Many nonlinear problems can be approximated as linear and in this case the rough sea level rise (think altitude/distance) so ridiculously dwarfs the radius of the earth (the square in your argument) that the linear approximation is pretty good.