You may want to look at my post on creating multiple masks from a single image map. You can take the same approach with a PF as well.
One possible approach is to create a PF mask for the overall distribution distribution of trees/ clearings. From here you then need to make a number of different versions of the same mask using colour adjust nodes to isolate different ranges values and combine these using multiply/add/subtract scalars to build up density masks for each population.
Take it slowly...
Younger trees are at a higher denity at the edges of the forest (lower values of mask)
Oldest trees are at a higher density at centre of the forest (high values of the mask)
Middle aged trees are in between these two populations.
I'll try and make up a simple demo mask for you.
Here are some intermediate keyframes of my leaf growth. The next series will be even closer to the tree to show a bit more of the detail changes.