Modeling of bottom-limited acoustic transmission in the Florida Straits

Abstract

Most tomography experiments rely on a ray interpretation for inversions. Ideally, separable and stable pulse arrival patterns can be used to measure arrival time along identifiable ray paths, but many areas of oceanographic interests have more complicated acoustic transmission that cannot be precisely modeled with ray theory. Acoustic transmission in the Florida Straits is complicated by multiple interactions with an irregular bottom, a strong range dependence of sound speed, and very energetic oceanographic variability that results in an order of magnitude greater fluctuation in sound speed than observed in the deep ocean. Yet, many experimental observations dating from the sixties reveal a remarkable temporal stability of acoustic transmission and a strong correlation between fluctuations of acoustic travel time and oceanographic variables, such as transport and temperature. In 1983, two 3-point reciprocal transmission tomography experiments were conducted for a period of 1 month, each over a short range of 25 km in 250-m water depth and a longer range of 45 km in 400 m. Channel pulse responses were measured at 12-min intervals for 30 days, along with temperature and current measurements along the path of propagation. Neither the short or long ranges exhibit clean identifiable arrivals associated with an individual ray path that can be used for inversions. The short range is characterized by a single spread out arrival composed of numbers of RBR path arrivals that overlap in time. The long range exhibits a similar RBR group, but with an additional six distinct earlier arrivals associated with the surface duct. After some temporal averaging, channel pulse responses are very stable and inversions appear to give precise measurements of average current and temperature. Presented here are broadband modeling results using normal mode, parabolic equation, and ray models with environmental measurements as imputs. For all ranges the forward problem can be solved with precision, and inversions are possible. The shortcomings of ray theory in describing this type of propagation become apparent, even for those ranges having negligible range dependence. While all the data and model predictions are for the Florida Straits and conditions of relatively short range and shallow water, results and methods may be transferable to long-range, deep-ocean experiments in situations where ray arrivals are closely spaced in time and cannot be separated or identified, or in the presence of a surface duct or other strong stratified channels.