Subseafloor hydrogeology is a major component of ongoing science programs and initiatives including IODP, MARGINS, RIDGE, and Neptune. Pore fluid pressure is the most fundamental physical parameter for determining the current state of hydrogeologic processes such as fluid and energy flow and the state of stress within marine sediments. It is also a primary indicator of transient tectonic processes ranging from earthquakes to aseismic creep. In spite of this, monitoring of pore pressure in marine sediments by the geoscience community has been limited to a few isolated studies and to a few instrumented ODP boreholes. However, recent advances in a number of technologies have contributed to make such measurements more accurate, reliable, and cost effective today than they once were and hence driven a renewed interest in developing pore pressure instruments for the deep sea (Earl Davis, Louis Geli, Warner Brückmann, Achim Kopf, pers. comm.). We propose here to develop and build a prototype pore pressure instrument package which can be deployed by freefall from a ship. It will build on knowledge gained from use of its predecessors, the PUPPI (Pop Up Pore Pressure Instrument) [Schultheiss, 1990; Schultheiss et al., 1985] and on the more recent SAPPI (Satellite-Linked Autonomous Pore Pressure Instrument) [Kaul et al., 2004] and IFREMER piezometer [Sultan et al., 2007]. A key feature will also be the ability to transfer data via acoustic modem to a surface ship or moored buoy (e.g. ORION) or to be connected to a seafloor cable network (e.g. MARS, NEPTUNE, ESONET) for real time monitoring. The ultimate goal of this development program, though outside the scope of this proposal, is to create a collection of such instruments to add to our existing marine hydrogeology facilities.

Linking Shallow Measurements to the Processes of Interest

Of what value are a series of shallow measurements if we have no means to extrapolate them to the larger picture/processes we are studying? Modern subaerial hydrogeology is dominated by modeling and its success is principally due to having good constraints on the boundary conditions and a significant number of tie points within those bounds. Marine hydrogeology, however, suffers from a critical lack of such constraints and tie points due to the extreme difficulty and high cost of obtaining those essential data. In this context the broadest applicability of the PUPPI-II is to provide the near-seabed boundary conditions to such models.

Monitoring pore pressure is certainly not limited to hydrologic modeling, however. For example, long term records of pressure at CORK installations at Nankai and Costa Rica exhibit pressure transients coincident with events recorded by on-land seismometers and/or GPS arrays [Davis et al., 2006; Davis and Villinger, 2006]. While technical difficulties with the CORK installations have precluded a quantitative interpretation of the Nankai events [Davis, in prep], such records have the potential to greatly increase our knowledge of the strain distribution associated with seismic and aseismic deformation through the seismic cycle. Dilation and compression associated with tectonic activity occurs not only at the depths monitored at boreholes but also throughout the sediment column and while pressure changes would be attenuated at shallow depths due to differences in matrix elastic properties, strain should still be detectable from shallow pore pressure monitors (see below). For example, an array of fluid flow meters detected a similar pattern of inflow and outflow events on the outer rise off Costa Rica correlated with earthquake activity [Tryon and Brown, 2005]. Such short-term (days to weeks) flow events can only be associated with shallow (<5-10 m) strain due to the rapid increase in hydraulic impedance with greater depth. These flow meters were merely recording the flow response due to shallow pore pressure changes such as would be monitored by the PUPPI-II.

Previous and Current Seafloor Piezometers

While a few seafloor piezometers have been built in the past [Bennett et al., 1985; Carpertier and Verdonk, 1986; Richards et al., 1975], the principal instrument for making seafloor pore pressure measurements was the Pop Up Pore Pressure Instrument (PUPPI) [Schultheiss, 1990; Schultheiss and McPhail, 1986; Schultheiss et al., 1985]. The PUPPI was designed at the former Institute of Oceanographic Sciences, UK, to measure residual pore pressures at a depth of 4-6 m in soft sediments with a resolution of ~10 Pa. It was designed as a free-fall instrument that is ballasted to penetrate a range of sediment types in water depths of up to 6,000 m. One or more pressure ports located at intervals on the lance are connected to one or more differential pressure transducers with the other side open to the sea. A valve sequentially opens other ports to the transducer when more than two ports are utilized. Pressure data is stored on a data logger at intervals as short as 2 s. An accelerometer records the penetration event allowing penetration depth and some sediment properties to be determined. Recovery was accomplished via an acoustic release mechanism which cuts the tubes to the ports and releases the instrument package from the lance and ballast which remain on the seafloor while the instrument floats to the surface for recovery. Subsequent to their initial development, a number of PUPPIs were built and used commercially by their inventor, Peter Schultheiss, who had established a private company, Geotek Ltd. During this time numerous upgrades and variations on the basic PUPPI design were accomplished including separating the lance and recorder package for use by DSRV Alvin for precise placement on seeps. They were used in a number of studies in different environments by both academics and industry [Schultheiss and McPhail, 1986; Shipboard Scientific Party, 1998]. Unfortunately the PUPPIs are no longer available due to the age of both the design and components. This vacancy has, however, led to an opportunity for the development of new instrumentation using the experience gained during the initial PUPPI development and operations.

