Rm. Jones et al., NONPERTURBATIVE OCEAN ACOUSTIC TOMOGRAPHY INVERSION OF 1000-KM PULSE-PROPAGATION IN THE PACIFIC-OCEAN, The Journal of the Acoustical Society of America, 96(5), 1994, pp. 3054-3063
A nonperturbative inversion was performed of acoustic tomography measu
rements made in the northeastern Pacific Ocean in July 1989, in which
acoustic transmissions from a 250-Hz broadband source located near the
sound-channel axis were recorded at a long vertical array of hydropho
nes 1000 km away. In contrast with a conventional inversion, this nonp
erturbative inversion does not assume that travel times are linearly r
elated to the sound-speed deviations from a background sound-speed mod
el. The inversion process involved three steps: (1) Measured pulse tra
vel times and the source and receiver locations were used to determine
the range average of the equivalent symmetric sound-slowness profile.
That part of the inversion used only curve fitting and Abel transform
s, and required independent (nontomographic) information only to help
identify the pulse arrivals. (2) Under the assumption that the range d
ependence of sound speed was small, we used the reciprocal of the rang
e-averaged sound-slowness profile to approximate the range average of
the sound-speed profile. (3) Constraining the sound speed below the so
und-channel axis to match climatological data and neglecting the range
dependence of sound speed below the sound-channel axis allowed us to
estimate the range average of the sound-speed profile above the sound-
channel axis. This inversion was compared with the range average of so
und speed calculated from CTD measurements made during the experiment
over a 10-day period. The agreement was good between 50- and 300-m dep
ths, but there were some disagreements near the surface and near the s
ound-channel axis.