Yes the Landsat imagery would be good for determining tree species, I have not used but am aware of some in-situ remote sensing devices that are for on the ground monitoring. I believe that you would be right though in that the leaf cover would crate too much of a challenge to use multi-spectral sensors. Looks like you will have to stick with the old school method of actually going to the sites in person!

It would be nice to build up a GIS community on Steemit where people can actually earn value for their contributions to others work around the world when answering or debating questions/ techniques.