stormcloud

Creates datasets for hydrologic modeling in the cloud.

---

Workflow

---

Adding Temperature Data

Write AORC data for precipitation, temperature, or both to DSS format

• Requires use of docker container as specified in Dockerfile.temp_precip

Identifying storms of interest

Identify top ranked storms tracked in meilisearch database, rerun zarr to DSS conversion and save to local path

• Requires use of docker container as specified in Dockerfile.top_storms_dss

Getting valid transposition geometry

Get the geometry defining all valid transposes within a transposition region for a given watershed and saves the geometry as a simple geojson polygon

• Requires use of docker container as specified in Dockerfile.transpose_geom

Plugins -- Process API support

Process API plugins exist which support:

• SST modeling runs
• DSS file extraction from zarr data
• Ranked document creation from SST modeling run output

Meilisearch

Meilisearch acts as a back end for the StormViewer site by holding, sorting, and filtering metadata associated with SST runs. In this repo, there are scripts which: - recreate the meilisearch index settings used by the project - query the meilisearch index for storms using user-provided filters - create an HMS grid package for a subsection of events from s3 data located using SST metadata - update the meilisearch index with ranked documents generated by the ranked document plugin

SST plugin directory

Temperature and precipitation extraction plugin directory

Metadata standardization plugin directory

HMS GRID package generation plugin directory

Ranked document plugin directory

---

To test plugins:

• edit dockerfile of plugin of interest, uncommenting the entrypoint execution block
• run build_and_run.sh followed by either 'sst', 'temp_precip', 'standardize_meta', 'hms_grid', or 'doc_rank' depending on which plugin you would like to test. Be aware that these tests result in outputs being written to s3, so test with caution