European Tribune

Display:
Some notes on the issues raised by the commentators:
  • ENSO - While the index divided to -2 in February it has recovered in the meantime now staying at 0. The temperature anomalies have been stronger in May and June.

  • Six months of data - While six months of data can not be used to fully characterize the year's anomaly they are enough to perceive where it will be relatively to past years. For 2008 to reach an anomaly similar to past years around +0.3 șC the following six months would need to average around +0.6 șC. There are just three months equal or above +0.6 șC in UAH's dataset and six in RSS', all occurring in 1997 - 1998. The probability of having just one month at +0.6 șC is thus below 2%, the probability of having six straight of such months is infinitesimal.

  • Trends - There's no trend fitted on any of the graphs. Many are the curves that can be fitted to each of these datasets, depending of the period considered and the fitting method used. To know what's ahead a good understanding of meteorologic phenomena and the mechanisms behind them is better than any trend.

  • Outlier - For an year with a null anomaly so far that's an hilarious characterization.


Vencit omnia veritas.
by Luis de Sousa (luis[dot]a[dot]de[dot]sousa[at]gmail[dot]com) on Mon Jul 7th, 2008 at 03:18:37 AM EST
  1. Comparing six-month to twelve-month averages is still sloppy. If you want to make a point, you could compare the relevant six months of all the past years to the six months you have data for this year. I don't know what that would yield; I haven't run the numbers. But then again, I'm not the one trying to prove a point using those data.

  2. You don't have to fit to see a trend in those data. Chi-by-eye more than suffices. As an aside, we know that there is an effect of solar activity, and we know that it is cyclic on the scales presented here. But it doesn't explain the underlying upwards trend (as another aside, it is - among other things - to smooth out this cyclic behaviour that one normally looks at global mean temperatures over a 30-year running average).

- Jake

Ceterum censeo Chicago esse delendam
by JakeS (JangoSierra 'at' gmail 'dot' com) on Mon Jul 7th, 2008 at 03:47:15 AM EST
[ Parent ]
six months of data Isn't it better to just do a moving average with a 12-month trailing window?

When the capital development of a country becomes a by-product of the activities of a casino, the job is likely to be ill-done. — John M. Keynes
by Migeru (migeru at eurotrib dot com) on Fri Jul 11th, 2008 at 04:12:44 PM EST
[ Parent ]

Display:
Login
. Make a new account
. Reset password
Recommended Diaries
Recent Diaries
Debates
Campaigns
Occasional Series