People interested in this might also be interested in the release notes for The Grove (a Blender tree gen plugin), as they describe a number of modelled behaviours of trees in nature: phototropism, gravitropism, forking, behaviour of sapwood and heartwood, etc.: https://www.thegrove3d.com/releases/the-grove-release-8/
I love that Blender is so programmable. As far as I'm aware, every action in Blender can be expressed in Python. The Grove in an absolutely beautiful representation of the power that comes with that.
A recall a undergrad project where I was helping write a C wrapper around an old FORTRAN library or potential energy calculations to make them accessible in Mathematica (and yes, I am aware of how wild that sentence is). We used Mathematica to generate some potential energy surfaces (they looked like the first picture here https://sav.blogs.unr.edu/3d/ ) and for an interesting visuals I exported the models to Blender. An interesting property of these graphs is that modeling a reaction is taking a point on that graph with some amount of energy and combining it with the gradient at that point. This is essentially the same as simulating a ball rolling around within the surface. So I simulated a ball in Blender, tracked it's motion, and compared it to real life data and "actual" simulations, and found them impressively similar.
I ended up winning some aware for presenting it, it was quite a lot of fun, and now I have a publication and conversation starter of an entry on my CV.
People interested in this might also be interested in the release notes for The Grove (a Blender tree gen plugin), as they describe a number of modelled behaviours of trees in nature: phototropism, gravitropism, forking, behaviour of sapwood and heartwood, etc.: https://www.thegrove3d.com/releases/the-grove-release-8/