climdata 0.3.2

This project automates the fetching and extraction of weather data from multiple sources — such as MSWX, DWD HYRAS, ERA5-Land, NASA-NEX-GDDP, and more — for a given location and time range.

climdata

climdata is a Python package designed to automate fetching, extraction, and processing of climate data from various sources, including MSWX, DWD HYRAS, ERA5-Land, and NASA-NEX-GDDP. It provides tools to retrieve data for specific locations and time ranges, facilitating climate analysis and research.

---

Key features

• Fetch and load datasets: MSWX, CMIP (cloud via intake), DWD, HYRAS
• Spatial extraction: point, box (region via config bounds), or shapefile (GeoJSON/Feature)
• Temporal subsetting via config or programmatic call
• Multi-format export: NetCDF, Zarr, CSV (standardized long format: variable, value, units)
• Hydra configuration + easy CLI overrides
• Helper to normalize AOI (GeoJSON → point / bbox / polygon)
• Provenance-friendly workflow (designed to be used with CI/CD workflows)

Install (development)

1. Clone repository

git clone <repo-url>
cd climdata

2. Create virtualenv and install deps

python -m venv .venv
source .venv/bin/activate
pip install -U pip
pip install -e ".[dev]"   # or pip install -r requirements.txt

Quick CLI (Hydra) usage

Hydra reads configs from conf/. Override any config value on the CLI.

Examples:

# Region extraction (saves NetCDF by default when region is used)
python examples/climdata_cli.py dataset=CMIP region=europe time_range.start_date=2010-01-01 time_range.end_date=2010-12-31

# Point extraction (saves CSV)
python examples/climdata_cli.py dataset=MSWX lat=52.5 lon=13.4 variables=['tas','pr'] time_range.start_date=2000-01-01

# HYRAS / DWD (point only)
python examples/climdata_cli.py dataset=HYRAS lat=52 lon=10

Notes:

• Use dataset=<MSWX|CMIP|DWD|HYRAS> in CLI.
• Override any config key: e.g. time_range.start_date=2000-01-01.
• DWD/HYRAS: region (box) extraction is not supported — script will raise an error if attempted.

Programmatic usage

Use the wrapper to compose configs, preprocess AOI, extract, and save.

from climdata.utils.wrapper import extract_data

# returns (cfg, filename, ds, index) when save_to_file=True
cfg, filename, ds, index = extract_data(cfg_name="config", overrides=["dataset=MSWX","lat=52.5","lon=13.4"])

Or use the dataset classes directly:

import climdata, xarray as xr
cmip = climdata.CMIP(cfg)
cmip.fetch()
cmip.load()
cmip.extract(box=cfg.bounds[cfg.region])
cmip.save_netcdf("output.nc")

Configs

• Config files live in climdata/conf/. There are dataset-specific config entry points e.g. config_cmip, config_mswx, etc.
• Filename templates are configurable in cfg.output:
  • cfg.output.filename_nc
  • cfg.output.filename_csv
  • cfg.output.filename_zarr

The wrapper generates filenames via get_output_filename(cfg, output_type, ...) using cfg.bounds, cfg.time_range, etc.

Output CSV format

CSV produced by save_csv is standardized to the long form with columns (where available):

• source_id, experiment_id, table_id, time, lat, lon, variable, value, units

This ensures a single value column and a variable column for stacked variables.

Common issues & tips

• NetCDF write ValueError (datetime encoding): call ds["time"].encoding.clear() before to_netcdf() (wrapper handles this).
• PermissionError writing files: ensure output directory is writable or write to /tmp/ (or adjust permissions).
• CMIP cloud access requires network access — use the Pangeo intake catalog URL already referenced in code.

AOI handling

preprocess_aoi(cfg) accepts:

• GeoJSON strings / Feature / FeatureCollection
• Point → sets cfg.lat, cfg.lon
• Polygon or bbox → sets cfg.bounds['custom'] and cfg.region='custom'

HYRAS support

HYRAS class mirrors MSWX design:

• fetch() / load() / extract(point=...) / save_csv() / save_netcdf()
• HYRAS extraction currently supports point extraction; attempt to use a region will raise an error.

Development & provenance

• CI: add GitHub Actions workflows to run tests and build/publish to PyPI.
• Keep config and runtime overrides in Hydra to enable reproducible runs.
• Include CITATION.cff, license, and a changelog for FAIR discoverability.

Contributing

• Run tests: pytest
• Style: follow repository linting config
• Open PRs against main with tests and a short changelog entry

License

Specify the license (e.g. MIT or Apache 2.0) in LICENSE.

