{
 "nbformat": 4,
 "nbformat_minor": 0,
 "metadata": {
  "kernelspec": {
   "name": "ir",
   "display_name": "R",
   "language": "R"
  },
  "language_info": {
   "name": "R"
  },
  "colab": {
   "provenance": [],
   "toc_visible": true
  }
 },
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Session 1 · Introduction to R for soil science\n",
    "**Module 1 · Introduction to R, spatial data and soil data preparation** · SoilFER Training\n",
    "\n",
    "[Course page](https://training.yigini.net/modules/01-r-soil-data/) · [Manual chapter](https://training.yigini.net/manual/introduction-to-soil-data-preparation-spatial-data.html) · [Original script](https://github.com/SoilFER/SoilFER-Training-Resources/blob/main/02_scripts/module1/Session1_Introduction_to_R.R)\n",
    "\n",
    "**Before you start**\n",
    "1. Check that the runtime is **R**: *Runtime → Change runtime type → R* (this notebook should open in R automatically).\n",
    "2. Run the **Setup** cell below once per session (≈1–3 minutes). It downloads the training project, installs the R packages and downloads the course rasters and MIR data (≈1.2 GB) and sets the working folder.\n",
    "3. Then run the cells in order with **Shift + Enter**.\n",
    "\n",
    "> Colab resets when you close it or after ~90 minutes without activity. Save results you want to keep with *Files → Download* (left sidebar), or re-run the Setup cell after a reset.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# ==== SoilFER · Colab setup (run first, once per session) ==================\n",
    "options(repos = c(CRAN = \"https://cloud.r-project.org\"), timeout = 3600)\n",
    "# Works in Google Colab and in any other Jupyter (JupyterHub, JupyterLab on your computer)\n",
    "on_colab <- nzchar(Sys.getenv(\"COLAB_RELEASE_TAG\")) || dir.exists(\"/content/sample_data\")\n",
    "root <- if (on_colab) \"/content/SoilFER-Training-Resources\" else path.expand(\"~/SoilFER-Training-Resources\")\n",
    "Sys.setenv(SOILFER_ROOT = root)\n",
    "\n",
    "# 1. Training project (scripts, small data, outputs, assignments)\n",
    "if (!dir.exists(root))\n",
    "  system(paste(\"git clone --depth 1 https://github.com/SoilFER/SoilFER-Training-Resources\", root))\n",
    "\n",
    "# 2. R packages (in Colab the R runtime installs ready-made binaries, so this is fast)\n",
    "pkgs <- c(\"aqp\", \"dplyr\", \"remotes\", \"sf\", \"source\", \"terra\", \"tidyr\", \"tidyverse\", \"win.binary\")\n",
    "need <- setdiff(pkgs, rownames(installed.packages()))\n",
    "if (length(need)) install.packages(need)\n",
    "\n",
    "# 3. Course rasters + MIR spectra (Google Drive folder of the SoilFER training, ≈1.2 GB)\n",
    "td <- file.path(root, \"01_data/module1/training_data\")\n",
    "if (!file.exists(file.path(td, \"MIR_KANSAS_data.xlsx\"))) {\n",
    "  system(\"python3 -m pip -q install gdown\")\n",
    "  drv <- file.path(dirname(root), \"soilfer_drive\")\n",
    "  system(paste(\"python3 -m gdown --folder --quiet https://drive.google.com/drive/folders/1K7tq9zX5HsqbqWcNoT27WtfPtehcKBCu -O\", shQuote(drv)))\n",
    "  f <- list.files(drv, recursive = TRUE, full.names = TRUE)\n",
    "  file.copy(f, td, overwrite = FALSE)\n",
    "}\n",
    "\n",
    "setwd(root)\n",
    "cat(\"Ready. Working folder:\", getwd(), \"\\n\")\n",
    "missing <- setdiff(pkgs, rownames(installed.packages()))\n",
    "if (length(missing)) message(\"Not installed: \", paste(missing, collapse = \", \"))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SoilFER Online Training Programme — Module 1"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "###############################################################################\n",
    "# SoilFER Online Training Programme — Module 1\n",
    "# SESSION 1: Introduction to R for Soil Science (1.5 hours)\n",
    "###############################################################################\n",
    "#\n",
    "# LEARNING OBJECTIVES\n",
    "# -------------------\n",
    "# By the end of this session, participants will be able to:\n",
    "#   1. Navigate the RStudio interface\n",
    "#   2. Install and load R packages\n",
    "#   3. Create and work with R objects and data types\n",
    "#   4. Use vectors, factors, matrices, data frames, and lists\n",
    "#   5. Apply arithmetic, comparison, and logical operators\n",
    "#   6. Write conditional statements and loops\n",
    "#   7. Create and call custom functions\n",
    "#   8. Manipulate data using base R and the tidyverse\n",
    "#   9. Handle missing values\n",
    "#  10. Reshape tabular data between wide and long formats\n",
    "#\n",
    "# DATASET\n",
    "# -------\n",
    "# All examples in this session use small toy datasets constructed inline.\n",
    "# The KSSL Kansas dataset is introduced in Session 2.\n",
    "#\n",
    "# TIMING GUIDE (approximate)\n",
    "# ---------------------------\n",
    "#  0:00 – 0:15  R Packages: install, load, find help\n",
    "#  0:15 – 0:35  Objects, data types, and data structures\n",
    "#  0:35 – 0:55  Operators and control structures\n",
    "#  0:55 – 1:15  Functions and data manipulation\n",
    "#  1:15 – 1:30  Missing values, reshaping, and BDL data\n",
    "###############################################################################\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 1 — PACKAGES"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 1 — PACKAGES\n",
