Polaris facilities for studies of earthquake hazards and continental dynamics Part II

Thomson 16751 POLARIS MFA proposal

POLARIS facilities for studies of earthquake hazards and continental dynamics - Part II
1. Description of the Portable Observatories for Lithosphere Analysis and Research Investigating Seismicity (POLARIS)

POLARIS is a Canadian university-government-industry geophysical consortium focused on investigation of the structure and dynamics of the Earth's lithosphere and predicting ground motion due to earthquakes. The original POLARIS concept is based on combined seismological and magnetotelluric (MT) studies of the Earth using remote observatories transmitting data over a VSAT satellite network and using Libra seismograph systems manufactured by Nanometrics Inc. of Ottawa, Ontario. An extension of this concept is currently being developed to upgrade some of the observatories with geodetic sensors measuring crustal deformation using the Global Positioning System (GPS).

Currently, 67 broad-band seismographs are installed at sites in three arrays in ON, BC, and the NWT and 30 MT instruments are being progressively deployed at these locations. They transmit continuously to central data acquisition hubs in London and Ottawa. These hubs both receive data and can control remotely the functions of each station. The POLARIS live data are ported over the Internet to the facilities of the National Earthquake Hazards Program (NEHP) of the Geological Survey of Canada (GSC) in Ottawa, to be processed and archived along with the GSC’s own data. All POLARIS data are rapidly made available to Canadian and foreign researchers via an AutoDRM, a web-based standard international data request form. The stations themselves will be moved as scientific projects progress, with MT analyses often requiring shorter periods of recording at a given site making these instruments somewhat more mobile.
POLARIS Observatory

Fig. 1. A typical POLARIS observatory is powered by 6 to 12 solar panels and is installed on solid bedrock. The seismometer vault is normally a few tens of meters from the VSAT antenna. Storage batteries, capacity about 15 days, are housed in a sealed battery box. A 12-channel GPS link provides precise timing for VSAT and seismic data. Provisions are made to add other, such as MT and geodetic, sensors to the system. Data from POLARIS observatories are available on the Internet within 5 seconds of recording.

The POLARIS infrastructure is unique in design, operation and data quality control. Most POLARIS observatories are powered by photovoltaic energy and demonstrate the functionality of the VSAT equipment in harsh Canadian environments. Seasonally-inaccessible northern stations can be remotely interrogated and problems diagnosed, often even corrected. For example, seismometer tilting can be compensated for electronically. The telemetry also allows rapid posting on the POLARIS website of, for example, seismograms from all globally-significant earthquakes of magnitude 5 or greater or all local and regional Canadian events only minutes after occurrence.

The POLARIS infrastructure is designed in a scalable and modular form. Additional remote geophysical observatories can and will be installed and accommodated with the current infrastructure. POLARIS hubs can readily expand to increase data type and volume. Other sensors, meteorological for example, can be added to POLARIS observatories and their data piped to the Internet for the user community.

Data Flow

Fig. 2. Typical POLARIS Data Flow starts at one of the two Hubs, UWO in this example. Data are stored locally at UWO, but final archive copies are made at NEHP, which provides AutoDRM service to the user community. Copies of the data are also available direct to the user in summary format from the POLARIS Website at the Continental Geoscience Division (CGD) of the GSC.

POLARIS Website

The POLARIS website (www.polarisnet.ca) is updated regularly. No later than 30 minutes after any local or global earthquake is reported, summary seismograms of the event at some key POLARIS observatories are available. These are useful for both research and education, and are used by POLARIS researchers to select events for further analysis. A copy of the digital data record can also be downloaded from the website. In addition to the seismograms, various graphs and tables are generated and updated daily for the purpose of monitoring the infrastructure and the state of health of the network. Other key details of the infrastructure are also posted on the website, such as the precise locations of the remote observatories and the instrument responses or sensitivity characteristics.

Data supplied to other users

The AutoDRM (Data Request Manager) provides a convenient automated route for users to access the full POLARIS data and allows tracking of users and data volumes. Over the past 8 months an average of 2.5 GB of data was requested each month (see reports at http://www.seismo.nrcan.gc.ca/nwfa/autodrm/polaris/). In addition to POLARIS members and other Canadians, users have come from the Netherlands, the U.S (government and universities), Switzerland, the U.K., Russia and Germany.

Fig. 3 . POLARIS maps, showing the arrays

in ON, BC and NWT as they are presently configured.

Outreach

The POLARIS team recognizes that communicating basic Earth Science information to the public is an important component of outreach activities. We believe that having an informed public has the double benefit of 1) reducing irrational worry that comes from lack of familiarity and understanding of earthquakes and 2) making the community receptive to disaster mitigation activities that can further promote well-educated, safe communities. We are currently working with staff at the Sudbury Science North museum to display in their "Dynamic Earth" exhibit the real-time ground motions recorded by the POLARIS seismograph in the Sudbury Basin. This prototype system could then be made available to other museums, and perhaps developed as an in-classroom adjunct to our proposed web-based educational resources for students, teachers and the general public (NSERC Promoscience application pending).

