I’ve created a regression for converting dbh to height and it’s working pretty well. One issue with it is that when I combine the data for young trees (which were measured in height) and bigger trees (which I converted), there is very little variance among the heights of larger trees; the conversion has served as a smoother of the data. This makes a regression for growth fit much more nicely, but unfortunately also creates an uneven distribution of residuals. I have attached a variance to the conversion to add some variance among its outputs, but there is still a skew in the residuals. I am currently looking into implications this may have.

I have also constructed a functioning kernel for population projection. I just need to figure out how to interpret and graph the data (specifically how to graph surfaces in R). In order to actually project the population I’m also going to need to learn how to integrate in R. This isn’t as simple as just finding an “integrate” function somewhere on the internet. There are several scholarly articles with different ways of summing differently sized intervals, and one specifically I’ll need to look at that deals with integrating the large matrices associated with large organisms (such as chestnut trees).