    "# =============================================================================\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 1.1  Installing Packages\n",
    "# -----------------------------------------------------------------------------\n",
    "# Packages need to be installed only ONCE. Use install.packages():\n",
    "\n",
    "# Install a single package\n",
    "install.packages(\"tidyverse\")\n",
    "\n",
    "# Install multiple packages at once\n",
    "install.packages(c(\"terra\", \"sf\", \"aqp\"))\n",
    "\n",
    "# Check which packages are already installed\n",
    "installed.packages()\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 1.2  Installing from GitHub or Other Sources\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# First, install remotes package (if not already installed)\n",
    "install.packages(\"remotes\")\n",
    "\n",
    "# Install an R package from GitHub\n",
    "remotes::install_github(\"rspatial/terra\")\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 1.3  Manual Installation (offline environments)\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Installing from a compressed source package file (e.g., mypackage_1.0.0.tar.gz)\n",
    "install.packages(\"path/to/mypackage_1.0.0.tar.gz\", repos = NULL, type = \"source\")\n",
    "\n",
    "# Installing from a Local .zip file (Windows Binary)\n",
    "install.packages(\"path/to/mypackage.zip\", repos = NULL, type = \"win.binary\")\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 1.4  Loading Packages\n",
    "# -----------------------------------------------------------------------------\n",
    "# After installation, LOAD a package each time you start a new R session:\n",
    "\n",
    "# Load tidyverse package\n",
    "library(tidyverse)\n",
    "\n",
    "# Load multiple packages\n",
    "library(terra)\n",
    "library(sf)\n",
    "library(aqp)\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 1.5  Finding Help on Packages\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Get help on a package\n",
    "help(package = \"tidyverse\")\n",
    "\n",
    "# Or use\n",
    "?tidyverse\n",
    "\n",
    "# View vignettes (tutorials) for a package\n",
    "vignette(package = \"ggplot2\")\n",
    "vignette(\"ggplot2\", package = \"ggplot2\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 2 — OBJECTS AND DATA TYPES"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 2 — OBJECTS AND DATA TYPES\n",
    "# =============================================================================\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 2.1  Creating Objects with Assignment\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Assign a value to a variable\n",
    "soil_depth <- 30\n",
    "soil_depth\n",
    "\n",
    "# Assignment `<-` to objects and `=` to function arguments\n",
    "soil_depth <- c(10,20,30)\n",
    "mean_soil_depth <- mean(x = soil_depth, na.rm = TRUE)\n",
    "mean_soil_depth\n",
    "\n",
    "# Good naming practices — descriptive, consistent\n",
    "# -   Object names must start with a letter\n",
    "# -   Names can contain letters, numbers, underscores `_`, and periods `.`\n",
    "# -   Names are case-sensitive: `SoilDepth` is different from `soildepth`\n",
    "# -   Avoid using reserved words like `TRUE`, `FALSE`, `NA`, `function`, etc.\n",
    "\n",
    "soil_ph <- 6.5\n",
    "organic_carbon_percent <- 2.1\n",
    "clay_content_gkg <- 350\n",
    "\n",
    "# Use consistent style\n",
    "plot_id <- \"P001\"      # snake_case (recommended)\n",
    "plotID <- \"P001\"       # camelCase (alternative)\n",
    "\n",
    "# Bad naming practices — too short, too long, unclear\n",
    "# Avoid — too short, unclear\n",
    "x <- 6.5\n",
    "a <- 2.1\n",
    "\n",
    "# Avoid — too long\n",
    "the_ph_value_of_the_topsoil_at_site_one <- 6.5\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 2.2  Data Types in R \n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Numeric values\n",
    "ph_value <- 6.8\n",
    "clay_percent <- 25.5\n",
    "temperature <- 15.2\n",
    "\n",
    "# Integer values (use L suffix)\n",
    "sample_count <- 100L\n",
    "plot_number <- 5L\n",
    "\n",
    "# Character values (use quotes)\n",
    "soil_type <- \"Acrisol\"\n",
    "location <- \"Kansas\"\n",
    "notes <- \"Sample collected from topsoil\"\n",
    "\n",
    "# Logical values\n",
    "is_valid <- TRUE\n",
    "has_missing_data <- FALSE\n",
    "\n",
    "# Dates: The ISO 8601 date format, YYYY-MM-DD\n",
    "sampling_date <- \"2024-03-12\"\n",
    "# Convert to date (%Y = 4-digit year; %y = 2-digit year)\n",
    "sampling_date <- as.Date(sampling_date, format = \"%Y-%m-%d\") \n",
    "class(sampling_date)\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 2.3  Checking Data Types\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Check the class (high-level type)\n",
    "class(ph_value)   # \"numeric\"\n",
    "class(soil_type)  # \"character\"\n",
    "class(is_valid)   # \"logical\"\n",
    "\n",
    "# Check the internal storage type\n",
    "typeof(sample_count)\n",
    "typeof(ph_value)\n",
    "\n",
    "# Helpful checks\n",
    "is.numeric(ph_value)\n",
    "is.character(soil_type)\n",
    "is.logical(is_valid)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 3 — DATA STRUCTURES"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 3 — DATA STRUCTURES\n",
