The Effects of Fault-Zone Cementation on Groundwater Flow at the Field Scale.

Fault zones are an important control on fluid flow, affecting groundwater supply, contaminant migration, and carbon storage. However, most models of fault seal do not consider fault zone cementation, despite the recognition that it is common and can dramatically reduce permeability. In order to study the field-scale hydrogeologic effects of fault zone cementation, we conducted a series of aquifer pumping tests in wells installed within tens of meters of the variably cemented Loma Blanca Fault, a normal fault in the Rio Grande Rift. In the southern half of the study area, the fault zone is cemented by calcite; the cemented zone is 2-8 m wide. In the center of the study area, the cemented fault zone is truncated at a buttress unconformity that laterally separates hydrostratigraphic units with a ∼40X difference in permeability. The fault zone north of the unconformity is not cemented. Constant rate pumping tests indicate that where the fault is cemented, it is a barrier to groundwater flow. This is an important demonstration that a fault with no clay in its core and similar sediment on both sides can be a barrier to groundwater flow by virtue of its cementation; most conceptual models for the hydrogeology of faults would predict that it would not be a barrier to groundwater flow. Additionally, the lateral permeability heterogeneity across the unconformity imposes another important control on the local flow field. This permeability discontinuity acts as either a no-flow boundary or a constant head boundary, depending on the location of pumping.

The Role of Fault-Zone Architectural Elements on Pore Pressure Propagation and Induced Seismicity.

We used hydrogeologic models to assess how fault-zone properties promote or inhibit the downward propagation of fluid overpressures from a basal reservoir injection well (150 m from fault zone, Q = 5000 m3 /day) into the underlying crystalline basement rocks. We varied the permeability of the fault-zone architectural components and a crystalline basement weathered layer as part of a numerical sensitivity study. Realistic conduit-barrier style fault zones effectively transmit elevated pore pressures associated with 4 years of continuous injection to depths of approximately 2.5 km within the crystalline basement while compartmentalizing fluid flow within the injection reservoir. The presence of a laterally continuous, relatively low-permeability altered/weathered basement horizon (kaltered layer = 0.1 × kbasement ) can limit the penetration depth of the pressure front to approximately 500 m. On the other hand, the presence of a discontinuous altered/weathered horizon that partially confines the injection reservoir without blocking the fault fluid conduit promotes downward propagation of pressures. Permeability enhancement via hydromechanical failure was found to increase the depth of early-time pressure front migration by a factor of 1.3 to 1.85. Dynamic permeability models may help explain seismicity at depths of greater than 10 km such as is observed within the Permian Basin, NM.

Reading a 400,000-year record of earthquake frequency for an intraplate fault.

Our understanding of the frequency of large earthquakes at timescales longer than instrumental and historical records is based mostly on paleoseismic studies of fast-moving plate-boundary faults. Similar study of intraplate faults has been limited until now, because intraplate earthquake recurrence intervals are generally long (10s to 100s of thousands of years) relative to conventional paleoseismic records determined by trenching. Long-term variations in the earthquake recurrence intervals of intraplate faults therefore are poorly understood. Longer paleoseismic records for intraplate faults are required both to better quantify their earthquake recurrence intervals and to test competing models of earthquake frequency (e.g., time-dependent, time-independent, and clustered). We present the results of U-Th dating of calcite veins in the Loma Blanca normal fault zone, Rio Grande rift, New Mexico, United States, that constrain earthquake recurrence intervals over much of the past ∼550 ka-the longest direct record of seismic frequency documented for any fault to date. The 13 distinct seismic events delineated by this effort demonstrate that for >400 ka, the Loma Blanca fault produced periodic large earthquakes, consistent with a time-dependent model of earthquake recurrence. However, this time-dependent series was interrupted by a cluster of earthquakes at ∼430 ka. The carbon isotope composition of calcite formed during this seismic cluster records rapid degassing of CO2, suggesting an interval of anomalous fluid source. In concert with U-Th dates recording decreased recurrence intervals, we infer seismicity during this interval records fault-valve behavior. These data provide insight into the long-term seismic behavior of the Loma Blanca fault and, by inference, other intraplate faults.