of the species (or in the case of women, portraits of handsome men): no contact required. Harvey in fact was ignoring millenia of scientific wisdom. Aristotle and the Greeks realized that a “mingling of the fluids” was needed for conception to occur.

    There is no dark age of science or civilization arising from the eccentrics amongst us: a dark age may arise, however, from the willful distortion of the scientific process for monetary and political ends. This happened under the Nazi regime. Unfortunately the lessons of history are being forgotten, as evidently the IPCC and other bodies believe the ends justifies the means – and the ends cannot be achieved without packaging propaganda as science.

  179. vigilantfish (07:17:49) :
    a dark age may arise, however, from the willful distortion of the scientific process for monetary and political ends.
    Such distortion would not be successful if the citizenry were scientific literate. Now, some of the illiterati are willfully illiterate, i.e. not willing to learn for various reasons, e.g. religious.
    There is another underlying problem, namely that modern science may be intellectually overwhelming and some people are drawn to simpler pictures that they can ‘understand’ [it is easier to 'understand' that angels push the planets around than some mathematical equations; or that 'thunderbolts' from the heavens are powering the Sun rather than complicated nuclear processes]. As has been said so often, half of Americans believe the Earth is 6000 years old. Right there, one loses that segment of the population.
    Regarding the IPCC and ‘other bodies’ there are people that oppose these on economic grounds rather on scientific grounds. I guess than there actually are more economic/political resistance than purely scientific. Getting the science right is a long-term project that may take centuries.

  180. Tom_R (06:19:36) :
    >> Leif Svalgaard (20:02:12) :
    The energy and momentum of a photon are related by E = pc, where p is the magnitude of the momentum vector p. The momentum p points in the direction of the photon’s propagation and has a magnitude of p = h/L, where h is Planck’s constant and L is the wave length. So, photons do have momentum and can give you a kick and increase your kinetic energy. <<

    This is as I remember it from my basic physics. What I don't remember is what happens to the photon. If an air molecule gains kinetic energy from a collision, the photon must lose energy in the collision. Does the photon come off of the collision with a longer wavelength?

    No the photon is absorbed it does not emerge from the collision at all.

  181. Phil M. (05:44:51)

    “My advice to everyone here: do not visit this website for science. Peer-reviewed literature is where this debate should be happening, not the blogosphere.”

    This thread by Steve Goddard has attracted a lot of criticism from (apparently) AGWers for stating a rather simple fact: snow extent is increasing, in direct contradiction of general circulation models which predict global warming. The very simplicity of the statement appears to be offensive to some who perhaps need their science to be more complex and opaque.

    It has exposed a raw nerve for a clear reason: the global climate is cooling. There are a number of stark and inescapable implications of this:

    1) The AGW body of theory is a total failure;
    2) Every “scientist”, however many peers they may be able to lay claim to, who has embraced AGW, is a failed scientist: their instincts and methodology have proved wrong.
    3) Every scientist who, having looked at the evidence, has instinctively decided that global warming is false and un-grounded, for whatever combination of reasons, has been vindicated as a true scientist. This includes a fair proportion of regular posters on this site, several of whom are climate scientists who have been removed from research positions unfairly by the AGW inquisition exposed in the CRU emails.

    It is one thing to be a technician; it is another to be a scientist. If you are a technician, you can have a powerful grasp of the workings of a scientific methodology and body of theory in a particular discipline. You might be a capable mathematician or programmer, but your world is essentially one of methodology. A scientist by contrast goes a little further than methodology, and thinks beyond one particular corner of a discipline or subject. Shocking as it might sound, a scientist requires an accurate intuition, an ability to “sense” the likely relative significance of a range of factors in a complex system for example. (Sound methodology is necessary for good science but not sufficient for good science.)

    As the climate continues to cool, the peer-reviewed literature on AGW will be shown for what it is – the self-interested output of professional politicians dressed up as scientists, weaving technical language to construct inductive arguments supporting a pre-decided outcome – indefinite global warming. Such theories have their own internal logic and sense of rightness, but being inductive, are held up by a series of assumptions that represent a fragile and flimsy foundation. Holding such views shows ignorance and contempt for the totatity of climate record extending all the way back to the Cambrian era, with narrative confined to only climate history in the last 1 or 2 centuries or decades.

