V3252039_Volpe UCSD-Scripps SOW 2025-06_PH_v2.pdf

Statement of Work

Support to Develop the Maritime Vessel Noise Module of the Maritime Analytics for Research and Innovation on Noise and Energy – Tool (MARINE-T)

Scripps Institution of Oceanography

PR# V3252039

Objective and Purpose

This project will support the development of the maritime vessel noise module of the United

States Department of Transportation (USDOT), Volpe National Transportation Systems Center

(Volpe Center) developed Maritime Analytics for Research and Innovation on Noise and Energy

– Tool (MARINE-T). This project will use field measurements from buoys located at three locations in U.S. waters to validate and calibrate the underwater noise module outputs. This work will support Volpe’s development of MARINE-T for the USDOT Maritime Administration

(MARAD).

The University of California San Diego Scripps Institution of Oceanography (SIO) will conduct this work by analyzing source level data from existing databases based on recordings at three monitoring stations located in the Santa Barbara Channel and Gulf coast, and compare those to the current JOMOPANS-ECHO (J-E) source level model which is used in MARINE-T. The J-E model is to a database of source levels from vessels operating in British Columbia, Canada. The

J-E model provides best fit predictions based on a large sample of empirical source level measurements collected at this single location over many years. The underlying dataset, as reflected in the reference paper for the J-E model and other published reports on the ECHO dataset, in some cases can still reflect substantial unpredicted variation among source levels with the same type, length and speed. The model’s fit is less confident for vessel types, lengths and speeds that are undersampled in the input data, and/or for which additional important drivers in predicting their source levels are unaccounted for (e.g., loading, machinery noise).

Variables that are not accounted for can include both those that affect the noise radiated by the vessel (vessel-specific design and operation characteristics) and those that affect the use of the monitoring data to accurately calculate source level (accounting for distance and oceanography between the source and the receiving recorder). In building systems that integrate information across Canadian and US monitoring locations to inform vessel operators as to their predicted underwater radiated noise (URN) source levels there is an opportunity to improve performance in both arenas: first, increase sample sizes for vessel types and operational modes that are currently not confidently predicted by the J-E model’s core output predictors (speed, length, type), and second identify measurement specifics (e.g., frequencies) that suggest the need for additional work to ensure comparability across systems.

SIO will utilize an existing source level database drawing from three monitoring locations in US waters that pairs acoustic data and Automated Identification System (AIS). Source levels are in one-third-octave bands ranging from 15 Hz to 4 kHz. Additional metrics based on existing output, such as radiated noise level and sound pressure level may also be analyzed and interpreted as prioritized for comparisons discussed below. Recognizing that source levels can vary significantly due to differences in ship types, operational conditions, designs, and environmental factors, this project seeks to integrate source level estimates for vessels operating in United States waters in the MARINE-T noise module being developed for MARAD.

To achieve these goals, SIO will collaborate with the Volpe Center acousticians to interpret the source levels estimated in United States waters and identify variations among different vessel types and operating conditions. Ultimately, this work will enhance the accuracy of the Volpe MARINE-T noise module by augmenting their existing application of the J-E source level model, incorporating information based on source levels of vessels operating in U.S. waters, which will support improved regulatory and environmental assessments.

2. Project Description/Tasks

SIO will partner with Volpe to explore and interpret SIO’s existing source level database in support of Volpe’s decision support tool development effort for MARAD. SIO will provide expertise to the Volpe group to evaluate differences between the acoustic signatures obtained from the Santa Barbara Channel and Gulf coast location monitoring stations and the J-E source level model. Integrating information from previous SIO work [reference here] directly comparing ECHO measurements with Santa Barbara measurements, and using published accounts of the underlying empirical data supporting the J-E model predictions, SIO will further assess whether adjustments to the fit of the J-E model are recommended based on their additional sampling and/or if further measurement control is needed to address comparability.