---

For further examples, see examples/ and the docs/ folder (usage, installation, faq).// filepath: /beegfs/muduchuru/pkgs_fnl/climdata/README.md

climdata

Lightweight toolkit to fetch, subset and export climate data (MSWX, CMIP, DWD, HYRAS).
Provides a Hydra-driven CLI, programmatic wrapper, cloud-native CMIP access, local dataset handling, and standardized CSV/NetCDF/Zarr exports.

Key features

• Fetch and load datasets: MSWX, CMIP (cloud via intake), DWD, HYRAS
• Spatial extraction: point, box (region via config bounds), or shapefile (GeoJSON/Feature)
• Temporal subsetting via config or programmatic call
• Multi-format export: NetCDF, Zarr, CSV (standardized long format: variable, value, units)
• Hydra configuration + easy CLI overrides
• Helper to normalize AOI (GeoJSON → point / bbox / polygon)
• Provenance-friendly workflow (designed to be used with CI/CD workflows)

Install (development)

1. Clone repository

git clone <repo-url>
cd climdata

2. Create virtualenv and install deps

python -m venv .venv
source .venv/bin/activate
pip install -U pip
pip install -e ".[dev]"   # or pip install -r requirements.txt

Quick CLI (Hydra) usage

Hydra reads configs from conf/. Override any config value on the CLI.

Examples:

# Region extraction (saves NetCDF by default when region is used)
python examples/climdata_cli.py dataset=CMIP region=europe time_range.start_date=2010-01-01 time_range.end_date=2010-12-31

# Point extraction (saves CSV)
python examples/climdata_cli.py dataset=MSWX lat=52.5 lon=13.4 variables=['tas','pr'] time_range.start_date=2000-01-01

# HYRAS / DWD (point only)
python examples/climdata_cli.py dataset=HYRAS lat=52 lon=10

Notes:

• Use dataset=<MSWX|CMIP|DWD|HYRAS> in CLI.
• Override any config key: e.g. time_range.start_date=2000-01-01.
• DWD/HYRAS: region (box) extraction is not supported — script will raise an error if attempted.

Programmatic usage

Use the wrapper to compose configs, preprocess AOI, extract, and save.

from climdata.utils.wrapper import extract_data

# returns (cfg, filename, ds, index) when save_to_file=True
cfg, filename, ds, index = extract_data(cfg_name="config", overrides=["dataset=MSWX","lat=52.5","lon=13.4"])

Or use the dataset classes directly:

import climdata, xarray as xr
cmip = climdata.CMIP(cfg)
cmip.fetch()
cmip.load()
cmip.extract(box=cfg.bounds[cfg.region])
cmip.save_netcdf("output.nc")

Configs

• Config files live in climdata/conf/. There are dataset-specific config entry points e.g. config_cmip, config_mswx, etc.
• Filename templates are configurable in cfg.output:
  • cfg.output.filename_nc
  • cfg.output.filename_csv
  • cfg.output.filename_zarr

The wrapper generates filenames via get_output_filename(cfg, output_type, ...) using cfg.bounds, cfg.time_range, etc.

Output CSV format

CSV produced by save_csv is standardized to the long form with columns (where available):

• source_id, experiment_id, table_id, time, lat, lon, variable, value, units

This ensures a single value column and a variable column for stacked variables.

Common issues & tips

• NetCDF write ValueError (datetime encoding): call ds["time"].encoding.clear() before to_netcdf() (wrapper handles this).
• PermissionError writing files: ensure output directory is writable or write to /tmp/ (or adjust permissions).
• CMIP cloud access requires network access — use the Pangeo intake catalog URL already referenced in code.

AOI handling

preprocess_aoi(cfg) accepts:

• GeoJSON strings / Feature / FeatureCollection
• Point → sets cfg.lat, cfg.lon
• Polygon or bbox → sets cfg.bounds['custom'] and cfg.region='custom'

HYRAS support

HYRAS class mirrors MSWX design:

• fetch() / load() / extract(point=...) / save_csv() / save_netcdf()
• HYRAS extraction currently supports point extraction; attempt to use a region will raise an error.

Development & provenance

• CI: add GitHub Actions workflows to run tests and build/publish to PyPI.
• Keep config and runtime overrides in Hydra to enable reproducible runs.
• Include CITATION.cff, license, and a changelog for FAIR discoverability.

Contributing

• Run tests: pytest
• Style: follow repository linting config
• Open PRs against main with tests and a short changelog entry

License

Specify the license (e.g. MIT or Apache 2.0) in LICENSE.

---

For further examples, see examples/ and the docs/ folder (usage, installation, faq).