To forecast3 the state of global biodiversity in 2100, we need to forecast how the drivers will change between now and then. Sala et al. [6] use the ``business as usual'' scenario generated by global models of climate, vegetation, and land use. So what they provide is a projection, not a forecast. They tell us what will happen if we project current trends into the future. They don't try to suggest how those trends might change. It's a reasonable approach, and it illustrates again that knowing the direction we're headed may be more important than knowing the final destination when trying to manage conservation problems. After all, the fact that we're managing them at all means that we are at least considering changing directions. So we need to be able to project where we'll arrive if we keep heading in the current direction and to compare it with where we would be if we changed our direction. You want a forecast when you can't (or don't want to) influence what's going to happen. You want a projection when you do want to have an influence.
They ranked the amount of change for each of the five drivers on a
scale of 1 (small) to 5 (large) (Table
1). Using the same scale (1 
low, 5
high), they then ranked the biodiversity impact of a unit change in
each of the five drivers in each of the ten
biomes (Table 2), where one unit is defined as:
C change or 30% change in
precipitation,
year,
: 2.5-fold increase (as projected) by
2100.