Wednesday, July 8, 2020
Water

ForestFit: An R package for modeling plant size distributions

0
8


Highlights

Bayesian inference for diameter distributions and height-diameter data.

R package with extensive model options.

Model fit diagnostics.

Accommodates grouped and ungrouped data.

Abstract

Modeling the diameter distribution of trees in forest stands is a frequent application in ecology that supports key biologically and economically relevant management decisions. The choice of model used to represent the diameter distribution and how to estimate its parameters has received much attention in the forestry literature; however, accessible software that facilitates comprehensive comparison of the myriad modeling approaches is not available. To this end, we developed an R package called ForestFit that simplifies estimation of common probability distributions used to model tree diameter distributions, including the two- and three-parameter Weibull distributions, Johnson’s SB distribution, Birnbaum–Saunders distribution, and finite mixture distributions. Frequentist and Bayesian techniques are provided for individual tree diameter data, as well as grouped data. Additional functionality facilitates fitting curves to height-diameter data. The package also provides a set of functions for computing probability distributions and simulating random realizations from common finite mixture models.

Keywords

Bayesian

Forest modeling

Grouped data

Parameter estimation

Mixture models

R package

Statistical distributions

View full text

© 2020 Elsevier Ltd. All rights reserved.



Source link