Year

2021

Season

Fall

Paper Type

Master's Thesis

College

College of Computing, Engineering & Construction

Degree Name

Master of Science in Civil Engineering (MSCE)

Department

Engineering

NACO controlled Corporate Body

University of North Florida. School of Engineering

First Advisor

Dr. Raphael Crowley, Ph.D., P.E.

Second Advisor

Ryan Shamet

Third Advisor

James Gelsleichter, Ph.D.

Department Chair

Osama Jadaan, Ph.D.

College Dean

William Klostermeyer, Ph.D.

Abstract

There has been a growing concern in recent years about the effects of anthropogenic noise due to pile driving on underwater wildlife. Current guidelines for mitigating hydroacoustic effects associated with these geotechnical events are based upon a relatively simple transmission loss formulation known as the Practical Spreading Loss Model (PSLM). This model is easy to implement, but it may produce overly conservative results. Sound data during pile drives from several sites in Florida showed much higher sound attenuation than predicted by the PSLM. The first part of this study focused on explaining this discrepancy using computational fluid dynamics. Specifically, synthetic pile drives were simulated using Siemens’ Star-CCM+. These models tracked sound decay from a single hammer blow that was imposed on a modeled pile using site-specific bathymetry data. Results showed that discrepancies between measured transmission loss coefficients and the PSLM could not be explained due to local bathymetry alone. However, if different sound absorption criteria were used at the sites’ mudlines, the model was able to replicate results. The data therefore suggest that geotechnical conditions may play a significant role in determining anthropogenic sound loss due to pile driving. 12 The second part of the study focused on using empirical data fitting to calibrate a physics based semi-empirical model for shallow water acoustics model by Rogers (1981) using a multidimensional curve fitting tool to model the difference between Rogers’ predictions and field data as a function of site-specific environment variables. The results produced a slightly improved model for transmission loss prediction but still faced the problem of an overabundance of parameters required for the input field to give reliable results. In an effort to address these issues, a new empirical model was developed to explain transmission loss by leveraging a large sound dataset collected in different sites in Florida and using linear regression to establish a relationship between the transmission loss coefficient and a source level dependent parameter (Ainslie, 2014). Results indicated the tool was able to compute more reliable transmission loss coefficients compared to the one currently in use by the NMFS calculator resulting in more accurate results for ranges with sound pressure levels below the thresholds. However, the new model showed an apparent dependency between sound attenuation and amplitude, and the physics associated with this apparent dependency require further investigation.

Share

COinS