Because Warmer Oceans Cause More Snow Or Something

Some of the nutters have been spewing some nonsense since even before the snow storm in the Northeast hit.  They’ve babbled all sorts of silliness…. warmer oceans increase the snow and whatnot.  Does the ocean temps control our snow?  Well, sure, within parameters, and about 1000 other factors.  For instance, if our oceans were all ice, then we probably wouldn’t have much snow.  But, we’d have a lot of other problems to deal with.  But, the oceans aren’t all iced over. 

We all know by now, or should by now that there hasn’t been any significant change in the snow cover for the Northern Hemisphere when looking at it year round.  But, I thought I’d check to see if there has been for the periods of time when it snows the most often in most of the Northern Hemisphere,  Nov through March.  Mostly because, obviously, it isn’t going to snow in July for most of the NH.  So, one is just recording melt during that time of the year.  We’re interested in if it snows more for this little excursion. 

image

source

Well, the nutters got it wrong again.  Well, wait, just to be sure, let’s examine the sea surface temps of the NH over the about the same time period……

image

source

This is the problem with the nutters.  They’re not very bright and are only two dimensional thinkers. (There are other terms)  They can’t go beyond silly little correlations and it screws up their conclusions.  They see temps and CO2 momentarily correlate and they say “aha!”  X must cause Y!!! 

Similarly, if one looks back up to the snow  graph, one can see a slight decrease in the snow from about the mid 1980s to the late 1990s.  This is probably why they had a couple of pinheads declare snow was a thing of the past and that the ski industry was all but dead.  But, since then, we’ve seen a slight and gradual increase.  But, the temps haven’t really changed any since from about the late 1990s to early 2000s. 

So, what does this tell us?  It tells us a few things.  One, snow isn’t all that dependent upon temps, within certain parameters.  And two, there’s at least one factor with is much or important than the slight deviations we see in our temps.  There are many, but the observations of snow extent deltas point to a regime change of some sort.  (Might that regime change have something to do with the delta in the temp trends?)

Look mom!!!  A near perfectly symmetrical 2 order poly fit!!!!  Sometimes, it gets tiresome dealing with these simplistic buffoons. 

Bonus points for why it might not be so obvious in the years prior!!! 

About these ads
This entry was posted in Climate. Bookmark the permalink.

34 Responses to Because Warmer Oceans Cause More Snow Or Something

  1. gator69 says:

    Leftists don’t really want to solve problems, they just like to whine about them, lord over everyone else, and make them miserable.

    Reminds me of someone I used to know, and why I am still single!

  2. tckev says:

    Look it’s simple –
    pay more tax and the government will improve the weather.

    /sarcoff

  3. HankH says:

    They can’t go beyond silly little correlations and it screws up their conclusions. They see temps and CO2 momentarily correlate and they say “aha!” X must cause Y!!!

    There’s several problems with their statistical skills: 1) the oft repeated correlation is not causation is the first to come to mind, 2) any linear trend that changes slope indicates that there are other independent variables involved which can be more statistically significant than CO2, 3) it is often ignorantly asked “if not CO2 then what?”, meaning if you can’t answer “what” then “what” doesn’t exist. Referring back to 2, “what” does exist whether it can be identified or not. This is a very basic rule of statistics.

    • suyts says:

      Thanks for articulating that, Hank. There’s got to be a way to verbalize this notion to allow some other people to see this silliness. .

      • HankH says:

        I find that people who resort to the argument of correlation are too ignorant to understand what any statistician would consider the most basic rules of correlation. I’ve debated this whole line of [lack of] reasoning for years. It always turns into “you’re not a climatologist” appeals to authority – like my being a biostatistician does not qualify me to weigh in on statistics when it comes to climate. Nothing has changed since Francis Galton and Karl Pearson (the fathers of statistical math) yet they insist global warming affects everything, including statistics.

      • suyts says:

        So true. Math, statistics, programing, even database management….. none of it stacks against a climatologist! Except, …. this is what a climatologist uses for his arguments.

      • HankH says:

        Lame arguments at that. That’s why became skeptical. I started out believing in GW but upon further investigation I found a bunch of snake oil salesmen masquerading as scientists pretending they know statistics.

        • suyts says:

          Exactly. I’m not a statistician, but, I know more than most of them. It got me looking at things and I looked for the data. It’s in a miserable condition and will remain as such. They’ve no desire to put this stuff in a proper data base nor maintain the integrity of the data. But, they’ll beg for computers to run their models…. based on crap data, and crap statistics. Once I saw that, I knew it was a farce. No self respecting statistician, engineer, computer programer, or database manager would run with this stuff.

        • HankH says:

          I totally agree. There’s no way I would base any conclusions on such crappy data, especially the made up (adjusted) data they use.

          I think the climategate files were very telling of how shoddy data management is in climatology. The big focus of most people was in defending characters like Michael Mann and Phil Jones. It seems the defenders are blissfully ignorant of how incredibly bad data were maintained and used. You would think scientists would be meticulous about such things. My personal view is if you can’t manage data you can’t do quality research. Ergo, the quality of climate research is abysmal.

        • suyts says:

          That’s the way I see it. I think they know if the data is properly kept and processed that the results would end their farce, so they don’t. Yeh, the climategate file with the programmer said it all.

        • HankH says:

          If I had written a Harry ReadMe file and disclosed what he did, every paper I’ve published would have to be retracted. :oops:

        • suyts says:

          Yep, the thing that struck me was the palpable frustration in his comments.