    Peer reviewed literature is where the politics is happening. Climate science that reflects reality is more likely to be found here.

  182. >> Phil. (08:48:21) :

    No the photon is absorbed it does not emerge from the collision at all. <<

    Then why is the sky blue?

  183. Ref – vigilantfish (07:17:49) :
    “…a dark age may arise, however, from the willful distortion of the scientific process for monetary and political ends.”
    _________________________
    Yes we all have to pinch ourselves every so often and remember we’re dealing with “people”.

    Ref – Leif Svalgaard (07:59:30) :
    “There is another underlying problem, namely that modern science may be intellectually overwhelming and some people are drawn to simpler pictures that they can ‘understand’ [it is easier to 'understand' that angels push the planets around than some mathematical equations; or that 'thunderbolts' from the heavens are powering the Sun rather than complicated nuclear processes]. As has been said so often, half of Americans believe the Earth is 6000 years old. Right there, one loses that segment of the population.”
    ________________________
    One of the first laws of Human Nature has the acronym KISS (Keep It Simple Stupid!). Einstein was brilliant at it; what a guy, who else could have said more with less? (E=mc^2) I fear that many “intellectuals” underestimate the intelligence of Joe & Jo the Plumbers. If nothing else, they & their kids are survivors. They can take anything the gods throw at them and in their magnanimity they support and pander to the stunted, crippled, sick among them who only know how to think well about one thing at a time. Life’s a beach, Leif! Don’t assume IQ tests have anything to do with wisdom or intelligence.

  184. Leif Svalgaard (07:59:30) :

    (…) As has been said so often, half of Americans believe the Earth is 6000 years old. Right there, one loses that segment of the population.(…)

    And the MSM also says there is an overwhelming scientific consensus that global warming is occurring, man is responsible, there is no debate, the science is settled, yada yada.

    Half of Americans (or more) believing in God and that God created the world? Credible. Half of Americans believing the Earth is 6000 years old? Not credible, as seen here “in the field.”

    The “elite” as found in the MSM and academia wanting to paint average Americans as ignorant unscientific hicks who obviously need to be led by “smarter” people, such as themselves, thus believing God created the world is labeled Creationism thus proof that half of Americans believe the Earth is 6000 years old? Credible.

    Why do “smart” people continue to imply, when they don’t outright state, that belief in God precludes one from understanding nuclear fusion in the Sun, and means they think angels push the planets around? When scientists violate scientific principles and insist there is no God without supplying proof, then these “scientists” find that people must obviously be unscientific because they continue to believe in God. Last I checked, even the most devoutly religious people in America believe lights come on when you turn on a light switch because of electrons moving in wires, not because moving the switch signals the angels in the bulbs to start glowing. Nor do they petition their electric company to stop charging them for power since the angels are doing the work. So why are “smart” people insisting that half of Americans are a permanently scientific-illiterate “lost cause”?

  185. Pascvaks (09:45:54) :
    One of the first laws of Human Nature has the acronym KISS (Keep It Simple Stupid!). Einstein was brilliant at it;
    Einstein once said: “keep it as simple as possible, but no simpler“.

    Don’t assume IQ tests have anything to do with wisdom or intelligence.
    Science has nothing to do with IQ-test, nor with wisdom, nor with intelligence.

  186. > Steve Goddard (21:13:04) :
    > The upwards trend started in 1989. Prior to that
    > there was a downwards trend. If I am showing a graph
    > of the most recent upwards trend, why would I include
    > years that aren’t part of that trend?

    If you want to assert that there was some kind of climatological regime change in 1989, there are statistical tests for distinguishing that from random noise, and it would help to have an explanation that comes from outside the snow data as to why 1989 should be a transition point. Monckton has rightly made great public sport of those who display selected subsets of noisy data in order to assert the existence of a trend.

    If you are showing a graph of the most recent upwards trend, you would include years that aren’t part of that trend in order to communicate honesty with the data.

    I have sliced and diced the snow cover data in all of the ways described by our colleagues above, including seasonally averaged variations from weekly-normalized means, with various appropriate smoothings. What emerges, as others including you have stated, is that maximum Winter snow cover has not changed much, while Spring and Summer has clearly decreased.