2. Research activites supported by POLARIS
2.1 Diamonds and ancient continental roots

Exploration geophysicists increasingly consider the passive recording of distant (teleseismic) earthquake signals as an alternative to active seismic profiling, because of the environmental and cost concerns in using large explosions. Specific application of teleseismic studies to diamond exploration was attempted previously in South Africa (e.g. James et al., 2001). However, the search for geological structures related to diamond formation, their preservation and their transport to the surface in kimberlite pipes requires the greater focus and resolution that we are only now able to contemplate. The nascent Canadian diamond industry typically uses global seismic tomography to map continental roots, followed by costly and labor intenstive petrological surveying for indicator minerals found in glacial deposits in its quest to locate areas of high diamond prospectivity. POLARIS array studies seek an alternative method to locate diamonds beneath an area, particularly if glacial deposits are either absent or too complex.

Various analytical techniques using teleseismic sources and widely-spaced arrays have in some cases reached moderate levels of sophistication compared with methods used in the petroleum exploration industry. Relevant techniques include: (1) searching for structural fabric and its orientation in the mantle using the differential propagation velocities of shear-wave components that pass through the Earth’s core (Silver & Savage, 1994) ; (2) locating discontinuities in the mantle that disrupt seismic waves and migrating scattered seismic energy back to the scattering structure (Bostock, 1998; Bostock et al., 2001); (3) mapping 3-D variations in either P- or S-wave velocities using tomographic methods (VanDeCar, 1991); and (4) determining velocity variations versus depth using long-period surface waves whose depth of penetration is proportional to their wavelength (Li, Forsyth & Fischer, 2003). All of these techniques can be independently and effectively applied to the same set of POLARIS array data to identify major mantle structures.

Fig.4. Seismic discontinuities identified beneath the central Slave craton. POLARIS stations are shown as black triangles. Coloured bars represent discontinuities identified beneath individual stations. Background shows velocity variations asdetermined by tomographic modeling with the method and stations used by Bank et al. (1999), here supplemented by the POLARIS stations. The rock column at right is based on geochemical analysis of garnets in kimberlite samples.

The POLARIS seismic array in the NWT has already revealed distinctive layers in the Slave craton (Fig. 4) that correlate in depth with a strong electrical conductor revealed by MT surveys (Jones et al., 2001; see also below), and with petrological layering derived from geochemical analysis of mantle garnets (Bank et al., 2000; Snyder et al., 2003). This layering appears better developed than elsewhere in the world, such as in South Africa. Further analysis of the POLARIS NWT array data will reveal details of both lateral and vertical variations in mantle structure with which to substantiate preliminary evolutionary models. In these models, underthrust Archean oceanic crust introduced carbon to pressures sufficient to create diamonds and the subsequent stabilization of the craton preserved these diamonds until kimberlite eruptions carried the diamonds to the surface (Davis et al., 2003). Recent precision dating of individual kimberlite eruptions indicates that they were erupted along fractures related to both old inherited structures and recent stress fields (Lockhart et al., 2003).

Elsewhere, the northern extension of the POLARIS Ontario array covers the western part of the Superior Province, the world's largest exposed Archean craton and a region where economic diamond deposits were recently discovered and mines are planned. All the analytical techniques used on NWT array data, to the extent they are successful, will be applied to the Ontario array data.
The electromagnetic MT method reveals electrical resistivity structure and fabric of the upper mantle that can also play an important role in development of diamond exploration models. Mantle resistivity is controlled dominantly by minor constituents such as grain-boundary carbon and graphite, free ionic fluids, partial melts or hydrous minerals. Thus resistivity structure is independent from and complementary to the elastic structure derived from seismic studies. The MT method maps the mantle resistivity using deep-penetrating, low-frequency electromagnetic signals created by the Earth’s magnetic field during magnetic storms.
Earlier MT responses measured by Lithoprobe revealed a remarkable anomaly in electrical conductivity in the sub-continental lithospheric mantle in the central Slave craton (Jones et al., 2001). This striking NE-SW anomaly forms a spatially-confined region of extremely-low resistivity, with its upper surface at depths of 80-120 km. The conductor correlates spatially with a geochemically-defined ultra-depleted harzburgitic layer. In the western Superior Province,

resistivity structure suggests the presence of a relatively deep continental keel in some locations and a conductive lithospheric mantle, similar to that observed in the Slave Province, in other locations.

