Overview | Method 1: SHAP (SHapley Additive exPlanations) | Mathematical definition | Key properties | Computational implementation | Method 2: LIME | Implementation | Selecting kernel_width | Method 3: Partial Dependence Plots (PDP) | Limitation: correlated predictors | Method 4: Accumulated Local Effects (ALE) | When to use ALE vs PDP | Method 5: Permutation variable importance | Definition | Choosing the right XAI method | Reporting checklist for book chapters | References

Introduction | Background | The xaiHydro package | Scope of this vignette | Installation | Simulated datasets | Daily streamflow dataset | Monthly drought index dataset | Model training | Random Forest — streamflow prediction | XGBoost — drought index prediction | The hydro explainer | Creating an explainer | Explainer for XGBoost (custom predict wrapper) | SHAP Analysis | Computing SHAP values | Global SHAP summary plot | Single-observation waterfall | LIME Explanation | Partial Dependence Profiles | PDP — marginal effects | ALE — preferred for correlated predictors | Variable Importance | Prediction Breakdown | Residual Diagnostics | Full automated XAI report | Complete drought index case study | Interpreting XAI results in a hydro-climate context | Physical interpretation of SHAP | ALE vs PDP selection guide | LIME local R² threshold | Session information | References

Overview | Method 1: SHAP (SHapley Additive exPlanations) | Mathematical definition | Key properties | Computational implementation | Method 2: LIME | Implementation | Selecting kernel_width | Method 3: Partial Dependence Plots (PDP) | Limitation: correlated predictors | Method 4: Accumulated Local Effects (ALE) | When to use ALE vs PDP | Method 5: Permutation variable importance | Definition | Choosing the right XAI method | Reporting checklist for book chapters | References

Introduction | Background | The xaiHydro package | Scope of this vignette | Installation | Simulated datasets | Daily streamflow dataset | Monthly drought index dataset | Model training | Random Forest — streamflow prediction | XGBoost — drought index prediction | The hydro explainer | Creating an explainer | Explainer for XGBoost (custom predict wrapper) | SHAP Analysis | Computing SHAP values | Global SHAP summary plot | Single-observation waterfall | LIME Explanation | Partial Dependence Profiles | PDP — marginal effects | ALE — preferred for correlated predictors | Variable Importance | Prediction Breakdown | Residual Diagnostics | Full automated XAI report | Complete drought index case study | Interpreting XAI results in a hydro-climate context | Physical interpretation of SHAP | ALE vs PDP selection guide | LIME local R² threshold | Session information | References

Overview | Installation | Quick start | Offline imputation: choosing a method | LLM-mode imputation | Domain-specific imputation | Choosing a model | Inspecting the audit trail | Large datasets | Tips for best results

Introduction | Package structure | Installation | Temporal Analysis | Mann-Kendall Trend Test — mk_test() | Sen's Slope Estimator — sens_slope() | Change-Point Detection — change_point_detection() | Seasonal Decomposition — seasonal_decompose_climate() | Rolling Trend Analysis — rolling_trend() | SNHT Homogeneity Test — temporal_homogeneity() | Multiple-Station Trend Significance — trend_significance() | Autocorrelation Analysis — autocorrelation_climate() | Spatial Analysis | Moran's I — morans_i() | Hot-Spot and Cold-Spot Detection — hot_cold_spots() | Spatial Interpolation — spatial_interpolate() | Spatial Trend Field — spatial_trend_field() | Climate Zone Classification — cluster_climate_zones() | Elevation Lapse Rate — elevation_lapse_rate() | Extreme Event Analysis | GEV Distribution — fit_gev() and rgev_sim() | Return Levels — return_period() | Peaks-Over-Threshold — peaks_over_threshold() | Heat Wave Detection — heat_wave_detection() | Cold Spell Detection — cold_spell_detection() | Drought Spell Detection — drought_spell() | Hill Tail-Index Estimator — extreme_value_index() | Climate Indices | Standardised Precipitation Index — spi() | SPEI — spei() | Simplified PDSI — pdsi_simple() | Heat Index — heat_index() | Wind Chill — wind_chill() | Frost Days — frost_days() | Growing Degree Days — growing_degree_days() | Diurnal Temperature Range — diurnal_temp_range() | Detection and Attribution | Signal Detection — detection_attribution() | EOF Fingerprint Analysis — fingerprint_analysis() | Optimal Fingerprint Regression — optimal_fingerprint() | Data Quality and Utilities | Gap Filling — fill_gaps_climate() | Homogenisation — homogenize_series() | Temporal Aggregation — aggregate_climate() | Anomaly Calculation — anomaly_baseline() | Standardisation — standardize_climate() | Comprehensive Summary — climate_summary() | References

Overview | 1. Loading Data from CSV | 2. Combining Trend Detection and Uncertainty | 3. Multi-Variable Response Surface | 4. Uncertainty-Adjusted Forecasting | 5. Custom stat_fn Closures | Example A — R-squared of an OLS model | Example B — Optimum N from a quadratic response curve | Example C — Mann-Kendall tau for soil OC | 6. Multi-Metric Model Comparison | 7. Complete Reproducible Pipeline | Session Info

