There's two ways. One is by analysing isotopes in pollution to find whether the carbon contains a "fingerprint" of human activity, but the easier way to understand is that alternatively there's only two places for the CO2 to have come from, the oceans or the land and if it was emitted from either there would have been a carbon reduction measured in either. There have been more than 20 published studies using 6 independent methods of the oceans, for example, and no carbon decrease has been found.
Yes, el nino and seasonal changes definitely have an effect. However, here's the trend, including each el nino year and seasonal variation:
<pic>http://upload.wikimedia.org/wikipedia/commons/8/88/Mauna_Loa_Carbon_Dioxide.png</pic>
As a trader, can you see a trend in this stock?
(In your cite he pulls a little sleight of hand and shows a graph not of the total CO2 in the atmosphere like this one, but a graph showing each year's growth of CO2 from the previous year. So his graph looks flat.
As well he omits a key piece of information about isotope monitoring -- he claims that this odd ratio could be due to a biological explanation -- he doesn't know that they've managed to pull C13 out of tree rings and ice cores and contrary to his claim the current wild ratio is not consistent with anything in history. Ooops for him!
