ALBERT

Description: Seafloor topography is neither spatially homogeneous, nor does it obey Gaussian statistics; deviations from both of these assumptions are important from a geological and acoustic point of view. It has been found that the distribution of topographic slopes can be used as a primary tool for understanding the sources and extent of spatial heterogeneities and patterns on the seafloor. The covariance function has been widely used to characterize seafloor topography, but requires the assumption of Gaussian joint probability statistics to be valid. For heterogeneous topography characterized by large transient signals such as steep scarps and volcanoes, the covariance becomes dominated by the transients; in contrast the family slope distributions can still be used to derive stable descriptors for regions with large transient signals, as well as regions containing asymmetric features, and regions with only limited sampling. Knowledge of slopes is useful because a direct relation exists between the covariance and the slope distributions at different spatial scales. Studies of the slope distribution provide a means of identifying the presence of the non‐Gaussian elements in the topography, and flagging their spatial locations. The methods used here are demonstrated by applying them to three small patches of topography located within 20 km of each other in the Eastern Pacific. It is found that dominant azimuthal directions and dip angles differ widely between the patches. In addition, asymmetries in the cross‐sectional shapes of faulted abyssal hills are documented. [Work supported by ONR.]