Pressure data recorded by long-term subseafloor observatories (CORKs) are a useful tool for understanding the state of the crustal hydrologic system. In Middle Valley, a sedimented rift at the northern end of the Juan de Fuca Ridge, thickly sedimented basaltic crust hosts an array of hydrologic regimes that have been monitored continuously by two CORKs since 1996.
This dissertation analyzes both recent trends in borehole and seafloor pressures, as well as several older datasets, in concert with local seismicity, physical properties of the crust, and hydrothermal circulation in an effort to understand the region’s ongoing dynamic eruptive cycle and hydrogeologic connectivity.
Middle Valley is a well-sedimented axial rift located at the northern end of the Juan de
Fuca Ridge, just south of the Sovanco Transform and the Nootka Fault . It is blanketed by a thick layer of sediment that varies from 200m to 2 km, composed of Pleistocene age turbidite deposits and hemipelagic sediment. The valley lies roughly along the axial strike of the unsedimented Endeavour segment located to the south. This, along with a relatively small amount of post-Pleistocene subsidence, indicates that Middle Valley was a primary center of spreading until recent times (Davis and Villinger, 1992).
ODP Site 857:
This site is located at the southern end of Middle Valley and was chosen as a representative ‘sealed’ region of basement through which hydrothermal flow occurs (Shipboard Scientific Party, 1992a). It lies on a back-tilted fault block 1.5 km east of the sediment-buried normal fault that forms the structural center of Middle Valley and lies within the thermal anomaly that runs parallel to the rift axis (Davis and Villinger, 1992). Sediment thickness at Site 857 is approximately 400-500 m. Local heat flow is approximately 1 W/m2 (Davis and Villinger, 1992).
THE CORK PROGRAM
Each CORK pressure data logger is programmable: the sampling rate can be changed if necessary. Following the initial deployment in 1991, Hole 858G was set to record pressure every ten minutes. This was changed to once per hour during the first data download in 1991 (Davis and Becker, 1994). Sampling rate at Hole 857D was also adjusted from once per hour to once every ten minutes for ~2 years of monitoring (2003-2005). The newest generation of data loggers is capable of up to 1 Hz sampling and is compatible with existing cabled seafloor observatory platforms (i.e. NEPTUNE Canada).
In addition to tidal loading experienced by the formation, local and distal seismicity can
perturb the background pressure state of the hydrologic system. This has been observed in the pressure records of many CORKs. In the case of Middle Valley, the effects of local earthquake swarms were recorded in 1991, 2001 (Davis et al., 2004), and 2004 and a distal swarm was recorded in 1999 (Davis et al., 2001). The effect of an earthquake swarm on a borehole pressure record is a function of the CORK’s location relative to the strain field created by the earthquake.
REMOVAL OF OCEANOGRAPHIC EFFECTS WITH MATLAB
T_tide MATLAB Code:
T_tide is a freely available MATLAB code used for predicting tides based on a given periodic dataset (Pawlowicz, Beardsley, & Lentz, 2002). Given inputted data and a known sampling rate, the code will generate a “best fit” tidal signal for the given values by calculating a set of tidal constants. An example of the tidal constants output from the code.
A SUBSEAFLOOR OBSERVATORY RECORD OF SEAFLOOR “UPLIFT” IN MIDDLE VALLEY, 2005-2010
Middle Valley has been the subject of two ODP expeditions (139 and 169) and contains a suite of hydrologic regimes: active hydrothermal venting from overpressured basement with underpressures at shallow levels at Dead Dog field (ODP Site 858/1036), a massive sulfide mound at Bent Hill (ODP Site 856/1035) that is also actively venting from overpressured basement, seawater recharge at boundary faults (ODP Site 855) and Site 857, which is located at the southern end of Middle Valley and was chosen as a representative sediment-sealed region of basement through which hydrothermal flow occurs (Shipboard Scientific Party, 1992a).
Data: Pressure Record Description and Significant Events:
The CORK at Hole 857D has been visited periodically since 1996 by DSV Alvin and the ROV JASON for the purpose of downloading pressure data from the internal logger. In addition, a new pressure monitoring system with a data logger was installed on the CORK’s formation fluid sampling port in June, 2010.
The new logger is currently recording one concurrent borehole and seafloor pressure sample per minute, and is capable of 1Hz data sampling when supplied with outside power (eg. NEPTUNE Canada cable).
MODELING THE POTENTIAL SEAFLOOR UPLIFT AT 857D
The Mogi Model:
To model the source of an uplifted surface, be it subaerial or submarine, it is important to
know the shape and lateral extent of the surface, as well as vertical and horizontal displacements in time. Since our knowledge of the lateral extent of uplift in Middle Valley is extremely limited, we must begin by considering a simple model to approximate the observed vertical change.
1996 CROSS-HOLE EXPERIMENT BETWEEN HOLES 857D AND 858G
In 1996, during the replacement of the CORKs in Holes 857D and 858G during ODP Leg 169, drilling operations were ordered such that an experiment be conducted to test the hydraulic connectivity of basement between Holes 857D and 858G. First, the CORK in Hole 858G was fully replaced. Then, while replacing the damaged CORK in Hole 857D, seawater was circulated down to the high permeability zone previously observed at 610 mbsf with the purpose of creating a pressure transient that would propagate to Hole 858G, and which would be recorded by the CORK.
Holes 857D and 858G:
While both sites are located in southern Middle Valley and are only separated by ~1.6 km,
the difference in hydrologic regimes between these two sites is rather remarkable. Hole 857D is the deeper ‘background’ hydrothermal site, exhibiting a ~300kPa formation underpressure relative to seafloor hydrostatic, has a hydrologic basement defined by interbedded sediments and basaltic sills, and contains several zones of extremely high permeability that are inferred to be fault zones deep in the hole (Langseth and Becker, 1994).
Post-defense analyses of Hole 857D high-resolution logger data downloaded in September 2012 indicate that the downhole logger’s seafloor record may in fact be an artifact of gauge failure or drift, as opposed to an actual uplift signal, despite the corroborating Alvin depth data.
This type of artifact has not been observed previously and we do not yet know the cause. Also, the apparent depth decrease recorded by the Alvin pressure gauge is especially perplexing, but we know nothing about the Paroscientific sensor installed on the submersible or the characteristics of its drift over time so it is unclear whether or not this decrease is real.
On the other hand the borehole data have been verified, as the trends be tween both loggers’ records match for the time period of data overlap (2010-2012). As mentioned in Chapter 6, if the uplift signal is not real then we must consider alternative theories for what caused the dramatic change in hydrothermal venting in Middle Valley between 2008 and 2010, as well as the 2005-2012 drop in formation pressure.
Source: University of Miami
Author: Katherine E. Inderbitzen