Subduction erosion, and the de-construction of continental crust: The Central America case and its global implications

Abstract

The relative rates of creation and destruction of continental crust at subduction zones are a key factor shaping the evolution of continental crust through time. Central America, arguably the best studied place where subduction erosion has been documented, is used here to assess past rates and modes of forearc recycling. Drilling from Guatemala to Costa Rica indicates that subduction erosion has been active since at least the early Miocene. Drilling also shows that the rates of subduction erosion have varied significantly both along strike and through time. The Integrated Ocean Drilling Program (IODP) Expedition 334 to southern Costa Rica documents unprecedented subduction erosion there — at rates larger than the fastest known rates of forearc accretion. In southern Costa Rica, accelerated subduction erosion of the upper plate initiated when the Panama Fracture Zone/Cocos Ridge, the latter being an over thickened aseismic ridge, arrived at the Middle America Trench. The forearc records this event with an unconformity at 2.2±0.2Ma. The recovered shelf sequence overlying the unconformity constrains a short (<2Myr) interval of extreme subsidence (~1200m) with a rapid pulse occurring during the first ~0.3Myr. This event removed an estimated 1.2×106km3 of forearc material at a rate of ~1125km3/Myr/km of trench during a time of rapid (~1035m/Myr) contemporaneous shelf sediment accumulation. Detrital apatite fission-track thermochronology on the sediments above the unconformity indicates the pattern of surficial sediment transport during this subduction erosion event. The fission track data show that sediments from the extinct and exhumed volcanic arc – the Cordillera de Talamanca – were able to immediately access the growing forearc basin after the onset of the 2.2Ma subduction erosion event. The onset of subduction of an aseismic ridge as occurred at 2.2Ma in southern Costa Rica is a fairly common tectonic event along a subduction margin. We suggest that similar rapid pulses of subduction erosion may punctuate the evolution of many margins, contributing disproportionately to crustal recycling at subduction zones. The (poorly) preserved geologic record of paleoforearcs needs to be reassessed with this mechanism in mind. It also implies that continental forearc material may be significantly consumed during short local bursts along a subduction margin, and furthermore, that margins abutting regions of frequent subduction of aseismic ridges, like the regions in the Western Pacific where the Darwin Rise currently subducts, should face disproportionate pulses of future subduction erosion and forearc recycling.