In addition to BGC projections for the individual GCM runs, we will want the ensemble BGC projection.
we will need two layers:
the vote-winner at subzone variant level
the vote-winner at zone level.
for example, if the distribution of votes for a grid cell is CDFmm (40%), CWHxm1 (30%), and CWHdm (30%), we would want the subzone level winner to be CDFmm, but the zone level winner to be CWH.
In addition to BGC projections for the individual GCM runs, we will want the ensemble BGC projection. we will need two layers:
for example, if the distribution of votes for a grid cell is CDFmm (40%), CWHxm1 (30%), and CWHdm (30%), we would want the subzone level winner to be CDFmm, but the zone level winner to be CWH.