The new generation of POLARIS MT soundings in the Slave and western Superior cratons will further define the mantle resistivity structure. A particularly exciting part of the project involves MT recordings made continuously for several years at selected Slave POLARIS sites. The responses derived at these sites will reveal resistivity structure as deep as the 410 km seismic discontinuity. Having the co-located seismic and MT data will also allow comparison of electrical-property fabric or anisotropy with elastic textural information drawn from the teleseismic data, leading to powerful constraints on geodynamical causes.
2.2 Seismic risk in Eastern Canada

The southern Great Lakes area is home to more than 20 million people in Canada and the U.S. Although not as seismically active as other areas of eastern North America, this region has experienced widely-felt earthquakes, such as the 1998 M=5.4 Pymatuning earthquake. The seismicity in this region has been characterized by Stevens (1994) as “intermittent scattered”, although other workers have attributed greater significance to apparent linear trends of seismicity in the historical catalog (e.g. Wallach et al. 1998). The nature of this seismic activity and its possible relationship to ancient fault zones is poorly known. This uncertainty, as well as uncertainties in the magnitude-recurrence parameters for local sources, contributes significantly to the overall uncertainty in the local hazard assessment (Adams & Halchuk, 2003)

The core Ontario POLARIS array covers the southern Ontario region, roughly from Ottawa to Windsor and from Niagara to Georgian Bay, with additional stations in northern Ontario funded through the FedNor project. Stations in Quebec are planned under a second proposal to CFI (POLARIS2, submitted Spring 2003). The 36 POLARIS stations operating in eastern Canada will provide greatly improved definition of the spatial and depth distribution of earthquakes and their occurrence rates, as well as crucial information on the generation and propagation of the seismic waves that cause damage to engineered structures.

The improved POLARIS monitoring capability has substantially reduced hypocentral location uncertainties for earthquakes, revealing linear trends of epicenters near the west end of Lake Ontario (Dineva et al., 2003). Whereas less well-constrained historical data (largely based on felt reports) place almost all events onshore, the emerging pattern of seismicity clearly shows that southern Ontario events are concentrated mainly beneath Lake Ontario or in adjacent areas to the south. This spatial relationship led Mereu et al. (2002) to suggest that the seismicity is moderated by the lubrication of fault zones by fluid circulation from the lake.

An innovative approach is currently under development at UWO to deconvolve the teleseismically-derived local site response, caused by resonant amplification in near-surface layers. This approach shows promise to improve magnitude estimates and the quality of arrival time-picks for event location. Other planned research at Canadian universities will further focus on characterizing regional earthquake source parameters, and determining what spatial relationships exist between contemporary seismicity and ancient structural elements in the underlying basement. Geodetic measurements planned under our POLARIS2 proposal will provide necessary constraints for geodynamic and structural/tectonic modelling, providing an improved physical basis for seismic hazard evaluation in the region.

POLARIS data will help to reduce uncertainty in engineering ground motion parameters used in design and retrofit of engineered structures by providing time histories for input into engineering analyses. Such histories, when simulated using limited knowledge of regional earthquake source and wave propagation characteristics, led in the past to large uncertainties in the earthquake shaking loads. A compromise between using simulated and real earthquake records for design will use the POLARIS data to modify real earthquake recordings from small events to model the shaking that would occur during large earthquakes.

2.3 Seismicity and structure of the Cascadia subduction fault

In the past 3 years, several landmark scientific discoveries concerning subduction zone structure and dynamics have been made in geophysical studies of Cascadia (S.W. British Columbia, Oregon, California). These include (1) observation of “silent slip” earthquakes along the Cascadia margin (Dragert et al., 2001), (2) evidence for widespread serpentinization of the forearc mantle (Bostock et al., 2002; Brocher et al., 2003), and (3) non-volcanic tremor apparently related to dehydration reactions in the down-going oceanic plate (Rogers & Dragert, 2003; see also Obara, 2002). All have important implications for our understanding of earthquake hazard in SW BC, the most earthquake prone region of Canada, and all were rendered possible through the deployment of dense networks of seismometers and GPS instrumentation.

POLARIS research in BC builds on these discoveries through the deployment of a dense array of seismometers, MT sensors and geodetic (GPS) instruments across southern Vancouver Island, northwest Washington and the BC lower mainland. The dense sampling permits application of newly-developed inverse scattering procedures to recover images of subsurface structure in unprecedented detail (Bostock et al., 2001). These procedures exploit recorded scattered teleseismic wavefields to provide an estimate of physical property changes to depths of 100km or more. In Figure 5 we display shear impedance perturbations across the POLARIS BC array. The oceanic crust of the subducting plate is clearly evident as a dipping low-velocity (red) layer at the western end of the profile. Near 35 km depth, the crust undergoes a change in character presumed due to eclogitization that, through dehydration, causes serpentinization and weakening of the overlying mantle. Serpentinization may play an important role in limiting the maximum size of megathrust earthquakes by preventing rupture to mantle depths.

Fig. 5. W-to-E cross section of the North Cascadia subduction zone beneath Vancouver. The western (left) edge of the model coincides with the western coastline of Vancouver Island. Red indicates slow velocity perturbations, blue fast ones. This image shows results of inversions assuming only P- to S-wave mode conversions during scattering. Boundaries between prominent red to blue transitions are Moho discontinuties.