  4. cdquarles says:

    James, it would be interesting to get an absolute graph to put next to the anomaly one. One of the issues with anomaly graphs is that it is easy to fudge the baseline (kind of like the Federal Budget). Without knowing what the baseline is for the anomaly graph, you can get the wrong impression. I would like to know, for instance if the SST baseline is 0 degrees C (32 F). That would give some kind of perspective to the anomaly graph. We could do the same for the land one, but I have heard different values exist for it (14 C, 15 C). What epoch the baseline is calculated would also be something we need to know (51 to 80, 61 to 90, etc).

    • suyts says:

      Yeh, I keep meaning to do that. I’ll try to make it a point in the near future.

      • HankH says:

        James, if you can find an unadjusted database(s) on CO2 and temperatures that span back to the 1800′s point me to it. I would like to do my own bivariate analysis of it for grins. The problem is most of the data has been adjusted, making it useless to say anything about natural relationships.

      • suyts says:

        The CO2 measurements will be tough. I don’t know that such exists. As to the unadjusted, I think Steve had a post on it, though, it might just be for the US data. If I have time I’ll try to run it down for you, but again, I don’t believe the CO2 measurements exist.

        But, here’s something for you to chew on. I once did a backcast of the CO2 measurements. I tried several different approaches. It has been increasing exponentially (sort of) so I thought it wouldn’t be that difficult to estimate what it was in the 1800s. I couldn’t get it to work. I kept coming in much lower than the imaginary 280 ppm. Which, we know isn’t possible. Modern plants don’t thrive well in very low CO2 environments. I think the assumptions of pre-industrialized CO2 levels are wrong. But, that’s about the limit of my skills and knowledge.

  5. HankH says:

    If you’re using a linear model it will definitely give you a much lower (than reality) backcast. You need to use a logistic function like a Sigmoid. Then curve fit the model to your existing CO2 data. Then you can use X (relative time) as your predictor to backcast. Here’s an example of a quick model output I did using a Sigmoid function:

    http://img694.imageshack.us/img694/9117/co2model.png

    It’s not fitted to actual data. To do a proper fit of the model to data I would use SPSS, plug in the real data, and test for covariance between the model output and the data to fine tune the model for a good fit.

    If you can point me to a datasource for CO2, I wouldn’t mind playing around with it and forward to you the results. The bold assumption in all of this, as you point out, is that 280 is indeed a real value. If it’s not then all bets are off.

    • suyts says:

      Here’s what I used for a source….. http://www.woodfortrees.org/data/esrl-co2
      Yes, the linear does put it way lower than what it’s suppose to be. I tried the logistic function, but, it didn’t seem to fit well. Of course, it could very well be that I was making some mistakes. It was a while back when I was playing with it. I wasn’t allowing for the 280 ppm to be assumed as true. I ended up with a good curve to about 1900 but I kept having to move it up and down the timeline to make it fit.

      • HankH says:

        Thanks! I’ll play with the data over the next few days.

        I’m probably thinking about the problem a little differently than you. Since you were testing your model to see how well it agrees with 280, then in my mind 280 would be the intercept for the model. Assuming 280 is a valid number and the model is tuned to fit well to the known data, I would then expect X to be a fairly good predictor to backcast from and it should agree, on average, with any known value for Y (CO2). Using my approach, if 280 isn’t right then the model would be pretty worthless for backcasting.

      • suyts says:

        Cool! I’ll be interested to see what you come up with, or, at least see where I might have gone wrong.

        • HankH says:

          Here ya go. Basically I used a Sigmoid function. I then tweaked the function until I had a good fit with the historical data. I think you were on the right path in your initial attempt. The only thing I think I did different was to set the intercept to 280 to constrain the lower tail of the curve. Anyway, the resulting graph:

          http://img507.imageshack.us/img507/5904/co2estimator.gif

          Compare it to this. I think it’s a pretty good fit.

          http://www.woodfortrees.org/plot/esrl-co2/from:1958.5/mean:12

        • suyts says:

          Thanks Hank! “The only thing I think I did different was to set the intercept to 280 to constrain the lower tail of the curve.”

          Right, but, then that’s the question I was trying to answer. Did they, or could they mathematically come up with the 280ppm with the current curve to begin with? Without setting the intercept at 280 it’s just a shot in the dark. So then, the question is, where did they get the 280ppm from to begin with?

        • HankH says:

          My understanding is they got that value from ice core data. NOAA claims it hovered around 280 ppm until 1850 when anthropogenic CO2 started to kick in.

          http://www.noaanews.noaa.gov/stories2008/20080423_methane.html (third paragraph).

          I don’t think they could have come up with an accurate value using the curve from historical data and knowing nothing else. The reason being that limited view of the curve wouldn’t really tell them much about the kurtosis – the rate at which the distribution approaches the tail of the curve, or the skewness – the asymmetry of the curve – if any. They would have had to guess at those, almost guaranteeing a wrong answer.

  6. cdquarles says:

    Early atmospheric carbon dioxide measurements were, from what few sources that I have been able to see, varied all over the map; and many are method dependent. Measurements basically began in the early to mid 1800s. Svante Arrhenius was the chemist who most of the warmists cite. There are issues with some of his assumptions. The 280 figure comes from ice cores. I sure would like to see some ice core methods papers and their raw data, for the act of coring has the potential to substantially bias measurements.

  7. Jim Masterson says:

    >>
    suyts says:
    February 12, 2013 at 11:34 am

    I’ve a lot of difficulty with ice core measurements.
    <<

    A lot of us do–especially Dr. Jaworowski.
    http://www.warwickhughes.com/icecore/

    Jim 8-)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s