Obliquely convergent plate motion and its relation with forearc sliver movement, El Salvador volcanic arc

The magmatic arc in El Salvador is interpreted to result from the subduction of the Cocos plate underneath the Caribbean plate along the Middle America trench. In addition, El Salvador contains a fore-arc sliver that moves 11 mm/yr westward relative to the back-arc. Well-defined strike-slip faults along the magmatic arc accommodate forearc sliver motion, but are offset at several locations by en echelon pull-apart step-overs with abundant normal faults. All basaltic-andesitic magmatic centers (San Miguel, San Vincente, San Salvador, Santa Ana) are located within these step-overs, while the two major rhyolitic calderas (Ilopango, Coatepeque) occur directly along the strike-slip faults. There are two puzzling aspects about the strike-slip tectonism. First, a silicic, shallow magma body that intrudes the San Miguel fault zone (part of the El Salvador fault system) was emplaced syn-tectonically (sigmoidal field and magnetic foliations, subhorizontally plunging magnetic lineations and dextral shear at the microscale). Within the dextrally sheared portion of the intrusion, an obsidian band with a 40Ar/39Ar age of 7.46 Ma indicates that dextral strike-slip tectonism in the Salvadoran arc has been an ongoing process for ~7.5 Ma. This casts significant doubt on whether Cocos ridge subduction (that started ~1 Ma ago) is the cause of the ongoing forearc movement. The potentially more significant problem is that the fore-arc sliver in El Salvador moves 11 mm/yr westward relative to the back-arc despite a nearly orthogonal angle of convergence (with a convergence rate of ~77 mm/yr) near El Salvador and absence of significant frictional coupling along the subduction interface. Further, GPS indicates that the Nicaraguan and Salvadoran forearcs define a semi-rigid sliver moving at nearly the same trench-parallel rates despite along-trench changes in the subduction angle. Consequently, it is tempting to attribute the movement of both forearc slivers to Cocos ridge subduction. However, as mentioned above, initiation of strike-slip movement during collision of the Cocos ridge appears incompatible with Miocene strike-slip deformation that occurs. These examples serve to highlight the difficulties in understanding forearc movement in obliquely convergent systems, even in relatively modern belts where the plate motions are well constrained.