Revisions to an empirical surface loss model using a correction for pH‐dependent attenuation

Empirically derived surface loss values are not obtained directly but are derived from propagation loss measurements that include at least two other components. Surface loss is determined by making assumptions with regard to spreading loss, volume attenuation (and bottom loss, if appropriate). Marsh and Schulkin [AVCO Marine Electronics Office Tech. Rep. (1962)] analyzed an extensive set of transmission loss data from surface duct propagation to obtain surface loss values. They assumed that attenuation was entirely due to magnesium sulfate and used the early Marsh–Schulkin [J. Brit. IRE 25, N6 (1963)] attenuation formula (Thorp, Mellen et al., and Francois Garrison attenuation formulas were not known at the time). Corrections to the Marsh–Schulkin surface loss data have been computed that account for the differences between the Marsh–Schulkin and Mellen et al. [NUSC TR 7293 (1983)] attenuation formulae. The revised values show less loss for all frequency‐wave‐height (F‐H) products. The Kuo [J. Acoust. Soc. Am. 36 (1964)] perturbation theory estimates of surface loss are shown to be in good agreement with the revised data at low F‐H products.