  187. kadaka (09:58:37) :
    Last I checked, even the most devoutly religious people in America believe lights come on when you turn on a light switch because of electrons moving in wires, not because moving the switch signals the angels in the bulbs to start glowing.
    How many did you interview?
    How were they selected? [how do you measure devoutness?]
    What was the exact question? Sometimes the wording can have an influence on the answer
    What is the error bar? i.e. how many agreed vs. not agreed?
    Do you plan to submit your finding to a peer-reviewed journal?

  188. Hi Steve,
    Thanks for getting back to me. It wasn’t clear in your article that was what you were trying to show. (I hadn’t read read your previous article I’m afraid and you didn’t point to it here.)

    I had a few other questions too, which I think are still relevant.

    I wondered if the trend in the Northern Hemisphere Annual Snow Cover Plot was sufficient to give you confidence that there was a trend – only it looks very slight, to tell the truth, and you don’t say what the error margin is.

    And I also wondered why you thought it might be useful to begin your plot from a minimum in the data. By picking unusual data don’t you think you might get unusual results?

  189. Ref – Leif Svalgaard (10:06:20) :
    Pascvaks (09:45:54) :
    P. – “One of the first laws of Human Nature has the acronym KISS (Keep It Simple Stupid!). Einstein was brilliant at it;..”
    L.S. – Einstein once said: “keep it as simple as possible, but no simpler“.
    _________________________
    My guess he was talking to a PhD; warning him not to underestimate “people”.

    P – “Don’t assume IQ tests have anything to do with wisdom or intelligence.”
    L.S. – “Science has nothing to do with IQ-test, nor with wisdom, nor with intelligence.”
    ___________________
    I’m sorry, what planet did you say you were from?

    PS: Ref your ref to WP article to Kadaka – :-)
    PPS: Exchanges like this are a waste of brain cells. I don’t have any I can afford to spare and you would be better occupied discovering something new for the rest of the tribe. God bless!:-)

  190. Leif Svalgaard (10:31:11) :

    kadaka (09:58:37) :
    Half of Americans believing the Earth is 6000 years old? Not credible

    http://www.washingtonpost.com/wp-dyn/content/article/2005/09/24/AR2005092401262.html

    Wow. I question the MSM presentation, and you rebut with an MSM article. Is this like when we “climate skeptics” question what the IPCC does with the research, and we get rebutted by references to the IPCC?

    Beginning analysis of spin…

    In the early 20th century, many creationist thinkers viewed Genesis as metaphorical, accepting that Earth formed over hundreds of thousands, even millions of years. But as society became more secular, and science offered an implicit challenge to fundamentalist beliefs, creationist leaders took a more literal line.

    Science was co-opted to promote atheism with the insistence that accepting science meant rejecting a belief in God, and believers pushed back against said “science.” Is that supposed to be surprising?

    Polls taken last year showed that 45 percent of Americans believe that God created humans in their present form 10,000 years ago (or less) and that man shares no common ancestor with the ape.

    As polls go, it helps in understanding the results to know who asked the questions, what groups were polled, and it is especially important to know the exact wording of the questions.

    As it stands, 10 thousand years is about the length of “recorded human history” thus is solid evidence of humans existing “in their present form.” If you take Genesis as metaphor, accept evolution and the actual science, with the caveat that God “tweaked” an existing pre-human primate to bring about humans, then yes, you could very well agree on a poll that “God created humans in their present form 10,000 years ago (or less)” and still not be a Young Earth Creationist.

    To recap, you are trying to affirm “…half of Americans believe the Earth is 6000 years old.” What have you provided? 10,000 years ago, not 6000, humans were created in their present form, not that the Earth is only that old.

    Plus, evolution as presented has changed over the years. My mother well remembers in school being explicity taught that man descended from the apes. A common ancestor for the seperate lines of humans and other large primates was a change in theory that came later. Thus, as originally taught, one may say that “man shares no common ancestor with the ape,” not be a Creationist, and only be faulted for considering a straight line yielded “no common ancestor.”

    One of the first things that science teaches us, is that the “reality” we experience with our senses isn’t even half of what really exists. Factor in current theory involving dark matter and energy, we don’t even realize a tenth of what is out there. Science needs to acknowledge its own limitations. At best it can state “we have found no evidence that a ‘higher power’ was involved at this time” and leave it at that.