    "# =============================================================================\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 3.1  Vectors\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "#### Creating Vectors ----\n",
    "\n",
    "# Create a numeric vector using c() (combine)\n",
    "ph_values <- c(5.2, 6.5, 7.1, 5.8, 6.9)\n",
    "print(ph_values)\n",
    "ph_values\n",
    "\n",
    "# Character vector\n",
    "soil_types <- c(\"Acrisol\", \"Ferralsol\", \"Vertisol\", \"Andosol\", \"Cambisol\")\n",
    "soil_types\n",
    "\n",
    "# Logical vector\n",
    "valid_samples <- c(TRUE, TRUE, FALSE, TRUE, TRUE)\n",
    "valid_samples\n",
    "\n",
    "# Create sequences\n",
    "depths <- 0:100                    # Integers from 0 to 100\n",
    "depths_seq <- seq(0, 100, by=10)   # 0, 10, 20, ..., 100\n",
    "\n",
    "#### Vector Operations ----\n",
    "\n",
    "# Arithmetic on vectors (element-wise)\n",
    "ph_values * 10\n",
    "ph_values + 1\n",
    "\n",
    "# Summary statistics\n",
    "mean(ph_values)\n",
    "median(ph_values)\n",
    "sd(ph_values)        # standard deviation\n",
    "min(ph_values)\n",
    "max(ph_values)\n",
    "# NAs affect the statistic results\n",
    "mean(c(ph_values,NA))\n",
    "mean(c(ph_values,NA), na.rm=TRUE) # Remove NAs from calculations\n",
    "\n",
    "#### Accessing Vector Elements ----\n",
    "\n",
    "# Access by index (position)\n",
    "ph_values           # Complete data\n",
    "ph_values[1]        # First element\n",
    "ph_values[3]        # Third element\n",
    "\n",
    "# Access multiple elements\n",
    "ph_values[c(1,3,5)]  # Elements 1, 3, and 5\n",
    "\n",
    "# Access by logical condition\n",
    "ph_values[ph_values > 6]     # All pH values greater than 6\n",
    "\n",
    "# Negative indices exclude elements\n",
    "ph_values[-1]       # All except first\n",
    "ph_values[-c(1,2)]  # All except first two\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 3.2  Factors\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Create a factor from character vector\n",
    "soil_class <- factor(c(\"Clay\", \"Loam\", \"Sand\", \"Clay\", \"Loam\"))\n",
    "soil_class\n",
    "\n",
    "# Check levels\n",
    "levels(soil_class)\n",
    "\n",
    "# Count observations per level\n",
    "table(soil_class)\n",
    "\n",
    "# Ordered factors (when order matters)\n",
    "texture_class <- factor(\n",
    "  c(\"Coarse\", \"Fine\", \"Medium\", \"Fine\", \"Coarse\"),\n",
    "  levels = c(\"Coarse\", \"Medium\", \"Fine\"),\n",
    "  ordered = TRUE\n",
    ")\n",
    "texture_class\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 3.3  Matrices\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Create a matrix\n",
    "soil_matrix <- matrix(\n",
    "  c(5.2, 25, 30,\n",
    "    6.5, 30, 28,\n",
    "    7.1, 18, 35),\n",
    "  nrow = 3,\n",
    "  ncol = 3,\n",
    "  byrow = TRUE\n",
    ")\n",
    "soil_matrix\n",
    "\n",
    "# Add column names\n",
    "colnames(soil_matrix) <- c(\"pH\", \"Clay\", \"Sand\")\n",
    "rownames(soil_matrix) <- c(\"Sample1\", \"Sample2\", \"Sample3\")\n",
    "soil_matrix\n",
    "\n",
    "# Access elements\n",
    "soil_matrix[1, 2]        # Row 1, Column 2\n",
    "soil_matrix[1, ]         # All of row 1\n",
    "soil_matrix[, 2]         # All of column 2\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 3.4  Data Frames\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Create a data frame\n",
    "soil_data <- data.frame(\n",
    "  plot_id = c(\"P001\", \"P002\", \"P003\", \"P004\", \"P005\"),\n",
    "  latitude = c(-1.25, -1.27, -1.23, -1.29, -1.26),\n",
    "  longitude = c(36.85, 36.83, 36.87, 36.81, 36.84),\n",
    "  ph = c(5.2, 6.5, 7.1, 5.8, 6.9),\n",
    "  organic_carbon = c(2.1, 3.2, 1.8, 2.7, 2.9),\n",
    "  clay_content = c(25, 30, 18, 42, 35),\n",
    "  soil_type = c(\"Acrisol\", \"Ferralsol\", \"Vertisol\", \"Andosol\", \"Cambisol\")\n",
    ")\n",
    "\n",
    "# View the data frame\n",
    "soil_data\n",
    "\n",
    "# View structure\n",
    "str(soil_data)\n",
    "\n",
    "# View first rows\n",
    "head(soil_data)\n",
    "\n",
    "# View last rows\n",
    "tail(soil_data)\n",
    "\n",
    "summary(soil_data)\n",
    "\n",
    "\n",
    "# Get dimensions\n",
    "dim(soil_data)         # rows, columns\n",
    "nrow(soil_data)        # number of rows\n",
    "ncol(soil_data)        # number of columns\n",
    "\n",
    "# Access columns and rows\n",
    "soil_data$ph\n",
    "soil_data[[\"ph\"]]\n",
    "\n",
    "# Access rows\n",
    "soil_data[1, ]           # First row\n",
    "soil_data[c(1,3,5), ]    # Rows 1, 3, and 5\n",
    "\n",
    "# Access specific cells\n",
    "soil_data[2, 4]          # Row 2, Column 4 (pH of second plot)\n",
    "\n",
    "# Subset based on conditions\n",
    "soil_data[soil_data$ph > 6, ]              # Plots with pH > 6\n",
    "soil_data[soil_data$soil_type == \"Acrisol\", ]  # Only Acrisols\n",
    "\n",
    "# Add a new column\n",
    "soil_data$silt_content <- c(40, 35, 52, 23, 30)\n",
    "\n",
    "# Calculate new columns from existing ones\n",
    "soil_data$clay_plus_silt <- soil_data$clay_content + soil_data$silt_content\n",
    "\n",
    "# View updated data frame\n",
    "head(soil_data)\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 3.5  Lists\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Create a list\n",
