pydsm

For working with DSM2 input (.inp) and output (.h5 and .dss) with Python (pandas).

• Free software: MIT license
• Documentation: https://cadwrdeltamodeling.github.io/pydsm

---

Features

• Reads DSM2 HDF5 tidefiles into pandas DataFrames
  • Hydro tidefile: time series data, input tables, and geometry tables
  • Qual tidefile: time series data and input tables
  • GTM tidefile: constituent concentration data
• Reads and writes DSM2 echo/input files into and from pandas DataFrames
• Working with DSS time series files: extract, compare, copy, extend, and convert data

---

Installation

Prerequisites

• Miniconda or Anaconda
• The cadwr-dms conda channel is required for pyhecdss and vtools3

Conda (recommended)

conda env create -f environment.yml
conda activate pydsm

This installs the core runtime plus visualization extras (geopandas, shapely).

pip

pip install pydsm

Install optional extras as needed:

pip install "pydsm[viz]"   # geopandas, shapely
pip install "pydsm[dev]"   # test + docs tools

---

Commands

The pydsm command-line tool provides the following subcommands:

pydsm --help

DSS file utilities

Command	Description
pydsm extract-dss <dssfile>	Extract time series from a DSS file; optionally filter by C-part, apply Godin filter, and resample to daily/monthly values. Output can be pickle (.gz/.zip/.bz2), HDF5 (.h5), or DSS.
pydsm compare-dss <dssfile1> <dssfile2>	Compare two DSS files on matching B/C-part pathnames and write goodness-of-fit metrics (RMSE, Nash-Sutcliffe, percent bias, etc.) to a CSV.
pydsm copy-all-dss <from_file> <to_file>	Copy all paths from one DSS file to another.
pydsm csv-to-dss <csv_file> <dss_file>	Convert a CSV time series file to a DSS file with configurable path parts, units, period type, resampling, and a numeric multiplier.
pydsm extend-dss-ts <dss_filename> <dss_ext_filename>	Extend time series in a DSS file by a number of days (default 366) by appending a shifted copy at the end.
pydsm repeating create <datafile>	Create a repeating annual time series from a CSV template year and write it to a DSS file.
pydsm repeating extend <datafile>	Extend an existing repeating time series in a DSS file forward to a given end year.

DSM2 input / echo files

Command	Description
pydsm pretty-print-input <input_file> [output_file]	Reformat (pretty-print) a DSM2 .inp echo file. Defaults to <basename>.pretty.inp.
pydsm create-dsm2-input-for-cd <dss_filename> <dsm2_input_filename> <file_field_string>	Generate a DSM2 .inp boundary file for consumptive demand from a DSS file.
pydsm chan-orient <channel_line_geojson_file> <hydro_echo_file>	Generate a channel orientation file (angles) from a GeoJSON channel geometry and a Hydro echo file.

HDF5 tidefile utilities

Command	Description
pydsm slice-hydro <infile> <outfile> <stime> <etime>	Slice a Hydro HDF5 tidefile to a time window (e.g. 1990-01-10 to 1990-03-31) and write a new tidefile.
pydsm update-hydro-tidefile-with-inp <hydro_tidefile> <input_file>	Patch an HDF5 Hydro tidefile's input table from a .inp file.
pydsm create-gtm-restart <tidefile> <target_time> <outfile>	Write a GTM/Qual restart file from an HDF5 tidefile at the nearest stored time step to target_time (e.g. 05FEB2020 0300).

---