Functionality Overview¶

Strain Data (Frames)¶

Download gravitational wave strain data from GWOSC

• 32-second and 4096-second frame files

• Multiple detectors: H1, L1, V1

• Automatic cache file generation

• Support for all observing runs (O1-O4)

See: Accessing strain data for detailed documentation.

Calibration Uncertainty Envelopes¶

Multiple Calibration Sources

• Local filesystem (LIGO detectors O2-O4)

• Frame files (Virgo O4+)

• PESummary metafiles

• Automatic observing run detection

Supported Formats

• ASCII text files (LIGO format)

• Frame-embedded calibration (Virgo)

• Bilby-compatible output format

• Per-detector version control

Observing Runs

• O2: LIGO (H1, L1) and Virgo (V1)

• O3a/O3b: LIGO and Virgo

• O4a/O4b/O4c: LIGO and Virgo (frame-based)

See: Accessing Calibration Uncertainty Envelopes for detailed documentation.

Posterior Samples¶

PESummary Integration

• Extract posterior samples from metafiles

• Support for multiple analyses per file

• Preserves original metafile format

• Automatic analysis selection

Use Cases

• Reproducing published results

• Building on previous analyses

• Comparing different waveform models

• Ensemble studies

See: Accessing posterior samples for detailed documentation.

Power Spectral Densities (PSDs)¶

PSD Extraction

• From PESummary metafiles

• Multiple output formats (ASCII, XML)

• Per-detector PSD files

• Compatible with inference codes

Output Formats

• Tab-delimited ASCII (.dat)

• XML format (.xml.gz)

• Bilby-compatible

• LALInference-compatible

See: Accessing posterior samples for detailed documentation.

Flexible YAML Configuration¶

Data Selection

• Specify multiple data types in a single config

• Mix and match data sources

• Filter by detector, time range, analysis

Time Configuration

• GPS time ranges

• Frame durations (32s or 4096s)

• Automatic time validation

Source Configuration

• PESummary metafiles

• Local file systems

• Frame files

• GWOSC public data

See: Configuration Reference for complete reference.

Asimov Workflow System¶

Blueprint-Based Configuration

• Seamless integration with asimov

• Event-based data retrieval

• Conditional logic support

• Placeholder substitution

Workflow Features

• Automatic job submission

• HTCondor integration

• Progress tracking

• Asset collection

Example Use Cases

• Parameter estimation pipelines

• Data quality studies

• Multi-event analyses

• Systematic studies

Command-Line Interface¶

Standalone Operation

• Direct YAML configuration

• No asimov required

• Quick data retrieval

• Scripting support

Example Commands

# Download frames for GW150914
$ gwdata --settings gw150914.yaml

# Extract posterior from metafile
$ gwdata --settings extract_posterior.yaml

See: Command Line Interface for detailed CLI documentation.

Multi-Version Support¶

• Version-specific calibration handling

• Per-detector version specification

• Backward compatibility

• Automatic version detection

Observing Run Detection¶

• Automatic run identification from GPS time

• Run-specific calibration retrieval

• Cross-run compatibility

• Future-proof design for new runs

Error Handling¶

• Graceful failure modes

• Informative error messages

• Missing data detection

• Validation checks

Authentication¶

• SciTokens support

• SciTokens integration

• Automatic credential handling

• Secure data access

Directory Structure¶

The tool creates organized output directories:

output/
├── frames/          # Strain data files
├── cache/           # Frame cache files
├── calibration/     # Calibration envelopes
├── posterior/       # Posterior samples
└── psds/            # Power spectral densities

File Naming Conventions¶

• Detector-specific filenames (e.g., H1.dat)

• Standard extensions (.gwf, .dat, .h5)

• Cache format compatible with LAL

• Human-readable organization

Reproducing Published Results¶

1. Download posterior from official data release

2. Extract PSDs and calibration from metafile

3. Use data to reproduce published plots

4. Compare with independent analyses

Parameter Estimation Pipeline¶

1. Download strain data for event

2. Retrieve appropriate calibration

3. Configure inference code

4. Run parameter estimation

5. Compare with published results

Open Data Tutorial¶

1. Download public GWOSC data

2. Get calibration for observing run

3. Demonstrate analysis techniques

4. Educational/outreach purposes

Multi-Event Study¶

1. Batch download data for multiple events

2. Standardized calibration retrieval

3. Consistent data format

4. Population analysis ready

Data Quality Investigation¶

1. Download strain data

2. Compare multiple analyses

3. Investigate systematic effects

4. Validate calibration

LIGO Detectors¶

• H1 (LIGO Hanford)

• L1 (LIGO Livingston)

Data Access: GWOSC (public), IGWN (private)

Calibration: Local filesystem (O2-O4)

Virgo Detector¶

• V1 (Virgo)

Data Access: GWOSC (public), IGWN (private)

Calibration:

• O2-O3: Local filesystem

• O4+: Frame files

KAGRA Detector¶

• K1 (KAGRA)

Data Access: GWOSC (public), IGWN (private)

Calibration: TBD based on collaboration distribution

Supported Runs¶

• O1 (2015-2016): Frames only, no calibration

• O2 (2016-2017): Full support

• O3a (2019): Full support

• O3b (2019-2020): Full support

• O4a (2023-2024): Full support

• O4b (2024-2025): Full support

• O4c (2025): Planned support (dates subject to confirmation)

Current Limitations¶

• O1 calibration not available

• KAGRA calibration pending

• Requires IGWN credentials for private data

• Network connectivity required for downloads

Known Issues¶

• HTCondor dependency for asimov integration

• Network timeouts for large frame downloads

• Version detection edge cases at run boundaries

Download Speed¶

• Depends on network bandwidth

• OSDF/Pelican optimized for large files

• Parallel downloads not currently supported

Storage Requirements¶

• 4096s frames: ~100MB per detector per file

• 32s frames: ~1MB per detector per file

• Calibration files: <1MB per detector

• PSDs: <1MB per detector

Planned Features¶

• Parallel frame downloads

• More data validation checks

• Additional output formats

• Enhanced error recovery

• Automated data quality checks

Community Contributions¶

We welcome contributions! See Contributing to asimov-gwdata for guidelines.