    "soil_analysis <- list(\n",
    "  site_name = \"Kansas Field\",\n",
    "  coordinates = c(lat = -1.25, lon = 36.85),\n",
    "  measurements = data.frame(\n",
    "    depth = c(0, 10, 20, 30),\n",
    "    ph = c(6.5, 6.2, 5.8, 5.5)\n",
    "  ),\n",
    "  notes = \"Collected during dry season\"\n",
    ")\n",
    "\n",
    "# View list structure\n",
    "str(soil_analysis)\n",
    "\n",
    "# Access list elements\n",
    "soil_analysis$site_name\n",
    "soil_analysis[[1]]                # First element\n",
    "soil_analysis[[\"measurements\"]]   # View measurements data frame\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 4 — OPERATORS"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 4 — OPERATORS\n",
    "# =============================================================================\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 4.1  Arithmetic Operators\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Basic arithmetic\n",
    "10 + 5      # Addition\n",
    "10 - 5      # Subtraction\n",
    "10 * 5      # Multiplication\n",
    "10 / 5      # Division\n",
    "10 ^ 2      # Exponentiation (10 squared)\n",
    "10 %% 3     # Modulus (remainder: 10 mod 3 = 1)\n",
    "10 %/% 3    # Integer division (10 divided by 3 = 3)\n",
    "\n",
    "# Order of operations (PEMDAS)\n",
    "result <- (10 + 5) * 2 / 4 - 1\n",
    "result\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 4.2  Comparison Operators\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Comparison operators return TRUE or FALSE\n",
    "5 == 5      # Equal to\n",
    "5 != 3      # Not equal to\n",
    "5 > 3       # Greater than\n",
    "5 < 3       # Less than\n",
    "5 >= 5      # Greater than or equal to\n",
    "5 <= 6      # Less than or equal to\n",
    "\n",
    "# Use in subsetting\n",
    "ph_values <- c(5.2, 6.5, 7.1, 5.8, 6.9)\n",
    "ph_values > 6                     # Logical vector\n",
    "ph_values[ph_values > 6]          # Values greater than 6\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 4.3  Logical Operators\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# AND operator: & (element-wise) or && (single values)\n",
    "TRUE & TRUE      # TRUE\n",
    "TRUE & FALSE     # FALSE\n",
    "\n",
    "# OR operator: | (element-wise) or || (single values)\n",
    "TRUE | FALSE     # TRUE\n",
    "FALSE | FALSE    # FALSE\n",
    "\n",
    "# NOT operator: !\n",
    "!TRUE            # FALSE\n",
    "!FALSE           # TRUE\n",
    "\n",
    "# Combining conditions\n",
    "ph_values <- c(5.2, 6.5, 7.1, 5.8, 6.9)\n",
    "clay_content <- c(25, 30, 18, 42, 35)\n",
    "\n",
    "# Find samples with pH > 6 AND clay > 25\n",
    "ph_values > 6 & clay_content > 25\n",
    "\n",
    "# Find samples with pH > 6 OR clay > 40\n",
    "ph_values > 6 | clay_content > 40\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 5 — CONTROL STRUCTURES"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 5 — CONTROL STRUCTURES\n",
    "# =============================================================================\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 5.1  If-Else Statements\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Basic if statement\n",
    "ph_value <- 7.5\n",
    "\n",
    "# Basic if statement\n",
    "if (ph_value > 7) {\n",
    "  print(\"Alkaline soil\")\n",
    "}\n",
    "\n",
    "# If-else\n",
    "if (ph_value > 7) {\n",
    "  print(\"Alkaline soil\")\n",
    "} else {\n",
    "  print(\"Neutral or acidic soil\")\n",
    "}\n",
    "\n",
    "# Multiple conditions (only the first TRUE branch runs)\n",
    "if (ph_value > 7.5) {\n",
    "  print(\"Strongly alkaline\")\n",
    "} else if (ph_value > 7) {\n",
    "  print(\"Slightly alkaline\")\n",
    "} else if (ph_value == 7) {\n",
    "  print(\"Neutral\")\n",
    "} else {\n",
    "  print(\"Acidic\")\n",
    "}\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 5.2  Vectorized If-Else: ifelse()\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Vectorized conditional assignment\n",
    "ph_values <- c(5.2, 6.5, 7.1, 5.8, 7.0)\n",
    "\n",
    "# Simple two-class example\n",
    "soil_reaction <- ifelse(ph_values > 7, \"Alkaline\", \"Not alkaline\")\n",
    "soil_reaction\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 5.3  Vectorized Cut: cut()\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "soil_class <- cut(\n",
    "  ph_values,\n",
    "  breaks = c(-Inf, 5.5, 7.0, Inf),\n",
    "  labels = c(\"Acidic\", \"Neutral\", \"Alkaline\"),\n",
    "  right = TRUE, include.lowest = TRUE\n",
    ")\n",
    "soil_class\n",
    "\n",
    "# Notes:\n",
    "# - `right = TRUE` means intervals are **right-closed** (e.g., `(5.5, 7.0]`).\n",
    "# - `include.lowest = TRUE` ensures the smallest value is included in the first interval.\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 5.4  For Loops\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Loop through a sequence\n",
    "for (i in 1:5) {\n",
    "  print(paste(\"Iteration:\", i))\n",
    "}\n",
    "\n",
    "# Loop through a vector\n",
    "soil_types <- c(\"Acrisol\", \"Ferralsol\", \"Vertisol\")\n",
    "\n",
    "for (soil in soil_types) {\n",
    "  print(paste(\"Soil type:\", soil))\n",
    "}\n",
    "\n",
    "# Loop with conditional logic\n",
    "ph_values <- c(5.2, 6.5, 7.1, 5.8, 6.9)\n",
    "\n",
