Size Constraints on Late Miocene to Pliocene Submarine Slope Failures along the Colombian Caribbean Subduction Margin as a Basis for Assessing Circum-Caribbean Impact of Future Tsunami Events

Abstract:

The Colombian Caribbean margin provides an ideal setting for the formation of large mass transport deposits (MTDs): 1) the Caribbean Plate is slowly subducting at rates of 20 mm/yr with infrequent large thrust earthquakes and a complete lack of subduction events in the 400-year-long historical record; 2) the margin is a broad zone of active faults including a ~50 km-wide accretionary prism and strike-slip faults landward of the prism; 3) the active margin is draped by the Magdalena delta and submarine fan fed by the Magdalena River, the 26^th largest in the world; and 4) the margin is over-steepened to slopes of up to 7° from the combination of tectonic activity and rapid rates of deltaic progradation. Using seismic data we have identified three late Miocene-Pliocene MTDs, the largest of which is between 4500 and 6000 km³, comparable in size to the well-studied Storegga slide of Norway. The tsunamigenic potential of future, analog MTD events are modeled using GeoWave tsunami modeling software. The largest and youngest of these MTDs, the Santa Marta slide, is used as an analog to infer the location and input parameters for the tsunami model. The event is modeled as a translational slide ~46 km long and ~37 km wide with the center of the slide located ~57 km W/NW from the mouth of the present day Magdalena River in water depths of 1500 m. The volume for the initial failure is conservatively estimated at ~680 km³ of material. The resulting tsunami wave from such an event has an initial maximum trough amplitude of -65.8 m and a peak amplitude of 19.2 m. The impact of such a tsunami would include: 1) Kingston, Jamaica (population 938K), tsunami height 7.5 m, peak arrival at 60 min.; 2) Santo Domingo, Dominican Republic (population 965K, height 6 m, peak arrival at 80 min.); and 3) Cartagena, Colombia (population 845K, height 21 m, peak arrival at 34 min.). A number of parameters to the model are varied to analyze sensitivity of modeling results to changes in slide depth, angle of failure, slide volume, and slide density.