    Instead we have “scientists” insisting science says there is no God, and that science and religion cannot coexist. We have this article, which takes poll results showing one thing, juxtaposes it with Young Earth Creationism, that ends up being cited as affirming that half of Americans believe the Earth is only 6000 years old, thus half of Americans are permanently scientifically illiterate.

    Let’s pull another fact from the article.

    Only 26 percent believe in the central tenet of evolution, that all life descended from a single ancestor.

    How is that unscientific? We haven’t even identified all the life on this planet, we are still discovering new life forms in extreme environments, and even ones overlooked where we had searched before. To say that all life must have evolved from the same microrganism ages ago is actually the unscientific position, as the same conditions that the one came from could have yielded others, and all life has yet to be identified thus we cannot draw such a conclusion about all life.

    And another…

    Another poll showed that 65 percent of Americans want creationism taught alongside evolution.

    So what is wrong with that? Open debate, presenting alternate theories, letting the research stand on its own… Isn’t that what we “climate skeptics” argue for in the presenting of climate science? If the scientific robustness is so obvious, then let it be obvious, toss it all out there and let people make up their own minds.

    Instead we get loud insistence that the science says this and only this, dissent is squashed as unscientific, demands that even the most basic of claims be backed with research, that should very well be peer-reviewed to count…

    Which brings us to Leif Svalgaard (10:43:13)

  191. >> Leif Svalgaard (10:31:11) :

    kadaka (09:58:37) :
    Half of Americans believing the Earth is 6000 years old? Not credible
    http://www.washingtonpost.com/wp-dyn/content/article/2005/09/24/AR2005092401262.html <<

    Are you referring to this statement?

    "Polls taken last year showed that 45 percent of Americans believe that God created humans in their present form 10,000 years ago (or less) and that man shares no common ancestor with the ape."

    Besides being 45% and not 50%, the question referred to how long humans have been around, and not the Earth itself. I have met people who stated they believed the Earth was only ~6000 years old (including a particle physicist, amazingly enough), but nowhere near 45%. My sample includes numerous co-workers at a can factory (summer job) who were happy to talk about the big bang when they found out I was an astronomy major, so I don't think it's just a selection effect.

  192. Ok, this has devolved into a science vs. religion discussion with Evolution and Creationism mixed in. This includes you Leif.

    Other moderators–why are you letting this through?

    This stops now.

    I mean it.

    Don’t make me go back and go through dozens of comments with with a scythe.

  193. Frequency Components of Rutgers University Global Snow Lab

    I removed an annual signal by working with the winter months of weeks 49 through 9, averageing the snow cover data. I subtracted the average value to work with the difference. The difference function initially declined and then inclined. A fit of a line through the data gave an R^2 of .000428 (very noisy data). Matching the difference function with a cosine curve with a period of 60 years and a 7.2 year shift gaive an R^2 of .0809, about 326 times higher than a straight line. I believe the NAO time period is on the order of 60 years.

  194. Re: charles the moderator (13:55:00)

    Charles,
    I am no fan of Creationism, especially Young Earth Creationism. I am not arguing science vs religion, from my perspective.

    I am complaining about science being used for an agenda, as we do with climate science specifically. I am upset by what I saw as half of my countrymen being declared as unable to understand science solely because of their religious views.

    That is all.

  195. Leif Svalgaard (07:59:30) :

    vigilantfish (07:17:49) :
    a dark age may arise, however, from the willful distortion of the scientific process for monetary and political ends.

    Such distortion would not be successful if the citizenry were scientific literate.

    Like in Germany?

  196. Jay,

    If you were measuring the height of a mountain, would you start the measurement at the bottom and end it at the top? Hopefully. You started the measurement at the bottom because that is the sensible place to do it.

    What if someone told you that you should have started your measurement on a neighboring peak of identical height, and that statistically speaking the topography is flat? Would you take them seriously?

  197. Phil M. (05:44:51) :

    My advice to everyone here: do not visit this website for science.

    Disregarding your obviously keen perception and acumen, I will continue to visit here; additionally, I may even see some valid science discussed.

    Perhaps, though, few would mind if you follow your own advice.

  198. Hi Steve,
    That’s an interesting response. If I wanted to measure the height of a mountain, then yes, I suppose I would start at the bottom and end at the top. What I wouldn’t do though, is suppose that I had discovered a meaningful trend and expect the mountain to keep on getting higher beyond the top. If you see what I mean.

