The future of the land carbon sink is a significant uncertainty in global change projections. Here, key controls on global terrestrial carbon storage are examined using a simple model of vegetation and soil. Equilibrium solutions are derived as a function of atmospheric CO2 and global temperature, these environmental variables are then linked in an idealized global change trajectory, and the lag between the dynamic and equilibrium solutions is derived for different linear rates of increase in atmospheric CO2. Terrestrial carbon storage is departing significantly from equilibrium because CO2 and temperature are increasing on a similar timescale to ecosystem change, and the lag is found to be proportional to the rate of forcing. Thus peak sizes of the land carbon sink, and any future land carbon source, are proportional to the rate of increase of CO2. A switch from a land carbon sink to a source occurs at a higher CO2 and temperature under more rapid forcing. The effects of parameter uncertainty in temperature sensitivities of photosynthesis, plant respiration and soil respiration, and structural uncertainty through the effect of fixing the ratio of plant respiration to photosynthesis are explored. In each case, the CO2 fertilization effect on photosynthesis is constrained to reproduce the 1990 atmospheric CO2 concentration within a closed global model. New literature compilations are presented for the temperature sensitivities of plant and soil respiration. A lower limit, Q10=1.29, for soil respiration significantly increases future land carbon storage. An upper limit, Q10=3.63, for soil respiration underpredicts the increase in carbon storage since the Last Glacial Maximum. Fixing the ratio of plant respiration to photosynthesis (R/P) at 0.5 generates the largest and most persistent land carbon sink, followed by the weakest land carbon source.