    "for (i in 1:length(ph_values)) {\n",
    "  if (ph_values[i] > 6) {\n",
    "    print(paste(\"Sample\", i, \"has pH =\", ph_values[i], \"(Acceptable)\"))\n",
    "  } else {\n",
    "    print(paste(\"Sample\", i, \"has pH =\", ph_values[i], \"(Too acidic)\"))\n",
    "  }\n",
    "}\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 5.5  While Loops\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# While loop continues until condition is FALSE\n",
    "counter <- 1\n",
    "\n",
    "while (counter <= 5) {\n",
    "  print(paste(\"Counter value:\", counter))\n",
    "  counter <- counter + 1   # Increment counter\n",
    "}\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 6 — FUNCTIONS"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 6 — FUNCTIONS\n",
    "# =============================================================================\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 6.1  Using Built-in Functions\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Statistical functions\n",
    "mean(c(5, 10, 15, 20))\n",
    "median(c(5, 10, 15, 20))\n",
    "sd(c(5, 10, 15, 20))\n",
    "sum(c(5, 10, 15, 20))\n",
    "\n",
    "# String functions\n",
    "toupper(\"acrisol\")\n",
    "tolower(\"FERRALSOL\")\n",
    "nchar(\"soil science\")         # Count characters\n",
    "\n",
    "# Math functions\n",
    "sqrt(16)\n",
    "log(10)\n",
    "exp(2)\n",
    "abs(-5)\n",
    "round(pi, 2)\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 6.2  Creating Custom Functions\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Example: Bulk density (mass / volume)\n",
    "calculate_bulk_density <- function(mass, volume) {\n",
    "  if (any(volume <= 0)) stop(\"volume must be > 0\")\n",
    "  mass / volume\n",
    "}\n",
    "\n",
    "calculate_bulk_density(mass = 150, volume = 100)\n",
    "\n",
    "# Function with a default argument\n",
    "classify_soil_ph <- function(ph, threshold = 7) {\n",
    "  if (ph > threshold) {\n",
    "    \"Alkaline\"\n",
    "  } else if (ph == threshold) {\n",
    "    \"Neutral\"\n",
    "  } else {\n",
    "    \"Acidic\"\n",
    "  }\n",
    "}\n",
    "\n",
    "classify_soil_ph(6.5)\n",
    "classify_soil_ph(6.5, threshold = 6)\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 6.3  Function Arguments and Defaults\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# SOC stock calculation (example)\n",
    "# Assumptions:\n",
    "# - soc_percent is in %\n",
    "# - bulk_density is in g/cm^3\n",
    "# - depth is in cm\n",
    "# Output: SOC stock in Mg/ha\n",
    "\n",
    "calculate_soc_stock <- function(soc_percent, bulk_density, depth, coarse_fragment = 0) {\n",
    "  if (any(coarse_fragment < 0 | coarse_fragment > 100)) stop(\"coarse_fragment must be between 0 and 100\")\n",
    "  soc_percent * bulk_density * depth * (1 - coarse_fragment / 100)\n",
    "}\n",
    "\n",
    "calculate_soc_stock(soc_percent = 2.5, bulk_density = 1.3, depth = 30)\n",
    "calculate_soc_stock(soc_percent = 2.5, bulk_density = 1.3, depth = 30, coarse_fragment = 15)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 7 — DATA MANIPULATION WITH BASE R"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 7 — DATA MANIPULATION WITH BASE R\n",
    "# =============================================================================\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 7.1  Subsetting and Filtering\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Example data frame\n",
    "soil_data <- data.frame(\n",
    "  plot_id = c(\"P001\", \"P002\", \"P003\", \"P004\", \"P005\"),\n",
    "  latitude = c(-1.25, -1.27, -1.23, -1.29, -1.26),\n",
    "  longitude = c(36.85, 36.83, 36.87, 36.81, 36.84),\n",
    "  ph = c(5.2, 6.5, 7.1, 5.8, 6.9),\n",
    "  clay_content = c(25, 30, 18, 42, 35),\n",
    "  soil_type = c(\"Acrisol\", \"Ferralsol\", \"Vertisol\", \"Andosol\", \"Cambisol\")\n",
    ")\n",
    "\n",
    "# Select column by index: Show second column\n",
    "soil_data[, 2]\n",
    "# Select column by index: Show first and third columns\n",
    "soil_data[, c(1,3)]\n",
    "# Select column  by name: Show texture columns\n",
    "soil_data[, c(\"ph\",\"clay_content\",\"soil_type\")]\n",
    "# Filter rows by index: keep first 2 records in the data frame\n",
    "soil_data[c(1:2),]\n",
    "\n",
    "# Filter rows by index: keep texture properties for the first 100 records in the data frame\n",
    "soil_data[c(1:2), c(\"ph\",\"clay_content\",\"soil_type\")]\n",
    "\n",
    "# Filter rows based on conditions\n",
    "high_ph <- soil_data[soil_data$ph > 6, ]\n",
    "high_ph\n",
    "\n",
    "# Multiple conditions with & and |\n",
    "high_ph_clay <- soil_data[soil_data$ph > 6 & soil_data$clay_content > 25, ]\n",
    "high_ph_clay\n",
    "\n",
    "# Select columns by index: Show second column\n",
    "subset_data <- soil_data[, 2]\n",
    "# Select several columns by index\n",
    "subset_data <- soil_data[, c(1:3)]\n",
    "\n",
    "# Select by name\n",
    "subset_data <- soil_data[, c(\"plot_id\", \"ph\", \"soil_type\")]\n",
    "subset_data\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 7.2  Deleting Columns and Rows\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Example data frame\n",
    "data <- data.frame(\n",
    "  Plot = c(\"P001\", \"P002\", \"P003\"),\n",
    "  Clay = c(25, 30, 18),\n",
    "  Silt = c(40, 35, 52),\n",
