EarthScope SFG Workflows

earthscope-sfg-workflows provides workflow orchestration and data management for EarthScope Seafloor Geodesy (SFG) pipelines: discovery, ingestion, preprocessing, mid-processing, modeling (GARPOS), and the campaign/station/network lifecycle that ties them together.

It is the orchestration counterpart to earthscope-sfg-tools, which provides the low-level parsing, file conversion, and data-model primitives that workflows compose.

What’s in this repo¶

Quick start¶

from pathlib import Path
from earthscope_sfg_workflows.workflows import WorkflowHandler

handler = WorkflowHandler(directory=Path("/path/to/SFGMain"))
handler.set_network_station_campaign("cascadia-gorda", "NCC1", "2024_A_1126")

handler.ingest_discover_archive()
handler.download_data()
handler.preprocess_run_pipeline_sv3()
handler.run_garpos()

The CLI is exposed as sfgtools:

sfgtools run path/to/manifest.json
sfgtools preprocess \
    --main-dir ./SFGMain \
    --network cascadia-gorda \
    --campaign 2024_A_1126 \
    --stations NCC1

Docs sections¶

• Installation — environment setup with Pixi, GARPOS, and PRIDE-PPPAR

• Development — linting, testing, and contributing

• API Reference — auto-generated module documentation

Architecture¶

The data layer follows the ports & adapters (hexagonal) pattern. Workflows interact only with three ports — AssetCatalogPort, FileStorePort, ArchiveSourcePort — never with concrete implementations.

WorkflowHandler wraps a Workspace, which manages a pool of StationSession instances keyed by (network, station). Services (IngestService, ProcessingService, SyncService) are lazy properties on each session, constructed on first access.

The Environment singleton (config/env_config.py) is the single source of truth for runtime settings.

Environment variables¶