  199. Roger Knights (16:55:55) :
    “willful distortion of the scientific process for monetary and political ends.
    Such distortion would not be successful if the citizenry were scientific literate.”
    Like in Germany?

    I didn’t know that Germans were distorting the scientific process any more than the US immigration rules against the feeble-minded…

  200. Phil M. (05:44:51) :

    actually i think the internet has done a pretty good job exposing the limitations and assumptions of the post. In short order. Now if the article had been in a journal, it would hang around as long as mann’s bogus de centering

  201. Charles the moderator

    Of course anti-religious intolerance and on the other hand religious bigotry should be excluded. However the creationism issue has been pushed into the political mainstream actively by a part of the Church. So it is fair game, surely?

    And using “religion” as a metaphor for (for instance) a political agenda and a new morality behind AGW, with historic parallels, need not be censored I would have thought.

    Reply: I didn’t want to go into specifics. This is not a site to debate religion vs. science. It is also not appropriate to lump religion as anti-science as I would concur this is bigotry. I personally know very religious medical researchers. ~ ctm.

  202. Phil M.

    An alternative to cherry-picking your start date, e.g. 1989 start for snow data, is the much more ambitious approach taken by the AGW movement – to change the historic record Stalin-air-brush style, to edit flat with a virtual iron the temperature record before the 1970s – much reduced 1930-40s warmth, no more MWP etc. Do you think this is preferable?

    Is it permissable to look at the evolution of any parameter with time? If so, then it is arguable that any start point after the beginning of the universe 13 odd billion years ago is “cherry-picking”. The accusation becomes meaningless.

    Since 1989 snow extent has increased. (Contrary to climate model predictions.) In its own terms a simple and valid statement. Live with it.

  203. DirkH (08:15:05) :
    Thank you very much for answering my questions. Of course I immediately have ten more questions.

    When I wondered:

    “Are Oxygen and Nitrogen warmed at all, when sunlight passes through them?”

    You responded:

    “No interaction between visible light and O2 or N2, otherwise air wouldn’t be transparent for visible light. So visible light that is reflected e.g. by snow will go outwards. There might be refraction by water droplets, so clouds will diffuse visible light. In the absence of clouds, the light reflected by snow (or white roofs, or any part of the surface) will just radiate into space without warming anything significantly.”

    However then I read, at the start of Roy Spencer’s UAH site:

    “Since 1979, NOAA satellites have been carrying instruments which measure the natural microwave thermal emissions from oxygen in the atmosphere.”

    I assume the oxygen must be able to hold heat, for them to be able to measure the heat it emits. This suggests that, while oxygen may not absorb heat directly from visible light, it does absorb heat through an indirect process, involving several steps.

    My question remains, is the atmosphere heated at all by visible light passing through it, and does the albedo-effect heat it twice, by bouncing descending long wave radiation back through it a second time?

    The place to test this would be just north of the Arctic Circle during the winter solstice. Although the sunlight never strikes the ground, just north of the Arctic Circle during the winter solstice, it passes very close to earth. In fact, (in theory,) if you stepped one step north of the Arctic Circle, an ant by your feet might be in shadow all day, but your face, an altitude of six feet, would briefly be bathed in sunshine at noon.

    Such sunshine passes through the atmosphere without ever touching the ground. It therefore would not be turned into infrared radiation by striking a physical surface. Would that air show any warming, during the period of time the sunshine passed through it?

    If it did, then the albedo-effect could be expected to doubly-warm the troposphere, for as long as the expanded snow-cover lasted. It then would create a sort of false-warming, for the atmosphere would appear to warm even as a great deal of heat was lost to outer space.

    To me, this is one answer to this post’s title, “Why Is Winter Snow Extent Interesting?” I don’t care so much about trend lines, as I care about the immediate effect of the snow cover.

    Another interesting thing to contemplate is how much heat will be sucked up (becoming latent heat) just melting all the snow.

  204. Leif Svalgaard (22:12:10) :
    “willful distortion of the scientific process for monetary and political ends.
    Such distortion would not be successful if the citizenry were scientific literate.”

    Roger Knights (16:55:55) :Like in Germany?