Introduction | Installation | Example Data | Module 1 — Trend Analysis (trendML) | Mann-Kendall test and Sen's slope | Visualise the trend | Individual components | Module 2 — Multivariate System Modelling (multiSysML) | Scaled OLS regression | Partial Least Squares (PLS) | Module 3 — Response Curve & Optimisation (responseML) | Economic optimum | Module 4 — Time Series Forecasting (tsML) | Module 5 — Panel Data & Treatment Effects (panelML) | Module 6 — Uncertainty & Sensitivity Analysis (uncertaintyML) | Module 7 — AutoML & Model Benchmarking (autoML) | Benchmarking on a hold-out set | Recommended Workflow | Session Info

Overview | 1. Train a model | 2. Predict | 3. Evaluate | 4. Cross-validation | 5. Tune hyperparameters | 6. Compare algorithms | 7. Feature importance | 8. Partial dependence | 9. Confusion matrix | 10. Regression example | Citation | Session info

Introduction | Supported Providers | Quick Start | Step 1 — Configure the provider | Step 2 — Inspect the built-in messy datasets | Stage 1 — Detect Issues | Issue type breakdown | Case inconsistencies | Typos and near-duplicates (Levenshtein) | Malformed email addresses | Numeric outliers (Tukey outer fence) | Stage 2 — Suggest Fixes | Stage 3 — Apply Fixes | Dry run mode | Stage 4 — Offline Detection (No API Key) | Stage 5 — Summary Report | Complete Workflow (One Pipeline) | API Key Storage Best Practice | Session Information | References

1. Introduction | 2. Data Generation | 3. Probability Sampling Designs | 3.1 Simple Random Sampling | 3.2 Stratified Sampling | 3.3 Cluster Sampling | 3.4 Probability Proportional to Size (PPS) | 4. Non-Probability Sampling | 5. Estimation Methods | 5.1 Mean and Total Estimation | 5.2 Ratio and Regression Estimators | 5.3 Horvitz–Thompson Estimator | 5.4 Stratified Estimator | 6. NRM-Specific Applications | 6.1 Biomass and Carbon Estimation | 6.2 Soil Loss Estimation | 7. Sampling Efficiency | 8. Spatial Sampling (Optional) | 9. Recommended Workflow | 10. Conclusion | References

Introduction | Key Concepts | Installation | The Example Dataset | Step 1: Validate Your Data | Step 2: Define Variable Configuration | Verify scoring curves | Step 3: Score All Variables | Step 4: Select the Minimum Data Set (MDS) | Step 5: Compute SQI Using All Six Methods | Linear Scoring | Regression-Based | PCA-Based | Fuzzy Logic | Entropy Weighting | TOPSIS | Compare All Methods | Step 6: Visualisation | Score Heatmap | SQI Bar Chart | Radar Profile | PCA Biplot | Step 7: Statistical Analysis | ANOVA | Sensitivity Analysis | Sensitivity Tornado Chart | Recommended Workflow Diagram | References

Introduction | Financial appraisal | Benefit-cost ratio | Net present value | Internal rate of return | Payback period | Marginal rate of return (CIMMYT method) | Modified BCR | Soil loss economic valuation | USLE-based soil loss cost | Nutrient replacement cost | Water resource valuation | Social indicators | Risk analysis | Sensitivity analysis | Switching value | Monte Carlo simulation | Benchmark datasets | State-wise BCR benchmarks | USLE parameters for Indian soils | Rainfall erosivity | SWC unit cost norms (PMKSY-WDC 2015) | Full pipeline and report | References

Multi-environment mixed model | Fixed genotype, random environment | Post-hoc comparisons with emmeans | Fitting multiple growth curves by group | Overlay curves on a single plot | Comparing dose-response curves across species | Exporting a comparison table to Word / Excel

Overview | 1 · Data preparation | Cleaning and outlier detection | Normalisation | Advanced outlier flagging | 2 · Linear models | Simple OLS | Diagnostic plots | Full residual panel | Mixed-effects model | Polynomial selection | 3 · Nonlinear models | Logistic growth | Gompertz curve | Asymptotic (Mitscherlich) model | Quadratic and optimum dose | Linear-plateau model | 4 · Dose-response models | Fit a 4-parameter log-logistic curve | Effective dose estimates | 5 · Model comparison | 6 · Extracting results for reports | Saving plots | Session information

Overview | 1. Built-in datasets | 2. Data validation | 3. Tillage analysis | 3.1 Summarise tillage operations | 3.2 Tillage depth trend | 3.3 Compaction detection | 4. Erosion analysis | 4.1 Track erosion depth | 4.2 Estimate soil loss | 4.3 Compare fields | 5. Visualisation | 5.1 Tillage depth timeline | 5.2 Erosion depth trend | 5.3 Organic matter trend | 5.4 Tillage vs erosion comparison | 6. Full analysis pipeline | References

Introduction | What is I30? | What is EI30? | Basic Usage | Loading the Package | Example Data | Calculate I30 | Calculate EI30 | Working with Different Kinetic Energy Equations | Processing Multiple Storm Events | Data Validation | Visualization | Rainfall Pattern | Intensity Profile | Advanced Usage | Using Explicit Time Intervals | Custom Event Identification | Typical Workflow | References

Key Terms and Definitions | Rainfall Intensity (I) | Maximum 30-minute Intensity (I30) | Kinetic Energy (E) | Erosivity Index (EI30) | Erosivity | Breakpoint Rainfall Data | USLE — Universal Soil Loss Equation | RUSLE — Revised Universal Soil Loss Equation | Hydrological Analysis | References