    "  pH   = c(5.2, 6.5, 7.1)\n",
    ")\n",
    "\n",
    "# --- Delete columns ---\n",
    "\n",
    "# Delete one column by name\n",
    "data$Clay <- NULL\n",
    "data$Clay <- c() # Empty vector\n",
    "data\n",
    "\n",
    "# Delete multiple columns by name\n",
    "data[, c(\"Silt\", \"pH\")] <- NULL\n",
    "data\n",
    "\n",
    "# --- Delete rows ---\n",
    "\n",
    "# Delete a row by index (e.g., remove the 2nd row)\n",
    "data <- data[-2, ]\n",
    "data\n",
    "\n",
    "# Delete rows based on a condition (e.g., remove rows with pH < 6)\n",
    "# Rebuild the dataframe\n",
    "data <- data.frame(\n",
    "  Plot = c(\"P001\", \"P002\", \"P003\"),\n",
    "  Clay = c(25, 30, 18),\n",
    "  Silt = c(40, 35, 52),\n",
    "  pH   = c(5.2, 6.5, 7.1)\n",
    ")\n",
    "data <- data[data$pH >= 6, ]\n",
    "data\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 7.3  Sorting Data\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Sort by pH (ascending)\n",
    "soil_data[order(soil_data$ph), ]\n",
    "\n",
    "# Sort by pH (descending)\n",
    "soil_data[order(-soil_data$ph), ]\n",
    "\n",
    "# Sort by multiple columns\n",
    "soil_data[order(soil_data$soil_type, soil_data$ph), ]\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 7.4  Aggregating Data\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Example data frame with repeated soil_type groups\n",
    "soil_data <- data.frame(\n",
    "  plot_id = paste0(\"P\", sprintf(\"%03d\", 1:12)),\n",
    "  latitude = c(-1.25, -1.26, -1.27, -1.28, -1.23, -1.24, -1.22, -1.21, -1.29, -1.30, -1.31, -1.32),\n",
    "  longitude = c(36.85, 36.86, 36.83, 36.84, 36.87, 36.88, 36.82, 36.81, 36.84, 36.83, 36.86, 36.85),\n",
    "  ph = c(5.2, 5.6, 6.5, 6.2, 7.1, 6.9, 5.8, 6.0, 6.9, 6.7, 5.4, 5.1),\n",
    "  clay_content = c(25, 28, 30, 33, 18, 20, 42, 40, 35, 34, 27, 26),\n",
    "  soil_type = c(\n",
    "    \"Acrisol\",\"Acrisol\",\n",
    "    \"Ferralsol\",\"Ferralsol\",\n",
    "    \"Vertisol\",\"Vertisol\",\n",
    "    \"Andosol\",\"Andosol\",\n",
    "    \"Cambisol\",\"Cambisol\",\n",
    "    \"Acrisol\",\"Acrisol\"\n",
    "  )\n",
    ")\n",
    "\n",
    "# Multiple summary values by group (mean and standard deviation of pH by soil_type)\n",
    "aggregate(\n",
    "  ph ~ soil_type,\n",
    "  data = soil_data,\n",
    "  FUN = function(x) c(mean = mean(x), sd = sd(x))\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 8 — DATA MANIPULATION WITH THE TIDYVERSE"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 8 — DATA MANIPULATION WITH THE TIDYVERSE\n",
    "# =============================================================================\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 8.1  Loading the Tidyverse\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Install if not already installed\n",
    "# install.packages(\"tidyverse\")\n",
    "\n",
    "# Load tidyverse\n",
    "library(tidyverse)\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 8.2  Tibbles: Modern Data Frames\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Create tibble directly\n",
    "soil_tbl <- tibble(\n",
    "  plot_id = c(\"P001\", \"P002\", \"P003\"),\n",
    "  ph = c(5.2, 6.5, 7.1),\n",
    "  clay = c(25, 30, 18)\n",
    ")\n",
    "soil_tbl\n",
    "\n",
    "soil_data <- data.frame(\n",
    "  plot_id = c(\"P001\", \"P002\", \"P003\"),\n",
    "  ph = c(5.2, 6.5, 7.1),\n",
    "  clay = c(25, 30, 18)\n",
    ")\n",
    "\n",
    "# Convert data frame to tibble\n",
    "soil_tbl <- as_tibble(soil_data)\n",
    "soil_tbl\n",
    "\n",
    "\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 8.3  The Pipe Operator: %>%\n",
    "# -----------------------------------------------------------------------------\n",
    "soil_data <- tibble(\n",
    "  plot_id = c(\"P001\", \"P002\", \"P003\"),\n",
    "  ph = c(5.2, 6.5, 7.1),\n",
    "  clay = c(25, 30, 18)\n",
    ")\n",
    "\n",
    "# Without pipe (nested functions)\n",
    "round(mean(soil_data$ph), 2)\n",
    "\n",
    "# With pipe (sequential operations)\n",
    "soil_data$ph %>%\n",
    "  mean() %>%\n",
    "  round(2)\n",
    "\n",
    "# More complex example\n",
    "soil_data %>%\n",
    "  filter(ph > 6) %>%\n",
    "  select(plot_id, ph, clay) %>%\n",
    "  arrange(desc(ph))\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 8.4  Data Manipulation with {dplyr}\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Select specific columns\n",
    "soil_data %>%\n",
    "  select(plot_id, ph, clay)\n",
    "\n",
    "# Select range of columns\n",
    "soil_data %>%\n",
    "  select(plot_id:ph)\n",
    "\n",
    "# Remove columns\n",
    "soil_data %>%\n",
    "  select(-plot_id, -ph)\n",
    "\n",
    "# Select columns matching pattern\n",
    "soil_data %>%\n",
    "  select(contains(\"p\"))\n",
    "\n",
    "# Filter rows based on condition\n",
    "soil_data %>%\n",
    "  filter(ph > 6)\n",
    "\n",
    "# Multiple conditions\n",
    "soil_data %>%\n",
    "  filter(ph > 6 & clay > 25)\n",
    "\n",
    "# Filter with OR\n",
    "soil_data %>%\n",
    "  filter(plot_id == \"P001\" | plot_id == \"P002\")\n",
    "\n",
    "# Use %in% for multiple values\n",
    "soil_data %>%\n",
    "  filter(plot_id %in% c(\"P001\", \"P002\"))\n",
    "\n",
    "# Rename columns for clarity\n",
    "rename(soil_data,\n",
    "    site_id = plot_id)\n",
    "\n",