    I didn’t know that Germans were distorting the scientific process any more than the US immigration rules against the feeble-minded…

    In the 1st half of the 20th century Germany was the most scientifically literate country around, but it distorted the scientific process in, for instance, rejecting relativity, stacking scientific positions in universities with stooges, and indulging itself in some fairly crackpot anthropology, archaeology, etc.

    I think that “scientific literacy” has had the effect, in the modern Western world, of making its educated and informed elites more gullible than the scientifically benighted about the pronunciomentos of consensus science on things like catastrophic global warming (and certain other eviro-alarms like all in-place asbestos being a threat).

    “Man is so necessarily mad that not to be mad is but another form or madness.” (Pascal)

    “Man is a reasoning, not a reasonable, animal.” (Alexander Hamilton)

    What’s needed to avoid catastrophe (such as enviro-alarmism and hyper-precautionism would entail) is reasonableness, not mere reasoning (science). What’s needed also is an awareness of the basic madness of individuals and groups, including scientific consensuses.

    What’s dangerous is the scientistic insinuation of people like Bill Nye the Science Guy that the modern world has discovered, in the scientific method, a reliable (self-correcting-in-the-long-run) truth detector. The global warming episode shows that it takes outsiders to perform that correction — ironically including flat-earthers and creationists. Such persons, who aren’t intimidated by the parvenu authority of science, find it easier to razz the emperor. This is why the benighted, backward US (personified in Senator Inhofe), and not enlightened Europe, has resisted warmist doctrine and legislation, and thereby saved the world from a catastrophe greater than any inflicted to date by unscientific forces.

    Science is the institution that has risen to the top of the heap as social arbiter of reality and trustworthiness. That has provided “Satan” with a wonderful disguise to pull another fast one on humanity. “It’s always the one you least suspect” that is Satan’s Slave. Science has become too full of itself, too prideful, too powerful, too inbred, too narrow and unreflective. This has provided its practitioners and acolytes, perverse and base as they are — being human and wanting to make a name for themselves and throw their weight around — with a temptation to do “Satan’s work”: inflicting madness and suffering on humanity.

    But that attempt was a minor stain on science in comparison with the explicit endorsement of Hansenism by official science. Therefore, sweeping changes are needed in science’s funding, institutional set-up, procedures, and incentives — including second-thinking about the commonplace that scientifically absurd ideas can have only a malign social effect. That’s too superficially plausible to actually be true. We seem to need such absurdities as a safety valve for our basic nuttiness; lacking them, we’re in danger of becoming stark staring sane and going on a “scientific” rampage.

  205. Leif Svalgaard (07:59:30) :
    (…) As has been said so often, half of Americans believe the Earth is 6000 years old. Right there, one loses that segment of the population.(…)

    Reply: I didn’t want to go into specifics. This is not a site to debate religion vs. science. – ctm

    Nowhere did I mention religion. I simply referred to the fact (and WP seems credible enough on this point – surely the builders of the museum referred to would not contest that number or would rather see it increase) that half of all Americans believe the Earth is only 6000 years old. In this connection it is irrelevant if the actual percentage is 45% or 50% or 44.6783% or if the 6000 is 10,000 or whatever. On the other side of the coin, there are more people in India than the above number of Americans who believe the Universe is 311 trillion years old. Both numbers are indications of science illiteracy, and that is my point and has nothing to do with religion as such.

  206. Jay,

    Nowhere in the article did I make any predictions of future trends. All I did was locate the mountain and measure it’s average slope. Tamino tried to use creative statistics to make the mountain disappear. Which one of us is being more honest?

  207. Roger Knights (05:30:37) :
    In the 1st half of the 20th century Germany was the most scientifically literate country around, but it distorted the scientific process in, for instance, rejecting relativity, stacking scientific positions in universities with stooges, and indulging itself in some fairly crackpot anthropology, archaeology, etc.
    Relativity was not universally accepted back then. Einstein didn’t get his Nobel Prize for relativity. And, BTW most Americans reject the consequences of GR even now.

    What’s needed also is an awareness of the basic madness of individuals and groups, including scientific consensuses.
    I think it is politicians that claim there is a scientific consensus. It is only in a situation where a people is illiterate that their leaders can lead them down the garden path. Your notion that it is the absurd crackpots that are now saving the world is in itself absurd.

