RODEO II - Vineyard Wind Data Analysis Workflow

William Jenkins, Ph.D.
Scripps Institution of Oceanography University of California San Diego

This repository contains code used for analyses that have been submitted for peer review to the following publication:

Jenkins, W. F. and Lin, Y.T. Passive acoustic tracking of a fin whale during pile driving at an offshore wind construction site. Scientific Reports (2026).

This repository depends on the rodeo Python package.

Workflow

The general workflow is as follows:

graph TD
   subgraph "Per-Sensor Processing"
      A[Construct inventory of acoustic data] --> B[Detect pile driving strikes]
      B --> C[Extract, process, and save strikes to HDF5 database]
      C --> D[Compute cross-correlations between strike pairs]
      D --> E[Build database of templates]
      E --> G[Denoise the signal using the templates]
      G --> H[Pulse compress the denoised signals with the whale call]
      F[Extract templates of whale calls] --> H
   end


   H --> I[Find peaks and compute TDOA estimates]
   I --> J[Localize the TDOA estimates]

Loading

Configuration

The entire workflow can be configured using two configuration files: config/config.toml and config/inventory.toml.

config/config.toml contains general settings for the workflow, such as paths to data directories, parameters for processing steps, and settings for parallelization. config/inventory.toml specifies the acoustic data files to be processed and the parameters for processing them.

Running workflow end-to-end

The entire workflow can be run end-to-end using the command:

workflow run --config config/config.toml

The optional --config flag specifies the path to the configuration file, which contains settings for the workflow. If the flag is not provided, the workflow will look for a default configuration file at config/config.toml.

Running individual steps

The individual steps of the workflow can be run using the following commands:

1. Run the ETL job:
   workflow etl
2. Run all data processing steps:
   workflow process
3. Run the localization step:
   workflow localize
4. Run the plotting step:
   workflow plot

Acknowledgements

Study concept, oversight, and funding were provided by the U.S. Department of the Interior, Bureau of Ocean Energy Management, Environmental Studies Program, Real-Time Opportunity for Development Environmental Observations (RODEO) Program, Washington, DC under Contract Number 140M0121D0003_CTO 140M0122F0030. Dr. David Bigger, Contracting Officers Representative.