I think I see the source of your confusion
Laurie's chart has some verbiage at the bottom about the "1979 to 2000 mean = 15.6 million sq km" This has nothing to do with the regression line drawn through the data, it's just a way of giving the viewer some idea of the areas involved since the chart is normalized to percentages. In the charts I link to, they graph area directly against time, and that kind of clarification is unnecessary.