Unique MT studies are also underway in this region, building upon earlier studies of the Cascadia subduction zone by Kurtz et al. (1990). In this recent POLARIS work by Unsworth and post-docs at the University of Alberta, long-period MT techniques have been applied for the first time to the region and preliminary results indicate that the new data may image the deep fluid distribution in the mantle wedge and asthenosphere.

The urban centers of southwestern BC, home to more than 2 million people and the economic core of the province, are at risk from three types of damaging earthquakes; (1) rare megathrust (M~9) earthquakes; (2) deep (~55 km) earthquakes (up to M~7) within the subducting Juan de Fuca plate; and (3) large (M>7) shallow earthquakes within the North American plate. In ongoing work we exploit the dense sampling of the POLARIS BC array to more accurately locate deep earthquakes relative to the subducting plate (Cassidy & Waldhauser, 2002). Although smaller than megathrust events, the deep earthquakes such as the Seattle-Nisqually event of February 2001 have great destructive potential. Like the newly discovered non-volcanic tremor, they are thought to result from phase changes in the subducting oceanic crust. Using 3-component broadband waveforms of POLARIS type, seismologists can now precisely locate the shallow earthquakes to within a few hundred meters, and thus identify an active fault beneath the Strait of Georgia (Cassidy et al., 2000) and search for active faults in the vicinity of Vancouver, BC.
Unique seismic attenuation studies that are underway in this region (Cassidy, Atkinson, Spence, and graduate students at the University of British Columbia, University of Victoria, and Carleton University) include examining site-specific differences between shaking from deep earthquakes and that from shallow earthquakes. These studies require several years of earthquake recording to provide improved estimates of ground shaking potential. Site response studies (Cassidy & Rogers, 1999; Atkinson & Cassidy, 2000) use POLARIS data to estimate the contribution of local geology to ground shaking. The POLARIS corridor in southwest BC samples the soft, thick soils of the Fraser River Delta, on which numerous critical structures were built. Such improved seismic hazard estimates in southwestern BC using the POLARIS infrastructure will reduce losses from future earthquakes, projected in the ten’s to hundred’s of billions of dollars.

2.4 Geomagnetic storm damage

Geomagnetic storms and associated solar effects represent a hazard to communication systems and electrical power transmission grids. POLARIS magnetotelluric observatories in Canada can improve the understanding of geomagnetically induced currents (GICs), the electric ground currents induced by geomagnetic storms. GICs contribute to the long-term degradation of power networks and to the corrosion of pipelines and other buried infrastructure through variations in the pipe-to-soil potentials. Extraneous currents entering power systems can cause failure of protective devices such as relays and circuit breakers and cause system shut-down. In extreme situations GICs can cause the collapse of a power system, for example as in the power outage of March 1989 in Québec and the collapse of the Québec-New England DC link in March 1991.

Modelling of GICs requires knowledge of: (1)geomagnetic source fields, including the statistical definition of events of particular size and an understanding of the possible wavelength structure of the events; (2) Earth electric fields induced by these sources and therefore large-scale Earth resistivity structure; (3) the geometry of the technological network, including the grounding points. Present GIC predictions for Canadian networks are based on relatively simple models of Earth resistivity structure and simplified models of geomagnetic source-fields. POLARIS electromagnetic infrastructure will refine GIC prediction through improved definition of the continent's large-scale resistivity structure and through correlation of measured GICs with surface magnetic and electric field recordings. By using the POLARIS magnetotelluric array for simultaneous measurements it will be possible to define the temporal, frequency and spatial variations in the magnetic and electric fields driving the GICs.
Several POLARIS geomagnetic-hazard projects will be completed in the next four years. In Manitoba, the emphasis will be on improving the resolution of large-scale resistivity structure in order to more accurately predict the earth potentials driving GICs. In Ontario, the focus will be examining correlation of recorded GICs with magnetic and electric field recordings. The POLARIS magnetotelluric infrastructure also faciliates interaction between solid-earth geophysicists and space-weather scientists in Canada. POLARIS instruments can be deployed in the form of a temporary magnetometer array and used in space weather studies to supplement magnetic field recordings at more-sparsely located permanent magnetic observatories.

2.5 Rapid Earthquake warning technology

With the POLARIS infrastructure in Ontario we will receive seismographic data from over 30 observatories and MT data from 2-5 observatories in real-time (delay of only a few seconds), permitting development of Rapid-Warning technology for earthquake ground shaking. The TriNet project of southern California, supported by dense instrumentation and collaboration between government, universities and industry, has demonstrated that it is possible to provide reliable information on the distribution of the intensity of ground shaking within 10 minutes of the occurrence of an earthquake (Wald et al. in Earthquake Spectra, 1999). In the August, 1999 magnitude 7 earthquake in the California desert, rapid-warning information was available to stop trains before they travelled into areas where track damage was likely to have occurred. California utilities have formulated detailed earthquake-response plans, in which the response actions are keyed to the data provided by the shake maps. Future developments may even allow warnings to be issued several seconds in advance of the most severe portion of the seismic shaking, allowing automatic safe shutdown of critical systems, such as those in nuclear power plants. Real-time spatial analysis of earthquake ground motion in densely populated regions can provide crucial and timely information to emergency-response organizations and operators of critical industrial facilities, allowing them to prioritize their responses and take appropriate measures to reduce loss of life and mitigate damage.

