Until it stopped changing slowly, and started changing 100x faster, in the past 50ish years.
Until it stopped changing slowly, and started changing 100x faster, in the past 50ish years.
No chart-based argument is ever fair if it does not show all available data and includes zero. Which is to say that it needs to cover about 4.5 billion years and show temperature in degrees Kelvin. Not that it isn't rhetorically effective; it's just not fair.
To restore the effectiveness, you'd probably have to put everything on a log scale, also chart the first derivative of the mean global temperature function, chart the partial pressure of CO2, and label all the mass extinction events.
http://commons.wikimedia.org/wiki/File:All_palaeotemps.png might be a good starting point.
The facts speak for themselves; no one should need to be tricked in order to believe them. And cherry-picking is a rhetorical trick. The rate of change in mean global temperature is at a level that is a hundred times more rapid than the warming and cooling rates seen in previous geological eras. And while we are still under the Eocene maximum for absolute mean temperature, biological evolution needs time to keep up with temperature change. Species that produce new generations on the scale of decades can possibly handle a 5 degree change, when it occurs over 100000 years, but not so much when it only takes 100 years. If you swing the needle on the temperature gauge fast enough, only the species that can produce new generations in a matter of hours will survive it.
Superfluous data can distract from salient points; for example it's simply not relevant to the discussion of life on earth what the weather was like when the earth was still a ball of lava. And sure, the temperature changed quite rapidly when the moon was blown off the side of the earth. But how is that important?
If you conceal data because it is distracting, the person from whom you are hiding it cannot know why you did that. They only know that they cannot see what you refused to show.
If it is not relevant, it is better to show it, and explain why it is not relevant as best as you are able. And that explanation can take the form of an arrow that marks the earliest time life was suspected to have first developed.
But in showing it, an observer might be able to discern temperature trends due to changes in the sun, meteor and comet bombardments, loss of core radionuclides, radiation into space, or other cosmological factors not attributable to humanity, or other forms of life. It might show a reason why the Earth today should be cooler now than it was at the Eocene maximum.
It's just more honest. You can still show a second graphic that cuts out the pieces you feel to be unimportant, and which highlights the pieces that best make your point.