I am analyzing data collected in the field for a habitat selection study of a large mammal. I have ~1300 plots that were sampled. I am trying to find a good way to quantify the diversity of tree heights within each plot. At each plot, we identified the best representative for each of 3 height classes (overstory, intermediate, and understory) and the height of each was recorded. If the stand was clearly unimodal or bimodal and lacked a representative then a NA was recorded for that class.
In this paper by Kuuluvainen et al. the authors quantified tree height diversity using the Shannon-Weaver formula. However, it appears to me that the Shannon-Weaver formula used in this context relies on the proportion of trees in each height class. Since we only measured one tree in each size class, the proportion of trees measured is the same for each class.
Are there any standardized statistical approaches to quantifying these types of variables with the way the data were sampled?
