Abstract
This presentation will discuss recent experience with geoacoustic inversion of transmission loss (TL) data from 50 to 5000 Hz from both deep and shallow water regions. For shallow water, a multi-layered geoacoustic profile with range dependent layer thicknesses is used. Layer thickness is not an inversion parameter but is known from ancillary data such as seismic profiling and chirp sonar. Issues related to range dependent sound speed profiles (SVPs) will be discussed as will observation of the time spread of multipath arrivals. For deep water, a single layer geoacoustic profile is assumed. Quality of inversion is assessed using comparison of measured and modeled TL and the estimation error. The estimation error is the dependence of the total rms error—summed over range, frequency, and receivers—on the values of a geoacoustic parameter within the bounds used to constrain inversion. The estimation error identifies geoacoustic parameters that are well constrained or poorly constrained by the TL data. Inversion was carried out using a genetic algorithm model [D. Harvey et al., Oceans '02 MTS/IEEE, 1, 358–362 (2002)]. Acoustic and environmental data were provided by the Naval Oceanographic Office. [Work funded by the Ocean Bottom Characterization Initiative through PEO C4I, PWM 120.]
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call