forestecology

forestecology implements neighborhood competition models to quantify effects of interspecific competition on tree growth in spatially explicit forest data.

Key Features:

• Model Specification: Supports specification of neighborhood competition models to represent spatial interactions among trees and interspecific competition effects on growth.
• Permutation Tests: Implements permutation tests to evaluate the effect of competitor species identity on tree growth.
• Spatial Cross-Validation: Provides spatial cross-validation methods to account for spatial autocorrelation when assessing model performance.
• Bayesian Linear Regression Model: Implements a Bayesian linear regression neighborhood competition model following Allen and Kim (PLoS One, 2020), including incorporation of prior information and estimation of parameter uncertainty.
• Tidyverse-like Structure: Exposes verb-named, pipeable functions in a tidyverse-like structure for modular analysis workflows.
• Standardized Inputs/Outputs: Uses simple features (sf) class for standardized geospatial inputs and outputs.
• S3 Object-Oriented Implementation: Implements models and results using R's S3 object system.

Scientific Applications:

• Forest dynamics and interspecific competition: Quantifies how interspecific competition influences individual tree growth and community structure.
• ForestGEO and Smithsonian plot analyses: Applies neighborhood competition models to the Smithsonian Conservation Biology Institute's large forest dynamics plot within the ForestGEO network and to other ForestGEO plots collected under similar protocols.

Methodology:

Fitting and assessment of neighborhood competition models; permutation tests for competitor-species effects; spatial cross-validation to account for spatial autocorrelation; a Bayesian linear regression neighborhood competition model following Allen and Kim (PLoS One, 2020); standardized sf inputs/outputs; implementation via R's S3 system.