Tree Survival Prediction Based on Arbuscular Mycorrhizal Fungi

Overview

This project analyzes how tree species and microbial communities affect the growth of arbuscular mycorrhizal fungi (AMF), a key indicator of tree health and survival. The analysis is based on a controlled greenhouse experiment using a two-way factorial design.

Research Question

Do tree species, microbial community type, and their interaction significantly influence AMF colonization?

Data

• 3 tree species
• 4 microbial treatments
• 480 total observations (40 replicates per treatment)
• Response variable: Percent AMF colonization measured at 9 weeks

Raw data are provided in data/raw/.
A cleaned analysis dataset is generated by the scripts and saved to data/processed/.

Methods

• Exploratory data analysis (boxplots, histograms)
• Two-way ANOVA with interaction
• Model diagnostics (standardized residuals, residuals vs fitted, Q–Q plot, residual histogram)
• Tukey-adjusted post hoc comparisons
• Post-hoc power analysis using ANOVA effect sizes

Key Findings

• Microbial community type had a strong and statistically significant effect on AMF growth
• Tree species and species–microbe interactions were also significant
• Combined and control microbes produced the highest AMF colonization
• Post-hoc power was high for microbial treatments but low for species, indicating potential Type II error for species effects

Reproduce Results

1. Clone or download this repository.
2. Open the project by double-clicking the .Rproj file (recommended).
3. Run the following in R:

source("scripts/00_run_all.R")