We are developing the tools required to rapidly calculate maps of ground shaking within the Ontario array, including the calculation of response spectra at selected locations. With a concentration of stations in Ontario’s economic core (Toronto area, the 401 corridor, and also the Ottawa area), we are well positioned to develop the necessary technology to provide shaking estimates, within a few minutes of an event, for the locations of critical facilities. This would allow rapid assessment of potential damage, and facilitate timely responses to mitigate damage. An improved system to alert provincial, municipal and industrial subscribers (e.g., Province of Ontario, major urban cities, Ontario Power Generation, Bruce Nuclear) of an event and rapidly disseminate the information is also under development. This technology is an ideal application for the satellite telemetry approach to data communications showcased in POLARIS. It is anticipated that this field of application will continue to grow in the future as a growing roster of sponsors subscribe to additional sites at or near selected critical facilities and it is extended to the populated regions of southwest British Columbia and Quebec.

2.6 Geodetic measurement of crustal strain and deformation processes

Continuously-operating geodetic Global Positioning System (GPS) networks provide good spatial and temporal resolution for the estimation of individual site ground velocities and trajectories relative to a global reference frame, as well as the spatial coverage needed to study local variations in the strain field. Precise measurement of crustal strain and horizontal deformation are obtained with sufficient accuracy (~ 1 mm per year) to map patterns of strain buildup prior to a large damaging earthquake, or crustal motions associated with postglacial adjustment. Seismicity and the more gradual deformation of the Earth are sensitive to different geological properties, and provide important, often complementary information on a variety of scale lengths. Strain measurements from GPS receivers can, for example, be used to characterize fault structures, mechanisms and depths. By making use of VSAT communications and data archival systems, geodetic GPS hardware can be incorporated directly into the POLARIS infrastructure very efficiently.

The three focus areas for research with the POLARIS infrastructure include western BC, southern Ontario, and the lower St. Lawrence region (Charlevoix) of Quebec. These areas provide a range of crustal deformation rates for tectonic studies in regions of recognized seismic hazard in Canada. The long- and short-term potential uses of POLARIS GPS deployment include regional strain estimates obtained from differential crustal motions in order to measure the accumulating stress available for seismic release; detailed strain measurements in areas of higher risk; deformation measurements in a stable intraplate tectonic environment such as the Charlevoix seismic zone and the constraints that deformation place on its seismicity; monitoring post-glacial rebound and global sea-level change; the measurement of North America tectonic plate velocity and stability; data assimilation into meteorological applications, such as weather and climate models; effects of groundwater migration and local subsidence; and comparison and augmentation of satellite remote sensing data, in particular InSar imaging.
2.7 Integrated analysis and modeling of continental dynamics

A new research focus is the construction of models for the evolution and dynamics of the lithosphere which incorporate geological constraints as well as new experimental results from the POLARIS geophysical data. Such understanding is crucial in order to allow practical applications of the new data sets in seismic hazard mitigation and mineral resource discovery to develop from a solid theoretical foundation. The seismographic and magnetotelluric data made available by POLARIS will bring dramatic improvements in the clarity/resolution of lithospheric structures in the upper few hundred km of the Earth. The derived acoustic and electrical properties and their spatial variability will be combined with knowledge of surface geology and the properties and behaviour of rocks from low (upper-crustal) to high (upper-mantle) temperatures and pressures. Knowledge of in situ stress, historical seismicity and fault movements will also be incorporated. Geographic Information Systems (GIS) will be used to manage the differing databases. Geodetic GPS recordings are particularly significant, because contemporary velocities will provide the constraint on crustal deformation required to calibrate the lithosphere-scale geodynamic models.

