ABYSSAL HILL TOPOGRAPHY AS AN INDICATOR OF EPISODICITY IN CRUSTAL ACCRETION AND DEFORMATION

Recent studies of seafloor morphology suggest that abyssal hills are created at the ridge axis by the interplay of magmatic and tectonic processes; therefore, abyssal hills should be ideal indicators of episodicity. In this study, we use abyssal hill topography created near the ridge axis and subsequently transported onto the ridge flanks as a time series reflecting the episodicity in magmatism/tectonism at the ridge axis. If the creation of an abyssal hill is viewed as a discrete event (i.e., a pulse of magmatism or faulting), we can parameterize episodicity by considering a sequence of events of magnitude M (proportional to the relief or cross-sectional area of a hill) separated by time intervals Δt (proportional to the spacing between hills). This analysis will allow us to decide if abyssal hills are periodic or if they are randomly distributed in time as a Poisson process. Ten profiles 200-1000 km long of unaveraged centerbeam Sea Beam, located along flowlines and covering a range of spreading rates (15-60 km/Ma half-rate), were used to measure locations and characteristics of the relief (cross-sectional area and height) of abyssal hills. The principal results of the analysis are that: (1) abyssal hills are clearly not periodic, and their distribution in time corresponds to a Poisson process with a characteristic frequency λ; the characteristic frequency λ is roughly proportional to the square of the spreading rate, and varies from 12.8 events/Ma at fast rates to 1.4 events/Ma at slow rates; and (2) while the magnitudes M of abyssal hills (cross-sectional areas and reliefs) are positively correlated with the time elapsed since the formation of the previous hill, there is no correlation between M and the amount of time to the next event. These results agree with a scenario in which magma or stress accumulate at a steady rate at the ridge axis, and are episodically released in discrete events of volcanism or faulting whose magnitude is proportional to the time elapsed since the last event. © 1990.