  208. In this thread climate (as described by time series) is awash with quotes of RSquared (R^2, R2) computed in virtually every case by performing a simple (linear) regression of supposedly homogeneous observations on date or time. It also seems to me that some posters may not be familiar with the background to RSquared and its relationship to “the real world”. It’s worth remembering that although RSquared is a genuine and useful statistical measure of “goodness of fit” of a model (any model) to the data, it is not at all informative about the value of the model, simply because in order to convert it to /useful/ inferential statistics you must also provide further information, to wit the quantity of information on which it is based. Statistical tables exist for this purpose, but there are various methods in the maths of regression analysis to do the same thing. If there’s not very much – say a few tens of observations – there’s a further complication because the data are a time series, with almost inevitably considerable serial correlation, so there has to be an adjustment to the quantity that describes the amount of information by methods that have been extensively discussed in Climate Audit. For large data sets, these adjustments, which affect the probability statements that can be made about the data, are I think unlikely to affect the outcome seriously. However, this does not, in my opinion, throw much further light on the scene.

    What would be more informative for most readers, I venture, is a simple statement of the estimated coefficient of interest (typically the increase/decrease of some parameter over a given time interval) together with its confidence intervals at a specified probability level, say 90% or 95%. This information provides an instant, simple and unmistakable message about the analysis. If the confidence interval includes zero, the estimated slope cannot be significant at levels greater than the one on which the interval is based.

    Something else that I always hope to see is a graphic that shows the the fitted (usually least squares) line together with the two hyperbolae that describe the confidence interval for the line, and the other hyperbolae that describe the interval within which a future observation from the same population can be expected to fall at the chosen confidence level. The software I use does this, and I remain surprised that nobody else presents such information. I don’t know how to post graphics to this thread, so can’t illustrate what I mean. For those of you who like to think of fitting higher order model to data, such plots can be most enlightening. The confidence intervals for extrapolations of a cubic fit can really be quite alarming.

    Now you can legitimately choose whatever confidence level you fancy. There’s absolutely nothing sacrosanct about 95% or 90%. These merely express the betting odds at which you are prepared to make a quantitative statement about the outcome of the analysis. You might well decide that 30% was adequate, because it provides narrow confidence bands and would apparently support your hypothesis despite rather nonedescript data. In the days when I was an industrial scientist involved with trying to persuade engineers, chemists and physicists to adopt statistical design of experiments to improve their productivity, I reached the conclusion that in their scouting experiments they instinctively tended to use an approximate 70% level on which to base their further actions. Who can blame them, if these actions were not potentially very expensive if they finally proved to be misguided. However, the roughly 2:1 chance of being right was sufficient to persuade them that they were “doing the right thing”.

    In earth-shaking climate science the choice of confidence levels in assessing the import of noisy data is something that I feel should be carefully and pro-actively contemplated by all those involved, included those influential non-scientist and non-statistician outsiders who take world-affecting decisions based on data that are not always what they seem.

    I hope that this makes sense to everyone!

  209. Robin Edwards,

    I used a stock market program which backfitted stock data with extremely precise fits using multiple order polynomials. Unfortunately, the equations were completely worthless at predicting future behaviour of the stocks.

    “with four parameters I can fit an elephant and with five I can make him wiggle his trunk”
    Johnny Von Neumann

  210. Steve Goddard.

    Of course, Steve! Did the software also produce the confidence interval plot? This is almost always a nasty surprise to aficionados of polynomials.

    I’d guess that with five parameters even Von Neumann would have had some problems with elephants, but it’s certainly a quote worth remembering.

    Robin

  211. Tom_R (09:38:24) :
    >> Phil. (08:48:21) :

    No the photon is absorbed it does not emerge from the collision at all. <<

    Then why is the sky blue?

    Rayleigh scattering.

  212. Moderators,
    In kadaka (12:43:24) : (awaiting moderation), “…state that, that it does say that…” should be “state that, that is does not say that…”
    Can you please correct that?
    Thanks!

  213. Been two days now.

    So are no further comments being accepted for this thread? Perhaps that should be noted.

    Oh, and if kadaka (12:43:24) (Feb 20) has been lost, can you delete kadaka (12:48:02)? It looks silly just hanging there.

    Thanks!

Comments are closed.