Can we hide an active fault within a geodetic network? Yes, we can.

Maria Beatrice Magnani

Submitted August 15, 2018, SCEC Contribution #8637, 2018 SCEC Annual Meeting Poster #130

Comparison of deformation rates around the New Madrid seismic zone (NMSZ), central U.S., indicates an apparent inconsistency between geologic and geodetic rates. In intraplate regions such as the central U.S., geologic rates, inherently longer-term, might be better indicators for future seismic activity than GPS data, which sample a shorter timeframe. Here I show how a Holocene active fault might go undetected within the geodetic network monitoring the central U.S. seismically active area. High resolution marine seismic reflection data acquired along the Mississippi river image a previously unknown NW-SE striking north-dipping fault displacing the base of the Quaternary alluvium by 15m with reverse sense of movement, and consistently deforming the Tertiary, Cretaceous and Paleozoic formations. The fault crosses the river south of the NMSZ, between Caruthersville, MO and Osceola, AR. Although the fault appears aseismic, geomorphological evidence from river channel sinuosity suggests that the latest slip occurred in the last 200 years. Here I explore whether movement along this fault can be detected by the GNSS Array for Mid-America (GAMA) network, which is specifically designed for tectonic studies in and around the NMSZ. GAMA stations straddle the fault with the closest stations located ~30 km to the north and ~50 km to the south. I model the amount and extent of surface displacement due to a shear fault in an isotropic half-space (Okada, 1992) and compare the results against the GAMA data as processed by Boyd et al., 2015. The model is applied to a 20 km-long fault that dips to the north ~60º, strikes N125º, and extends to 10 km depth. Both dip slip and strike slip dislocations are tested. Results show that a dislocation rate of 4 mm/yr is required to produce a detectable surface signal. Based on recent coring of the floodplain alluvium south of the study area, the age of the oldest Quaternary deposit along the most recent meander belt is 14.3ka±0.5ka. If fault slip has been steady state during the Holocene, then a slip rate of 1mm/yr is reasonable for the imaged fault, resulting in undetectable displacement rates at the nearest geodetic stations. My results indicate that it is possible for a fault of substantial dimension (~200 km2 area) with a Holocene slip rate exceeding 1mm/yr to hide within the geodetic network. Undetected by the GPS stations and currently aseismic, such fault could therefore represent an unrecognized seismic hazard.

Key Words
Intraplate deformation, geologic strain rates, geodetic strain rates

Citation
Magnani, M. (2018, 08). Can we hide an active fault within a geodetic network? Yes, we can.. Poster Presentation at 2018 SCEC Annual Meeting.


Related Projects & Working Groups
Tectonic Geodesy