In the past few years a renewed interest in this type of instrument has prompted at least four new deep-sea piezometers. As part of the Keck program, the Canadian Pacific Geosciences Center developed a three-port piezometer that was deployed on the east flank of the Juan de Fuca plate to monitor for pressure transients associated with tectonic events (Earl Davis, pers comm). This free fall instrument utilized 4 absolute pressure transducers and a CORK data logger. The frame and probe are expendable and the instrument package was recovered by Alvin in September 2005. Similarly, Heiner Villinger at University of Bremen has led the development of an expendable seafloor piezometer that sends its data via two pop-up data buoys to an Iridium satellite based data link [Kaul et al., 2004]. While the deployment of the instrument at a mud volcano in the Gulf of Cadiz was a complete success, three earlier deployments failed to send data and its data capacity is small, limiting its usefulness for deployments longer than a month. Another instrument has been developed by Nabil Sultan and Louis Geli at IFREMER, France [Sultan et al., 2007] that is very similar to the PUPPI. This system is currently employed in a long-term industry sponsored slope stability study. A prototype of another variation on this theme has been developed by Warner Brückmann’s group at GEOMAR [Brückmann et al., 2005]. All of these instruments, however, rely on the same differential pressure transducer which has a relatively low resolution. Clearly there is strong current interest in this type of research with an emphasis on long-term monitoring. We have good working relationships with Davis and the Kopf/Villinger team at Bremen and have ongoing collaborations with Geli and Brückmann and expect that there will continue to be a mutually beneficial exchange of technology among the different research groups.

Practical Design Considerations

The primary requirement of a seafloor piezometer is that it accurately measure the ambient pressure differential between the seafloor and some point beneath the sea floor. To accomplish this a probe of some sort must be inserted into the sediments as deep as practical. This will inevitably cause some disturbance of the ambient pore pressure as sediments are compacted. This pressure pulse decays at a rate that depends on the properties of the sediment and can take many hours to days. This extended time frame for making measurements has led to the need for an autonomous instrument package with some means of returning the data, and preferably at least the expensive parts of the instrument, back to the surface.

There are many potential problems associated with the accurate measurement of pore pressure in marine sediments. 1) The need for high precision and low compliance along with very high ambient pressure makes pressure transducer choice critical. Absolute pressure transducers have, until very recently, lacked the resolution necessary for this work but differential pressure transducers can easily be overpressured and often have relatively high compliance. 2) Pressure transducers and their associated volumes (tubes, valves) are sensitive to extremely small temperature changes so must be well insulated or accurately compensated. 3) Small differences in water density between the ambient pore water and seawater can lead to an apparent differential pressure. A one point change in salinity would produce an apparent differential pressure equivalent to about a 3 mm/yr flow rate. A representative gravity core should be taken, when possible, both for pore fluid salinity measurements and for testing penetration potential. Samples from this core could also be used for physical properties testing to better constrain the strength and permeability. 4) For dynamic measurements, the time lag associated with a pressure measurement system can be problematic. The pressure transducer and associated valves and tubing should have as little volume as possible and negligible compliance and, for shallow water work, it is also critical that there be no free gas present in the system. Of course the sediments themselves constitute a part of the system and so measurements in gassy sediments pose considerable problems for dynamic measurements for which we are aware of no solution. 5) In the presence of bottom currents there is a non-uniform and varying pressure field surrounding any ocean bottom instrument which can lead to incorrect measurements of bottom pressure. This can be remedied by minimizing instrument size and/or having the pressure port at some distance from the instrument.

NEXT GENERATION PUPPI

The new instrument, PUPPI-II, will be based on the successful design philosophies of the original instrument but significantly updated based on new technologies and nearly 20 years experience. Nominally, the instrument will consist of two sections: a lower section consisting of the disposable weight stack and lance with pressure ports and an upper instrument section which houses the pressure transducers, acoustic release, logger, acoustic modem, any optional equipment, and the floatation and recovery aids. Upon acoustic command the hydraulic and mechanical links between the two sections are disengaged and the upper section floats to the surface for recovery. This system has proven reliable and robust and, aside from significant upgrades to components, will remain little changed from the original basic configuration.

One significant new feature will be the incorporation of an acoustic modem for data transfer. Upon deployment the user will quickly be able to determine if the deployment was successful, i.e., vertical, sufficient penetration, and collecting data. For longer deployments this allows one to download the data and look at it prior to making a decision regarding recovery. If an instrument is currently collecting interesting data, leaving it deployed would often be desired. This also allows the instruments to be part of an instrument array communicating with a buoy and, through satellite link, to the internet. In areas where a cable network is in place, i.e., the Neptune and Mars programs, the modem would be replaced by a link to the cable.

Our key design philosophy is that the instrument be reliable, inexpensive, and modular. These last two are desired because we plan to be able to ultimately create enough of these instruments to be able to deploy arrays. Making them modular allows optimizing each instrument for the specific requirements of the deployment. Specifications may change somewhat during instrument development, however we currently expect the PUPPI-II to have the following specifications: