Register to Comment
Page 3910 of 6680 FirstFirst ... 291034103810386039003908390939103911391239203960401044104910 ... LastLast
Results 39,091 to 39,100 of 66791
Like Tree14135Likes
  1. #39091
    storybud1 storybud1 is offline
    storybud1's Avatar
    Join Date
    Oct 2011
    Posts
    6,548

    Tax anything made with coal power,
    any car under 100gm pays no tax
    Offer a billion Euro prize for a cleaner energy plan to reduce emissions by 50% in said Country, heck offer 20 billion for an ultraclean engine/power generator

    Is it so hard to offer incentives/necessity , it is the mother of invention , that is all we need.
    Sign in or Register Now to reply

  2. #39092
    barry schwarz barry schwarz is offline

    Join Date
    Aug 2012
    Posts
    1,414

    Quote Originally Posted by oneshotleft View Post


    The climate models used by policymakers around the world to estimate the economic and social costs of CO2 emissions have to be improved according to Thomas Sterner, professor of Environmental Economics at the School of Business, Economics and Law, University of Gothenburg, and six other scientists in the prestigious journal Nature.

    Claim: Climate Models Underestimate Costs to Future Generations | Watts Up With That?
    The graph is of the mid-troposphere, not surface temps. There are known issues with tropospheric data and and/or modeling, which has nothing to do with GHGs.
    Sign in or Register Now to reply

  3. #39093
    barry schwarz barry schwarz is offline

    Join Date
    Aug 2012
    Posts
    1,414

    Quote Originally Posted by owedtojoy View Post
    There is a very good trend calculator here.

    It gives you error bars that include allowance for autocorrelation. Temperature data is autocorrelated, i.e. a warm year will tend to be followed by a warm year because the surface cannot cool fast enough (ditto for colder years). Therefore the independent observations assumption behind Ordinary Least Squares is not true .... but that is not a disaster.

    The major effect is to inflate the variance of the trend estimate. Which is why this calculator is so useful. Not sure if Nick Stokes' trend calculator does that.

    Temperature trends
    I often use that regression calculator.

    One of the easiest ways to reduce autocorrelation is to plot annual instead of monthly temps. Autocorrelation is strongest at daily/monthly resolution.
    Last edited by barry schwarz; 9th April 2014 at 02:17 AM.
    Sign in or Register Now to reply

  4. #39094
    barry schwarz barry schwarz is offline

    Join Date
    Aug 2012
    Posts
    1,414

    Bit more fun with graphics. Just for giggles.

    UAH 1979 to 1998 (incl) trend with "error bars."



    With the slowdown in temps after 1997/8, we'd expect to see values hovering in the bottom half of the bounds.

    Sign in or Register Now to reply

  5. #39095
    barry schwarz barry schwarz is offline

    Join Date
    Aug 2012
    Posts
    1,414

    Another look at Christy/Spencer's model obs comparison.

    Models are CMIP5, based on RCP, which starts forecasting from 2006 - everything before is hindcasting.

    Note: for the near term, it really doesn't matter which RCP you use, as they show little divergence for the first few decades of the 21st century.

    So what happens when you baseline using a period 1986-2005 (20 years, ending right before forecast period)?



    FTR, I don't claim this is the most appropriate baselining, just that other choices can be made, which affects the results.

    And what does it look like when you use 5-year running means, as Christy/Spencer did?



    Ok, that looks a lot more like their graph.

    So what happens if you use the Christy/Spencer baseline choice of 1979-1983, and zero all the models at 1983 (which is NOT a valid baseline choice)?



    Look familiar?



    John Christy, Richard McNider and Roy Spencer graph

    While there is some validity in choosing the baseline period 1986-2005, it is not necessarily the most appropriate choice, but it at least goes right up to the forecast period. C/S do not explain their choice, and do not match model baseline to UAH.

    The point is, that while rebaselining has no effect on trends, it does affect model obs comparisons.

    So now we come back to Trainwreck's postulate - that model trends deviate significantly from observations for the period 1983-2013.

    Still waiting for TR to support his case.....
    Last edited by barry schwarz; 9th April 2014 at 04:53 AM.
    Sign in or Register Now to reply

  6. #39096
    correr correr is offline

    Join Date
    Dec 2010
    Posts
    885

    I wonder what happened to the MV Akademik Shokalskiy the expedition ship that the climate change alarmists used to sail to the Antarctica to 'prove that the ice there was melting and instead had an epic fail by getting stuck and requiring that the loons on board be rescued by helicopter, priceless.



    She appears to be still stuck from the latest article I could find.

    https://www.google.ie/url?sa=t&rct=j...oltlM1d04uqxAw
    Sign in or Register Now to reply

  7. #39097
    barry schwarz barry schwarz is offline

    Join Date
    Aug 2012
    Posts
    1,414

    Trainwreck,

    1) What is the trend period Spencer is referring to in his model/obs comparison? (1983-2014)

    2) What is the UAH trend for that period? (0.163C/dec +/- 0.084)

    3) What is the mean model trend for that period?

    With uncertainties, please.
    Sign in or Register Now to reply

  8. #39098
    barry schwarz barry schwarz is offline

    Join Date
    Aug 2012
    Posts
    1,414

    Quote Originally Posted by correr View Post
    I wonder what happened to the MV Akademik Shokalskiy the expedition ship that the climate change alarmists used to sail to the Antarctica to 'prove that the ice there was melting and instead had an epic fail by getting stuck and requiring that the loons on board be rescued by helicopter, priceless.



    She appears to be still stuck from the latest article I could find.

    https://www.google.ie/url?sa=t&rct=j...oltlM1d04uqxAw
    From the article:

    The Shokalskiy arrived safely back in Bluff on 14 January.
    Sign in or Register Now to reply

  9. #39099
    barry schwarz barry schwarz is offline

    Join Date
    Aug 2012
    Posts
    1,414

    owed, you've probably seen this already, but Nick Stokes discusses various models for calculating standard error from a linear regression.

    moyhu: Uncertainty of temperature trends.

    Mentions SkS standard error method.
    Sign in or Register Now to reply

  10. #39100
    Earthling Earthling is offline
    Earthling's Avatar
    Join Date
    Aug 2012
    Posts
    7,840

    Quote Originally Posted by barry schwarz View Post
    Bit more fun with graphics. Just for giggles.
    And your own amusement, putting you on par with Sid and Odious.
    Last edited by Earthling; 9th April 2014 at 07:51 AM.
    Sign in or Register Now to reply

Sign in to CommentRegister to Comment