The lithosphere-scale models in turn will provide boundary conditions for the more focused, shallow crustal dynamical models that are of direct use in engineering-stress analyses. These latter models require a region to be meaningfully partitioned into structural domains (or elements), with which the mechanical responses to imposed tectonic boundary conditions can be investigated. This geomechanical modelling will involve existing computer tools and potentially also exisiting analogue (scale-model) facilities. New, high-performance computational facilities will greatly facilitate this aspect of the research. This innovative multi-scale approach to mechanical analysis will form a vastly improved basis for understanding the complex dynamics of the region and fundamental linkages between regional geology and seismicity.
3. How POLARIS impacts the national research community
POLARIS is the only research facility in Canada that addresses the needs (in particular hardware, instrumentation) of solid Earth geophysicists with interests in deep Earth structure and earthquake science. The facility was originally conceived and developed by a group of ten geophysicists from across Canada representing both universities and government. Prior to POLARIS, individual researchers had limited access to small numbers of instruments, usually acquired through NSERC grants or from the Geological Survey of Canada. POLARIS now affords the Canadian geophysical community access to modern state-of-the art geophysical equipment in quantities necessary for the undertaking of forefront research. Several broadly-similar programs exist internationally (Australia, Germany, United Kingdom, U.S.A), but access is virtually always restricted to respective nationals. Moreover, long-term maintenance and support of the POLARIS facility is essential to the preservation of a competitive edge for Canadian solid Earth geophysicists and indeed other geoscientists within the international research community.
POLARIS bears some superficial resemblance to Lithoprobe, Canada's National Geoscience project, and now in the closing year of its 20-year duration. However, POLARIS and Lithoprobe are quite distinct in both focus and essence. POLARIS is an equipment facility that supports the research programs of Canadian geophysicists engaged in the study of earthquakes, earth structure and dynamics, whereas Lithoprobe has been a research program per se with the objective of deciphering the geological history of the North America.
It has always been the intent of POLARIS that the facility be a community resource, accessible to all interested Canadian geophysicists. Individual researchers have neither the time nor resources to manage the large numbers of instruments required for modern geophysical research. Thirty or more stations are typical of modern field campaigns. It is therefore both expedient and economically advantageous that geophysical instrumentation be managed through a central facility with dedicated technical support. Although the initial POLARIS research program has committed instrumentation to the projects described in section 2.0, these projects will draw to a close as of 2005 and instrumentation will be allocated to new projects on a proposal-driven basis (see section 4).
POLARIS provides an unparalleled vehicle for international collaboration. The remoteness of some POLARIS stations provides ideal environments for recording electromagnetic and seismic waves because noise levels are very low. We anticipate that many researchers that develop new analytical code will choose POLARIS data for testing. Researchers at UBC, MIT, UWO and the University of Manitoba already have or will do so, and many seismologists worldwide are aware of the quality of POLARIS data. It is anticipated that a significant proportion of POLARIS instrumentation may be deployed in southern Canada in conjunction with the EarthScope-USArray program (www.earthscope.org) to map deep Earth structure in a corridor including the the northern United States planned for 2006-2010. Some POLARIS instrumentation may also provide a useful land-based complement to sea-floor seismological observatories off the Pacific Coast planned for 2007 as part of the joint U.S.-Canada NEPTUNE program (www.neptune.washington.edu).
The telemetry design of the POLARIS field observatories is well suited to the remoteness of the Slave craton and most parts of the NWT and Nunavut, a major part of the Canadian landmass. MT measurements or earthquakes can be recorded and transmitted to research centers throughout the year, notably during the dark winter months when temperatures of –20 to –40°C make field access impossible. Many analytical techniques require that the seismic wavefield be sampled every 5-20 km and therefore a large number of stations must record data continuously for 1-2 years or longer.

4. Management of the POLARIS facility
The nucleus of the management structure for the POLARIS facility is the standing national steering committee that includes members from Carleton University, University of Western Ontario, University of British Columbia, University of Alberta, University of Manitoba, the Geological Survey of Canada (the Natural Hazards and Emergency Response and the Northern Resource Development Programs) and industry representatives. The Committee is currently chaired by Atkinson from Carleton University, with Asudeh of the Geological Survey acting as Program Manager. The POLARIS steering committee meets by telephone once each month, and in person at least once per year. A national workshop is held annually to review scientific and operational activities of POLARIS during the previous months, and to discuss plans and make preparations for continued operations in the following year. It often coincides with the annual meeting of the Canadian Geophysical Union.
The POLARIS infrastructure is designed to be redeployed at intervals of months to years depending on the application. As the initial deployment phase of POLARIS is completed in 2005, the steering committee’s role shifts to one of managing the allocation of instruments. The steering committee will review proposals for the use of POLARIS equipment, seeking independent review of proposals as needed, and decide on the deployment schedule. The steering committee oversees maintenance and ongoing infrastructure costs, and facility usage.
POLARIS equipment is available primarily to researchers from Canadian universities, governments and industry, but also to non-Canadian academic users at different levels of user fees (see Budget Justification). This user fee structure is fully integrated and compatible with the relevant Service Modules within the Geological Survey of Canada. As a truly national facility, POLARIS infrastructure is made available to investigators on the basis of the merit of their research proposals. Researchers who borrow POLARIS equipment are responsible for all costs associated with deployment and retrieval of the equipment, and other costs directly associated with a particular experiment. Researchers can also hire POLARIS to conduct full field deployment and retrieval. Ongoing operational costs are being sought under this current application.

5. Contribution of POLARIS to the training of highly qualified personnel
POLARIS will play a major role in the training of the next generation of Canadian scientists in areas of solid Earth geophysics and earthquake hazards. Since 2001 the project has led to the development of four new faculty positions in Canadian universities and the decision of a number of other scientists to accept academic positions in Canada. It has led to the training, in full or in part, of 3 postdoctoral fellows and research associates, 2 international visiting scholars, 4 Ph.D. students, 6 M.Sc. students, and more than 10 undergraduate thesis and Summer student projects at universities in Ontario, Manitoba, and British Columbia. In addition 4 technologists at Canadian universities and the Geological Survey of Canada have been recruited and trained using POLARIS infrastructure and additional technologists have used POLARIS equipment as a significant component of their duties. Over the coming years POLARIS will make even greater contributions as the new scientists recruit and train “second generation” personnel. Over the next three years it is anticipated that at least 20 to 25 graduate theses and a comparable number of undergraduate theses will be based on research employing the POLARIS facility.

POLARIS will allow the training of personnel in areas of high demand for Canadian society and industry. Canada is blessed with a great variety of geological environments, spanning all of Earth's history and most of its geological processes, and containing vast natural resources. From a scientific perspective, this renders Canada a perfect natural laboratory within which to undertake studies of continental structure. From an economic standpoint, it means that large areas of the Canadian Shield represent prospective targets for diamond exploration. From the standpoint of social welfare and economics, it means that a number of major cities are situated in regions of moderate to high earthquake hazard. POLARIS infrastructure will allow us to train Canadian geophysicists to fully exploit these unique geological circumstances and thereby become (i) leaders in the study of continental lithosphere, (ii) pioneers in the application of geophysical techniques to diamond exploration, and (iii) innovators in the assessment and mitigation of seismic risk.

Research trainees must, as part of their research, become acquainted with the language and concepts of related disciplines, notably inverse theory, petrophysics, petrology, geochemistry and civil engineering. This will provide the broadly-based scientific communication skills that are essential to success as the nature of research becomes increasingly collaborative. Their multi-disciplinary training will make POLARIS trainees highly flexible scientists capable of working in a wide range of scientific and engineering endeavours. Moreover, through the links forged by project scientists with industry (e.g., diamond mining, provincial hydro authorities), trainees will be exposed to industrial concerns, practices and requirements at an early stage in their careers. This class of training is important to the economic and social welfare of Canada. In addition to their application in seismic risk assessment and to understanding the mantle source of diamonds, the same specialized skills are also required of practitioners in the increasingly important field of environmental hazard evaluation, and in petroleum and base-metal exploration.
POLARIS geophysical equipment provides personnel with valuable hands-on field training and the opportunity for them to analyse important new data sets. At the most basic level, research trainees and scientists will use equipment that is at the forefront of geophysical research. In structural, seismicity and geomagnetic studies, all stages of data acquisition, reduction, processing and interpretation will rely critically upon the participation of students and postdoctoral fellows in addition to research scientists. During the course of their studies, trainees will be exposed to (i) instruction in geophysical instrumentation techniques by highly qualified technical personnel; (ii) the rigorous demands of field deployment; (iii) the challenging solution of signal processing and inverse problems; (iv) the integration of results with other geoscientific information; and (v) dissemination of results to the scientific community, industry partners and the general public.
POLARIS infrastructure will provide a setting unique in Canada for the training of highly qualified research personnel. The value of field-based geoscience programs to such training is well illustrated by Canada's Lithoprobe program, now nearing its completion. A high proportion of the many undergraduate and graduate students who worked within that program have gone on to complete higher degrees, take up academic positions, and occupy important positions in government geological surveys, petroleum and mineral exploration companies, and environmental and engineering consulting companies throughout Canada.

6. Synergy
Modern geophysical surveys require ever increasing numbers of individual sensors and take full advantage of the ever increasing capacity of computers to analyze digital data. Mantle-scale or earthquake studies cannot realistically involve either the numbers or the density of sensors typically used by the petroleum exploration industry. However, arrays with hundreds of elements equivalent to POLARIS observatories deployed for several years are certainly required, and these cannot be purchased or effectively maintained by a single academic institution. For example, the USArray of Earthscope in the USA will be operated by the Incorporated Research Institutions for Seismology (IRIS), a consortium of tens of universities and research organizations. The ready solution is a major facility that naturally draws together all the practitioners of the specialized field, and requires that they interact among themselves in prioritizing equipment usage. The central facility also brings together students, faculty, government researchers and exploration industry managers at the various levels required to operate and use the facility and its equipment. POLARIS exemplifies just this linkage among like-goaled research and development groups in a large country with finite technological and human resources.
As a consequence of organising to meet the practical needs, the various groupings of geophysical specialists are encouraged to interact more readily in planning scientific campaigns. Multi-disciplinary research among seismology, electromagnetics, geodesy, geodynamics, earthquake engineering and tectonics ensues much more readily in this environment, and students gain easy access to multiple research and job opportunities when involved in any component of the research using the POLARIS facility.

References (including co-applicants' works stemming from POLARIS activities)

Atkinson, G.M., and J.F. Cassidy, Integrated use of seismograph and strong motion data to determine soil amplification: Response of the Fraser Delta to the Duvall and Georgia Strait earthquakes, Bulletin of the Seismological Society of America, 90, 1028-1040, 2000.

Adams, J., and Halchuk, S., Fourth generation seismic hazard maps of Canada: Values for over 650 localities intended for the 2005 National Building code of Canada Geological Survey of Canada Open File 4459, 155 pp., 2003.

