Understanding the RAPID2 Interface¶

RAPID2 provides a unified set of Python functions and a corresponding Command Line Interface (CLI) dispatcher to streamline routing workflows.

This note was written for rapid2 2.0.0b5.

Command Reference¶

The table below maps each terminal command to its corresponding Python function and describes its primary purpose. Tools marked as Planned are under development and are not yet available.

For conciseness, we useimport rapid2 as r2 here.

Terminal Command	Python Import	Purpose
`rapid2 core run`	`r2.core.run`	Run the core model.
`rapid2 core ltir`	`r2.core.ltir`	Apply inflow bias correction.
`rapid2 pull sandbox`	`r2.pull.sandbox`	Download Sandbox data.
`rapid2 pull tutorial`	Planned	Download Mississippi tutorial.
`rapid2 pull gldas2`	`r2.pull.gldas2`	Download raw GLDAS2 data.
`rapid2 prep gldas2`	`r2.prep.gldas2`	Reformat GLDAS2 time & units.
`rapid2 prep couple`	`r2.prep.couple`	Map LSM grids to networks.
`rapid2 prep sandbox`	`r2.prep.sandbox`	Generate synthetic inflow.
`rapid2 prep coldinit`	`r2.prep.coldinit`	Generate cold-start states.
`rapid2 prep warminit`	Planned	Generate warm-start states.
`rapid2 prep sample`	`r2.prep.sample`	Align NetCDF data to gauges.
`rapid2 prep ltir`	`r2.prep.ltir`	Compute LTIR scalars.
`rapid2 eval graph`	`r2.eval.graph`	Generate SVG hydrographs.
`rapid2 eval cmpncf`	`r2.eval.cmpncf`	Compare NetCDF files.
`rapid2 eval cmppqt`	`r2.eval.cmppqt`	Compare Parquet files.
`rapid2 v1v2 static`	`r2.v1v2.static`	Upgrade RAPID1 CSV to Parquet.
`rapid2 v1v2 inflow`	`r2.v1v2.inflow`	Convert volumes to flow rates.
`rapid2 v1v2 namelist`	Planned	Fortran namelist to YAML.

This maps to the directory structure in the code repository:

src/rapid2/
├── base/        # Low-level mathematical and I/O building blocks
├── pull/        # Data download
├── prep/        # Data preparation
├── core/        # Routing, bias correction, and data assimilation
├── eval/        # Analysis and plotting
├── v1v2/        # Migration from legacy RAPID1 to RAPID2
└── cmds/        # Command Line Interface (CLI)

Core Workflow Concepts¶

RAPID2 organizes functionality into command categories available through both the Python API and the CLI:

pull commands acquire source or synthetic datasets.
prep commands transform, align, and configure input data for routing, including grid coupling and initial-state generation.
core commands perform routing and bias correction.
eval commands support comparison, visualization, and other analysis of model outputs.
v1v2 commands migrate legacy RAPID1 datasets and configuration files to RAPID2 formats.

In Python, these capabilities are exposed through the rapid2 namespace, such as r2.prep.couple(). The CLI provides corresponding dispatcher commands, such as rapid2 prep couple. The CLI is intended to be a thin interface to the same underlying Python functionality.