    "# You can rename multiple columns at once\n",
    "soil_data %>%\n",
    " rename(\n",
    "  location = plot_id,\n",
    "  acidity = ph,\n",
    "  clay_percent = clay\n",
    " ) \n",
    "\n",
    "# Create new columns\n",
    "soil_data %>%\n",
    "  mutate(\n",
    "    ph_class = ifelse(ph > 7, \"Alkaline\", \"Acidic\"),\n",
    "    clay_gr_kg = clay * 10 \n",
    "  )\n",
    "\n",
    "# Modify existing columns\n",
    "soil_data %>%\n",
    "  mutate(\n",
    "    ph = round(ph, 1),\n",
    "    plot_id = tolower(plot_id)\n",
    "  )\n",
    "\n",
    "# Sort ascending\n",
    "soil_data %>%\n",
    "  arrange(ph)\n",
    "\n",
    "# Sort descending\n",
    "soil_data %>%\n",
    "  arrange(desc(ph))\n",
    "\n",
    "# Multiple sort keys\n",
    "soil_data %>%\n",
    "  arrange(ph, desc(clay))\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 8.5  Grouping and Summarizing Data\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Create example dataset\n",
    "soil_data <- data.frame(\n",
    "site = c(\"Forest_A\", \"Forest_B\", \"Grassland_A\", \"Grassland_B\", \"Urban_A\", \"Urban_B\"),\n",
    "ecosystem = c(\"Forest\", \"Forest\", \"Grassland\", \"Grassland\", \"Urban\", \"Urban\"),\n",
    "pH = c(6.2, 6.8, 7.1, 6.9, 5.8, 6.0),\n",
    "organic_carbon = c(3.2, 2.8, 2.1, 2.4, 1.5, 1.8)\n",
    ")\n",
    "\n",
    "# Summarize soil properties\n",
    "soil_data %>%\n",
    "  summarize(\n",
    "    mean_ph = mean(pH),\n",
    "    sd_ph = sd(pH),\n",
    "    min_soc = min(organic_carbon),\n",
    "    max_soc = max(organic_carbon),\n",
    "    n_samples = n()\n",
    "  )\n",
    "\n",
    "# Group by and summarize\n",
    "soil_data %>%\n",
    "  group_by(ecosystem) %>%\n",
    "  summarize(\n",
    "    mean_ph = mean(pH),\n",
    "    mean_soc = mean(organic_carbon), # Average organic carbon\n",
    "    count = n(),  \n",
    "  .groups = \"drop\"                   # Remove grouping\n",
    ")\n",
    "\n",
    "# Alternative: count()\n",
    "soil_data <- data.frame(\n",
    "site = c(\"Forest_A\", \"Forest_B\", \"Grassland_A\", \"Grassland_B\", \"Urban_A\", \"Urban_B\"),\n",
    "ecosystem = c(\"Forest\", \"Forest\", \"Grassland\", \"Grassland\", \"Urban\", \"Urban\"),\n",
    "pH = c(6.2, 6.8, 7.1, 6.9, 5.8, 6.0),\n",
    "organic_carbon = c(3.2, 2.8, 2.1, 2.4, 1.5, 1.8)\n",
    ")\n",
    "\n",
    "# Count observations by group\n",
    "count(soil_data, ecosystem)\n",
    "\n",
    "# Count by multiple groups\n",
    "count(soil_data, ecosystem, site)\n",
    "\n",
    "# Count with weights of a column (sum of carbon instead of count)\n",
    "soil_data %>%\n",
    " count(ecosystem, wt = organic_carbon, name = \"total SOC\")\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 8.6  Combining Data Frames (Joins)\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Soil data example\n",
    "soil_basic <- data.frame(\n",
    " site_id = c(\"A\", \"B\", \"C\", \"D\"),\n",
    " pH = c(6.2, 6.8, 7.1, 6.9),\n",
    " organic_carbon = c(3.2, 2.8, 2.1, 2.4)\n",
    ")\n",
    "# Additional measurements (note: includes data on site E, but not on site A)\n",
    "nutrients <- data.frame(\n",
    " site_id = c(\"B\", \"C\", \"D\", \"E\"),\n",
    " phosphorus = c(0.15, 0.12, 0.18, 0.14),\n",
    " potassium = c(0.8, 0.9, 0.7, 0.6)\n",
    ")\n",
    "# Site information\n",
    "site_info <- data.frame(\n",
    " site_id = c(\"A\", \"B\", \"C\", \"D\"),\n",
    " ecosystem = c(\"Forest\", \"Forest\", \"Grassland\", \"Grassland\"),\n",
    " elevation = c(450, 520, 380, 420)\n",
    ")\n",
    "\n",
    "# Join examples (same keys, different rules for which rows are kept)\n",
    "\n",
    "# 1) `left_join()`: Keep all rows from soil_basic\n",
    "left_join(soil_basic, nutrients, by = \"site_id\")\n",
    "\n",
    "# 2) `right_join()`: Keep all rows from nutrients\n",
    "right_join(soil_basic, nutrients, by = \"site_id\")\n",
    "\n",
    "# 3) `inner_join()`: Keep only rows that exist in both tables\n",
    "inner_join(soil_basic, nutrients, by = \"site_id\")\n",
    "\n",
    "# 4) `full_join()`: Keep all rows from both tables (missing values become NA)\n",
    "full_join(soil_basic, nutrients, by = \"site_id\")\n",
    "\n",
    "## Joining more than two tables sequentially\n",
    "soil_basic %>%\n",
    "  left_join(nutrients, by = \"site_id\") %>%\n",
    "  left_join(site_info, by = \"site_id\")\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 8.7  Stacking Data with bind_rows()\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Data from different time periods\n",
    "spring_data <- data.frame(\n",
    " site = c(\"A\", \"B\"),\n",
    " season = \"Spring\",\n",
    " pH = c(6.1, 6.7),\n",
    " temperature = c(12.5, 11.8)\n",
    ")\n",
    "\n",
    "summer_data <- data.frame(\n",
    " site = c(\"A\", \"B\"),\n",
    " season = \"Summer\", \n",
    " pH = c(6.3, 6.9),\n",
    " temperature = c(18.2, 17.5)\n",
    ")\n",
    "\n",
    "# Combine the datasets\n",
    "bind_rows(spring_data, summer_data)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 9 — MISSING DATA"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 9 — MISSING DATA\n",
    "# =============================================================================\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 9.1  Identifying Missing Data\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Create data with missing values\n",
