QUBR_exsitu_gen_diversity

Overview

This repository contains code related to the 2024-2025 RaMP project, which assessed outplanted seedling success in Quercus brandegeei.

Contents

Set Up

Imports data from Daniel’s database (2023 and 2024 tabs) on 1927 adopted seedlings and data from November 2024 field work on 157 outplanted seedlings.

Data Cleaning

Daniel’s database

• The script renames columns to simplified versions in English.
• Calculates a DateDied (conservative, median, and liberal versions) based on different assumptions of how quickly a seedling died after its last positive ID
• Calculates a TimeAlive (conservative, median, and liberal) for each seedling
• Calculates a PotentialTimeAlive of how long each seedling has been planted for
• Identifies an Outcome of whether each seedling is Dead or Alive
• Identifies priority sites with potentially living seedlings to prioritize during November 2024 field work

Nov 2024 field data

• Renames columns to match other databases
• Standardizes character responses and assigns numeric values to continuous variables (height, health, canopy cover)
• Replaces height measurements with a range of heights
• Adds information back into Daniel’s dataset on the individuals we monitored in Nov 2024
• Recalculates TimeAlive (conservative, median, and liberal) and DateDied (conservative, median, and liberal) including information from Monitor4 (Nov 2024)

Survivorship For Loop

• Creates a df that increases by 1 day at a time and identifies how many seedlings were alive that many days after outplanting
• Plots a survivorship curve including all of the seedlings that have been outplanted to date
• We ended up not using the survivorship curve in analysis and replaced it with a waterfall figure

Functionalized For Loop

• This is another way of creating the Survivorship for loop using less lines of code

Visualizing Outliers

Chi-Squared Test

• Is there any significance to whether a seedling was watered after planting on overall survivorship?

Interval Regression

Ordinal Regression

• Analyzed Condition and Canopy variables (with and without dead individuals included)
• Found that the best explanatory model includes Canopy and Ranch

Poster Figures

• Figures that were used in presentations and on conference poster
  • Survivorship curve
  • Waterfall survivorship figure (conservative, median, liberal interpretations of DateDied): preferred alternative to Survivorship curve
  • Comparing the Frequency of each Condition of seedlings faceted by Ranch
  • Comparing the Frequency of each Condition of seedlings faceted by Canopy
  • Stacked bar charts
    • number of individuals in each Canopy class, colored by Region
    • number of individuals in each Canopy class, colored by Ranch & Region
    • number of individuals in each Condition class, colored by Region
    • number of individuals in each Condition class, colored by Ranch & Region

Maps

• Distribution of individuals alive as of November 2024
• Distribution of all individuals historically outplanted