rivia 0.3.0

A modern Python interface for HEC-RAS Visualization, Information, and Automation

RIVIA

RAS Interface for Visualization, Information, and Automation

A modern, modular Python library for interacting with HEC-RAS hydraulic modeling software.

Documentation

Full documentation is available at rivia.readthedocs.io.

Overview

rivia provides a clean, Pythonic interface for working with HEC-RAS projects:

• Control HEC-RAS via COM automation — open projects, switch plans, run simulations
• Read and write HEC-RAS text input files (.prj, .g*, .p*, .f*, .u*)
• Access HDF5 results — water surface, velocity, depth, and other outputs
• Export rasters — pixel-perfect RASMapper-equivalent rasters (WSE, depth, velocity, etc.)
• Export terrain — mosaic and modify terrain from HEC-RAS terrain HDF files

Requirements

• Windows — HEC-RAS is Windows-only
• Python 3.10+
• HEC-RAS 5.x or later installed

Installation

pip install rivia

With geospatial extras (required for raster export):

pip install rivia[geo]

Quick Example

import numpy as np
from rivia.model import Project

# Open a HEC-RAS project
model = Project("path/to/project.prj")
print(model.version)       # e.g. "6.30"

# Switch plans and run
model.set_plan(short_id="BC")
model.run(hide_window=False)

# Export a WSE raster (pixel-perfect RASMapper equivalent)
vrt = model.export_wse(timestep=None, render_mode="sloping")

hdf = model.results          # UnsteadyPlan

# 2D flow areas
area = hdf.flow_areas["Perimeter 1"]
area.max_water_surface       # pd.DataFrame  — max WSE per cell, columns [value, time]
area.get_depth(timestep=0)   # np.ndarray   — water depth snapshot

# 1D cross sections — mapping interval
xs = hdf.cross_sections("mapping")["Butte Cr Upper 7"]
xs.wse              # pd.Series  — WSE indexed by pd.DatetimeIndex
xs.flow             # pd.Series
xs.velocity_channel # pd.Series

# 1D cross sections — Post Process Profiles (rich hydraulic output)
coll = hdf.cross_sections("post_process")
coll.wse            # pd.DataFrame  — location × ["max_wse", 0, 1, …] profiles
coll.velocity_channel
coll.profile_table("EG Slope")   # any variable by name

xs_pp = coll[0]     # integer index, name "Butte Cr Upper 7", or (river, reach, rs) tuple — all equivalent
xs_pp.wse           # pd.Series  — timeseries only, Max WS excluded
xs_pp.energy_grade  # pd.Series

# Storage areas — DSS hydrograph interval
sa = hdf.storage_areas("output")["Reservoir 1"]
sa.wse              # pd.Series  — WSE indexed by pd.DatetimeIndex

# max_wse / min_wse are only available for output="mapping" (Base Output summary)
sa_map = hdf.storage_areas("mapping")["Reservoir 1"]
sa_map.max_wse      # pd.DataFrame  — columns [value, time]

# Structures — DSS Profile interval
structs = hdf.structures("profile")
inl = structs.inlines["Butte Cr Upper 100"]
inl.stage_hw        # pd.Series  — headwater stage
inl.flow_total      # pd.Series  — total flow

conn = structs.connections["Dam"]
conn.stage_hw       # pd.Series
conn.stage_tw       # pd.Series

# WSE and flow profiles along / across a line
# xy can be an (n, 2) ndarray or any object with a __geo_interface__ (e.g. shapely LineString)
xy = np.array([[500, 200], [600, 250], [700, 300]])
df     = model.wse_along_line(xy, timestep="max", interval=1.0)  # pd.DataFrame — station, wse, …
series = model.flow_across_line(xy)                               # pd.Series — discharge time series

Package Structure

rivia/
├── controller/  # COM interface to run/control HEC-RAS
├── model/       # Project — primary project interface; read/write text input files, read HDF results
├── hdf/         # Read HEC-RAS HDF5 geometry and result files
├── geo/         # Geospatial operations: raster export (geopandas/rasterio)
└── utils/       # Shared helpers

Development

git clone https://github.com/gyanz/rivia.git
cd rivia
pip install -e ".[dev,geo,docs]"

# Run tests
pytest tests/ -x --tb=short

# Lint
ruff check src/

# Type check
mypy src/rivia

# Build docs
sphinx-build -b html docs docs/_build/html

License

Copyright 2025 Gyan Basyal and WEST Consultants, Inc.

Licensed under the Apache License, Version 2.0.