acd7

Refactoring of the Acadian Variant of FVS

Introduction

The Acadian Variant of FVS ($ACD_0$) officially exists as R¹ code and is maintained by Aaron Weiskittel and Ben Rice (among others).

While $ACD_0$ performs well and is better than FVSne², there are several things of concern:

• Use of a plot-level basal area increment equation to limit basal area growth emerging from tree-level diameter growth estimates;
• Over-estimation of survival for over-stocked stands;
• Use of a cap on height growth to regulate unreasonably high height growth rates;
• Some problematic formulations of component models;
• A number of bugs and inconsistencies in the R code; and
• Slow execution times.

Methodology: Developing $ACD_7$

• Re-coded the model in C++ (to the C++ 23 standard) and built an R package interface;
• Refactoring:
  • Abandoned mixed-model approach and fit species independently
  • Fit equations for:
    • diameter growth;
    • individual tree survival;
    • height growth;
    • height to crown base change;
    • height prediction;
• Developed an algorithm to solve high expansion factor bias;
• Worked with Ben Rice's FIA plot benchmarking data set to validate and test the model.

R Package

$ACD_7$ is accessible through an R package acdR available in this repository as both a Windows binary and Source tarball. The package depends on the following libraries:

• Rcpp (>= 1.0.9)
• RSQLite
• dplyr

Repo Contents

acd7
|
|-- acdR: R package source
|
|-- cl_acd: command line interface source
|
|-- db_acd: SQLite interface source and example files
|
|-- documentation: pdfs of fitting documentation for component equations\

Note: all C++ source code tested with g++ 13.2.0.

Examples for running the R package and the command line interfaces can be found in README.txt.

Repo Development Path

1. Insure cross-platform build integrity
2. Build in runtime optimizations
3. Chasing model residuals

Footnotes

1. R Core Team (2023). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ ↩

2. FVS Staff. 2008 (revised April 16, 2024). Northeast (NE) Variant Overview – Forest Vegetation Simulator. Internal Rep. Fort Collins, CO: U. S. Department of Agriculture, Forest Service, Forest Management Service Center. 56p ↩