SapwoodModel

This software package fits a model to sapwood data from scots pine (Pinus sylvestris L.) using heartwood data and tree ring width using a nonlinear model described in Edvardsson et al. (2021). Three models are implemented:

sapwood_fit_plw() - Parabolic-linear model with constant variance and lognormal residuals that uses tree ring width, described as Model 1 in the article

sapwood_fit_pl() - Parabolic-linear model with constant variance and lognormal residuals that does not use tree ring width, described as Model 2 in the article

sapwood_fit_l() - Linear model with constant variance and lognormal residuals, described as Model 3 in the article

Installation

# Install development version from GitHub
devtools::install_github("karirogg/SapwoodModel")

Getting started

Get to know the package by trying to fit your first model to the data. Fetch some data by calling

library(SapwoodModel)
data(smaland)

and run the model with:

model <- sapwood_fit_pl(S~H, smaland)

Note that sapwood_fit_pl can be replaced with sapwood_fit_l or sapwood_fit_plw (see ?sapwood_fit for usage documentation).

From here, you can see some information about the fit by running

summary(model)
plot(model)
predict(model, tibble(H = c(220,300), remaining = c(60,0)))
model$parameter_CI

and more (see ?plot.sapwood_fit, ?predict.sapwood_fit etc.)

References

Edvardsson, J., Rögnvaldsson, K., Helgadóttir, E. Þ., Linderson, H., & Hrafnkelsson, B. (2022). A statistical model for the prediction of the number of sapwood rings in scots pine (pinus sylvestris l.) Dendrochronologia, 74, 125963. https://doi.org/10.1016/j.dendro.2022.125963