Triangulation-to-GPS and GPS-to-GPS geodesy in Trinidad, West Indies: Neotectonics, seismic risk, and geologic implications

We estimated horizontal velocities at 25 sites first surveyed in a 1901-1903 British Ordnance Survey triangulation and then resurveyed with GPS in 1994-1995 to identify Trinidad's principal active on-land faults, quantify fault-slip-rates, and test for elastic locking. Our best-fit single-fault elastic dislocation model put 12 ± 3 mm/yr of dextral strike-slip on the Central Range Fault (1-2 km locking depth), an apparently aseismic active fault. The estimated motions also showed statistically insignificant horizontal motion (2.2 ± 1.8 mm/yr of right-slip; 2.7 ± 2.0 mm/yr of N-S shortening) on the eastward on-strike extension of the El Pilar Fault, known to be the active transform fault in Venezuela. Repeat GPS measurements made between 1994 and 2005 at two sites spanning the island north to south showed a 14 ± 3 mm/yr eastward (plate-motion-parallel) dextral velocity differential, consistent with our best-fit historic (1901-1995) fault-slip-rate. Paleoseismology trenching demonstrates that the Central Range Fault cuts <5000-year-old sediment and is capped by ∼550-year-old sediment, suggesting that it may be locked and may have ruptured at least once during this time interval. About ∼5 mm/yr of slip could be taken up on the Los Bajos Fault and additional faults in the offshore south of Trinidad. The existing 1901-1995 and 1994-2005 geodetic data alone cannot resolve whether the Central Range Fault is essentially creeping (≤1-2 km locking depth) or locked to a more standard depth of 10 km.