Bank, C. G., M.G. Bostock, R.M. Ellis, J.F. Cassidy, A reconnaissance teleseismic study of the upper mantle and transition zone beneath the Archean Slave craton in NW Canada, Tectonophysics, 319, 151-166, 2000.

Bostock, M. G., Mantle stratigraphy and evolution of the Slave province, J. Geophys. Res., 103, 21,183-21,200, 1998.

Bostock MG, Rondenay S, Shragge J, Multiparameter two-dimensional inversion of scattered teleseismic body waves 1. Theory for oblique incidence, J. Geophysical Research, 106, 30771-30782, 2001.

Bostock MG, Hyndman RD, Rondenay S, Peacock, SM, An inverted continental Moho and serpentinization of the forearc mantle, Nature, 417, 536-538, 2002.

Brocher TM, Parsons T, Trehu AM, Snelson CM, Fisher MA, Seismic evidence for widespread serpentinized forearc upper mantle along the Cascadia margin, Geology, 31, 267-270, 2003.

Cassidy, J.F. , and G.C. Rogers, Seismic site responsein the greater Vancouver, British Columbia area: Spectral ratios from moderate earthquakes, Canadian Geotechnical Journal, 36, 195-209, 1999.

Cassidy, J.F., Rogers, G.C., and F. Waldhauser, Characterization of active faulting beneath the Strait of Georgia, British Columbia, Bulletin of the Seismological Society of America, 90, 1188-1199, 2000.

Cassidy, J.F., and F. Waldhauser, Evidence for both crustal and mantle earthquakes in the subducting Juan de Fuca plate, Geophys. Res. Letts., 30(2), 1095, 2002.

Davis, W., Jones, A. G., Bleeker, W., & Grütter, H., (in press), Lithospheric development in the Slave craton: a linked crustal and mantle perspective, Lithos.

Dineva, S., Eaton, D. and Mereu, R., Seismicity of the southern Great Lakes: Revised Earthquake Hypocenters and Possible Tectonic Controls. Bull. Seismol. Soc. Am., submitted, 2003.

Dragert H, Wang KL, James TS, A silent slip event on the deeper Cascadia subduction interface, Science, 292, 1525-1528, 2001.

James, D.E., M.J. Fouch, J.C. VanDecar, S. van der Lee, and the Kaapvaal Seismic Group, Tectospheric structure beneath southern Africa, Geophys. Res. Lett., 28, 2485-2488, 2001.

Jones, A. G., Ferguson, I. J., Chave, A. D., Evans, R. L., McNeice, G. W., Electric lithosphere of the Slave craton. Geology, 29, 423-426, 2001.

Kurtz, R.D., DeLaurier, D. and Gupta, J. The electrical conductivity distribution beneath Vancouver Island: A region of active plate subduction, J. Geophys. Res., 95, 10929-10946, 1990.

Li, A., D.W. Forsyth, and K.M. Fischer, Shear wave structure and azimuthal anisotropy beneath eastern North America from Rayleigh wave tomography, J. Geophys. Res., in press, 2003.

Lockhart, G., Grütter, H., & Carlson, J., Mapping the Mantle Lithosphere for Diamond Potential (extended abstract), 8^th International Kimberlite Conference, Victoria, B.C., 22-27 June, 2003.

Mereu, R.F., Asmis, H.W., Dunn, B., Brunet, J. Eaton, D., Dineva S. and Yapp, A., The seismicity of Western Lake Ontario Area: Results from the Southern Ontario Seismic Network (SOSN), 1992-2001, Seism. Res. Lett. 73, no. 4, 534-551, 2002.

Obara K, Nonvolcanic deep tremor associated with subduction in southwest Japan, Science, 296, 1679-1681, 2002.

Rogers G, and Dragert H, Episodic tremor and slip on the Cascadia subduction zone: The chatter of silent slip, Science, 300, 1942-1943, 2003.

Silver, P. G., and M.K. Savage, The interpretation of shear-wave splitting parameters in the presence of two anisotropic layers, Geophys. J. Int., 119, 949-963, 1994.

Snyder, D., Bostock, M. & Lockhart, G., Mapping the Mantle Lithosphere for Diamond Potential (extended abstract), 8^th International Kimberlite Conference, Victoria, B.C., 22-27 June, 2003.

Stevens, A.E., Earthquakes in the Lake Ontario region: Intermittent scattered activity, not persistent trends, Geoscience Canada 21, no. 3, 105-111, 1994.

VanDecar, J. C., Upper-mantle structure of the Cascadia subduction zone from non-linear teleseismic travel-time inversion, Ph.D. thesis, 165 pp., Univ. of Wash., Seattle, June 1991.

Wallach, J.L., Mohajer, A.A. and Thomas, R.L., Linear zones, seismicity, and the possibility of a major earthquake in the intraplate western Lake Ontario area of eastern North America, Can. J. Earth Sci. 35, 762-786, 1998.