Introduction

This package provides tools for stationary and nonstationary flood frequency analysis (FFA) of annual maximum series (AMS) data. The package supports flexible workflows with the following core components:

1. Explore AMS data for approach selection (S-FFA or NS-FFA) using the eda_* methods:
   • Detect change points.
   • Identify trends in the mean and variance.
2. Select a suitable probability model using the select_* methods.
3. Fit model parameters using the fit_* methods.
4. Quantify uncertainty using the uncertainty_* methods.
5. Evaluate model performance using model_diagnostics().

Additional utility functions support visualization and computation:

• plot_* methods produce diagnostic and summary plots.
• ams_* methods transform or segment AMS time series.
• quantile_*, loglik_*, and lmom_* implement distribution-specific computation.

This package assumes familiarity with statistical techniques used in FFA, including L-moments, maximum likelihood estimation, and parametric bootstrap. For an explanation of these methods, see the FFA Framework wiki. For examples, see the accompanying vignettes on exploratory data analysis and flood frequency analysis.