    "soil_data_na <- data.frame(\n",
    "  plot = c(\"P001\", \"P002\", \"P003\", \"P004\"),\n",
    "  ph = c(5.2, NA, 7.1, 6.5),\n",
    "  clay = c(25, 30, NA, 35)\n",
    ")\n",
    "\n",
    "# Check for missing values\n",
    "is.na(soil_data_na)\n",
    "\n",
    "# Count missing values per column\n",
    "colSums(is.na(soil_data_na))\n",
    "\n",
    "# Identify complete cases (rows with no missing values)\n",
    "complete.cases(soil_data_na)\n",
    "\n",
    "# Extract complete cases\n",
    "soil_complete <- soil_data_na[complete.cases(soil_data_na), ]\n",
    "soil_complete\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 9.2  Handling Missing Data (Base R)\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Remove rows with any missing values (base R)\n",
    "na.omit(soil_data_na)\n",
    "\n",
    "# Remove rows where a specific column is missing (keep rows with non-missing pH)\n",
    "soil_data_na[!is.na(soil_data_na$ph), ]\n",
    "\n",
    "# Replace missing pH values with the mean pH (na.rm = TRUE ignores NA in the mean)\n",
    "soil_data_na$ph[is.na(soil_data_na$ph)] <- mean(soil_data_na$ph, na.rm = TRUE)\n",
    "soil_data_na\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 9.3  Handling Missing Data (Tidyverse)\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Remove all rows with any NA\n",
    "soil_data_na %>%\n",
    "  drop_na()\n",
    "\n",
    "# Remove rows where ph is NA\n",
    "soil_data_na %>%\n",
    "  drop_na(ph)\n",
    "\n",
    "# Replace NA values with specified values\n",
    "soil_data_na %>%\n",
    "  replace_na(list(ph = 6.0, clay = 30))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 10 — DATA RESHAPING WITH {tidyr}"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 10 — DATA RESHAPING WITH {tidyr}\n",
    "# =============================================================================\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 10.1  pivot_longer(): Wide to Long\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "library(dplyr) # for piping %>%\n",
    "library(tidyr) # for pivoting\n",
    "# Wide format data\n",
    "wide_soil <- data.frame(\n",
    " site = c(\"A\", \"B\", \"C\"),\n",
    " ecosystem = c(\"Forest\", \"Grassland\", \"Urban\"),\n",
    " pH = c(6.2, 7.1, 5.8),\n",
    " carbon = c(3.2, 2.1, 1.5),\n",
    " nitrogen = c(0.25, 0.18, 0.12)\n",
    ")\n",
    "wide_soil\n",
    "\n",
    "# Convert to long format\n",
    "long_soil <- wide_soil %>%\n",
    " pivot_longer(\n",
    "  cols = c(pH, carbon, nitrogen),     # Columns to pivot\n",
    "  names_to = \"measurement_type\",      # Name for the variable column\n",
    "  values_to = \"value\"           # Name for the values column\n",
    " )\n",
    "long_soil\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 10.2  pivot_wider(): Long to Wide\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# Convert back to wide format\n",
    "long_soil %>%\n",
    " pivot_wider(\n",
    "  names_from = measurement_type, # Column containing variable names\n",
    "  values_from = value            # Column containing values\n",
    " )\n",
    "\n",
    "# -----------------------------------------------------------------------------\n",
    "# 10.3  Advanced Pivoting: Multiple Measurements per Site (Replicates)\n",
    "# -----------------------------------------------------------------------------\n",
    "\n",
    "# More complex examples with multiple measurements per site\n",
    "field_data <- data.frame(\n",
    " site = rep(c(\"Forest\", \"Grassland\"), each = 6),\n",
    " measurement = rep(c(\"pH\", \"carbon\", \"nitrogen\"), 4),\n",
    " replicate = rep(c(\"R1\", \"R2\"), 6),\n",
    " value = c(6.2, 6.1, 3.2, 3.0, 0.25, 0.23, 7.1, 7.0, 2.1, 2.3, 0.18, 0.19)\n",
    ")\n",
    "field_data\n",
    "\n",
    "# Pivot to have measurements as columns\n",
    "print(\"Pivoted data:\")\n",
    "field_data %>%\n",
    " pivot_wider(\n",
    "  names_from = measurement,\n",
    "  values_from = value\n",
    " )\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PART 11 — HANDLING BELOW DETECTION LIMIT (BDL) DATA"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "execution_count": null,
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# PART 11 — HANDLING BELOW DETECTION LIMIT (BDL) DATA\n",
    "# =============================================================================\n",
    "\n",
    "soil_lab <- data.frame(\n",
    "  plot_id = c(\"P001\", \"P002\", \"P003\", \"P004\"),\n",
    "  no3_mgkg_raw = c(\"0.12\", \"<0.05\", \"0.31\", \"<0.05\"),\n",
    "  stringsAsFactors = FALSE\n",
    ")\n",
    "soil_lab\n",
    "\n",
    "soil_lab <- soil_lab %>%\n",
    "  mutate(\n",
    "    censored = str_detect(no3_mgkg_raw, \"^\\\\s*<\"),\n",
    "    dl = if_else(censored,\n",
    "                 as.numeric(str_remove(no3_mgkg_raw, \"^\\\\s*<\\\\s*\")),\n",
    "                 NA_real_),\n",
    "    no3_mgkg = if_else(censored, dl / 2, as.numeric(no3_mgkg_raw))\n",
    "  )\n",
    "\n",
    "soil_lab\n",
    "\n",
    "\n",
    "###############################################################################\n",
    "# END OF SESSION 1\n",
    "# Next session: Session 2 — Soil Data Preparation with KSSL Dataset (Part 1)\n",
    "###############################################################################\n"
   ]
  }
 ]
}