Even when the numbers are right there for you, you prefer to make them up to incorrectly assert a linear fit.
from the data linked at the Mauna Loa monitoring site:
(
ftp://ftp.cmdl.noaa.gov/ccg/co2/trends/co2_annmean_mlo.txt)
2000 369.41
2001 371.07 0.45% change from 2000
2002 373.16 0.56% change from 2001
2003 375.81 0.71% change from 2002
2004 377.54 0.46% change from 2003
2005 379.78 0.59% change from 2004
2006 381.86 0.54% change from 2005
2007 383.73 0.49% change from 2006
2008 385.54 0.47% change from 2007
2009 387.35 0.47% change from 2008
2010 389.78 0.63% change from 2009
There is no 1% per annum, linear "best fit" to the data, even at this grossly truncated segment level of examination (and as a general RoT, the smaller the data sampling and the more averaged the numbers in that sampling, the easier it is to force a linear curve onto the data). At the scale of climate relevent time periods (3 X 20 = 60 years) there is a slight apparent geometric progression taking place, and if this is coupled to a multiple century (3 x100 = 300 years) data plotting, a logarithmic nature is more apparent in the data character.