The specific tasks are as follows:

Task 1: Interpret Existing Source Level Database

Provide the Volpe Center acousticians with existing source level estimates from the Santa

Barbara Channel and Gulf coast location; work with acousticians to interpret source levels relative to specific ship types and operations represented in the Volpe model; identify specific ship types, operations, and frequencies to compare with Volpe model predictions (largely based on current J-E model). This will largely entail matching a reasonable sampling of Volpe model predictions for vessels tracked by AIS in the 2022-based model with the same vessels measured empirically by SIO in Southern California or off southern Louisiana and in the lanes approaching

Galveston, Texas in the Gulf coast location. Given the large number of point samples that could be compared, this collaboration will entail consideration of a reasonable sample size across types and operational conditions of interest to evaluate the performance of the model. It should be emphasized that it is expected that these point measurements and model-based predictions will vary. This ground truthing should therefore consider what sampling is needed to evaluate the relative performance of the model across categories of vessels in the Volpe model/US-based dataset.

Task 2: Compare Source Level Database to JOMOPANS-ECHO Model

Evaluate compatibility of SIO’s source level database with the J-E model and (based on prior work and published reports) the underlying ECHO dataset; determine which ship types, operations, and frequencies are better sampled when the US-based datasets are included (e.g., vessel types, speeds, other operational states); for vessel types, speeds, operational states with common robust representation across both US and Canadian datasets, identify specific frequencies that show more variation based on recording location (indicated methodological variation not captured in calculations); as appropriate, suggest modifications to the Volpe model that would better represent vessel source levels inclusive of the SIO source level database. The ultimate goal for the Volpe model is for predictions to best reflect data available from many comparable databases, allowing for derivation of source level associated with types, sizes and operational modes that are not fully represented in any one location. Thus, the goal is to provide initial recommendations for adjustments to the JE coefficients that would be inclusive of the SIO results.

Task 3: Final Reporting

A final report will document the interpretation of the SIO source level database in terms of ship type, operations, and frequency. Additional resolution provided by the SIO source levels and vessel type/frequency/operations that show high variance with predictions based solely on the

J-E model will be outlined. Potential paths to increase accuracy of the Volpe model to reflect vessel source levels regardless of operational location more accurately will be discussed.

Meaning, the existing J-E model has some implicit assumptions about the locations where the vessel data come from, and this work will help to generalize the model to be more general to not just the specific areas where the data underlying the J-E model were collected.

3. Project Timeline

The project will be broken into the 3 tasks listed in Table 1 below and described in section 2.

Tasks shown may be completed concurrently depending on schedule execution.

Table 1 – Tasks and Completion Dates

Task Number and Name Estimated Start Date

Estimated Completion Date

Task 1: Interpret SIO Source Level Database Date of Award One Month After Award

Task 2: Compare Source Level Database to J-E Model

One Month After Award

Ten Weeks After Award

Task 3: Final Reporting Ten Weeks After Award

Four Months After

4. Reporting

Upon written acceptance of a contract, Volpe and SIO will arrange a kick-off meeting to finalize details of project timing.

SIO will provide monthly updates to Volpe e-mail; provide a final written report in accordance with this scope; coordinate regular conference calls as needed; share data as it may be pertinent to ongoing discussions and coordinate final data gathering and assembly into the final report.

5. Deliverables

SIO will provide four monthly reports to Volpe showing the progress on each task and deliverable. A draft final report will be delivered to Volpe upon the conclusion of the data interpretation and analysis effort. After review by Volpe, SIO will review all comments and suggested edits and incorporate changes into the final report which will be delivered within 2 weeks of receiving the Volpe comments. The final report will describe the objective of the project, detail the activities undertaken for each task, and describe the technical outcomes of the project in depth. The report will further include a discussion of project uncertainties and their outcomes, other issues that arose during any stage of the project, and lessons learned from the analysis and comparisons. Deliverables will be submitted by email to the Contracting Officer’s Representative, Peter Herzig, at Peter.Herzig@dot.gov.

Table 2 – Deliverables

Task and Deliverable Estimated Completion Date

Four Monthly Reports ~15th of Each Month

Draft Final Report Ten Weeks After Award

Final Report Four Months After