What is the nature and location of the plate boundary between North America and Eurasia and do microplates possibly exist in between?

Introduction

Definition of problem
The North America-Eurasia plate boundary is one of the world's most poorly understood plate boundary zones. The precise location of the northwest boundary of the North American plate is still a major problem in geology today. In oceanic areas the boundaries between major plates are readily defined by the distribution of earthquakes and bathymetric features and are typically no more than a few kilometers in width. Where plate boundaries bisect continental masses, however, the place where one plate ends and another begins is generally much more difficult to locate with confidence. The problem of identifying a plate boundary within a continent is heightened when the relative velocity of the two plates is small, as in the case of Eurasia and North America. Seismicity is the primary basis for identifying the North American-Eurasian plate boundary in the Atlantic and Arctic oceans. However, what happens to the plate boundary as it continues from the Arctic Ocean onto the Eurasian continent is by no means clear from seismicity (Figure 1).
Intracontinental convergent plate boundaries tend to involve complex assemblages of moderate-scale blocks or microplates, which move semi-rigidly with respect to the adjacent plates, as observed throughout the Tethyan belt (England and Jackson, 1989). There are a number of ways in which northeast Asia may be subdivided into plates consistent with the seismicity and recent tectonic activity. In addition to the Eurasian and North American plate, several smaller plates have been proposed in this region, for example the Okhotsk and Bering plates (e.g. Cook et al.,1986, Minster et al., 1974). The questions to be addressed by this proposal are: Previous authors have proposed qualitative tectonic models for this plate boundary area based on seismicity and geologic observations (e.g. Riegel et al., 1993; Mackey et al., 1997). Many authors have proposed Euler vectors (poles of rotation and angular velocities) for Eurasia-North America which vary considerably (e.g. Minster et al., 1974; Savostin et al., 1983), and Savostin et al. (1983) proposed a Euler vector for North America-Okhotsk. However, these poles and rotation rates are based on fracture zones, magnetic anomalies, seismicity and global inversions, all of which are indirect evidence and averaged over a range of different time periods. I propose to use constraints from GPS data to test the most recent tectonic models and to quantify the present day relative plate motions involved, something that has not been directly measured before in this part of the world.

Significance of problem

The North America-Eurasia plate boundary is one of the world's most poorly understood plate boundary zones. The precise location of the northwest boundary of the North American plate is still a major problem of Arctic geology today. It has been recognized since the early development of plate tectonics that plate boundaries in continental areas are substantially wider than those in oceanic plates (e.g. Isacks et al., 1968). The underlying causes for this diffuse rather ill-defined nature of intracontinental plate boundaries, as well as the most appropriate way of describing continental deformation (i.e. continuum deformation versus microplate or block behavior), remains the subject of ongoing debate (e.g. Avouac and Tapponier, 1993; Thatcher, 1995). GPS is providing an important new tool to quantify continental deformation to a precision and on a scale unprecedented in the Earth sciences (e.g. Hager et al., 1991). Using GPS to study the kinematics of continental deformation both in northeast Russia and in western Alaska will, in turn, provide constraints on rheological models of the continental lithosphere and the forces responsible for active deformation. Such knowledge would grant us insight into the nature of complex continental plate boundaries, such as the Arabia-Africa-Eurasia and India-Eurasia collision zones and even the small-scale Yakutat block-North America collision. It may then be possible to start to unravel the past tectonic record, which, once understood, provides a wealth of information by which present tectonics can be interpreted and future tectonic motions can be predicted.
A better knowledge of the tectonics of the area is invaluable for the construction of global plate tectonic models, which are used to evaluate current plate motions and to predict future plate motions worldwide. Complete specification of the boundary of each plate is also necessary to conduct tests of driving force models for plate tectonics. On a regional scale, a more complete tectonic picture of this area is valuable for the creation of hazard maps. No quantitative earthquake hazard map is in existence for this plate boundary zone area. The Bering Strait region, for example, is the most poorly studied region(s) in North America with a high level of seismic activity (Mackey et al., 1997). One magnitude seven and several magnitude six events have occurred in the Bering Strait region (dates??). Large earthquakes such as these are potentially tsunamagenic and could cause damage to villages on the western Alaska and eastern Chukotka coasts. Sakhalin Island has recently been recognized as having high oil potential. (The high seismicity rate and) Future large earthquakes that might occur (there) in this region could cause considerable damage to the high-cost equipment related to drilling for oil and potentially produce an oil spill that would devastate the fragile arctic ecosystem. The seismic potential of the entire region can only be more completely understood once the nature of the plate boundary is determined.
 

Present Understanding of the Tectonics of the Eurasian-North American plate boundary

Early plate boundary ideas
Relative motion between the Eurasian and North American plates currently produces divergence in the North Atlantic and convergence in northeast Siberia. The boundary between North America and Eurasia enters continental Asia at the southern end of the Laptev Sea in Buor Khaya Gulf. A zone of seismicity extends southeast to the Sea of Okhotsk, trends down though Sakhalin Island and on to a triple junction at the Japan-Kurile trench (Figure 1??) This zone was (has been) proposed by many authors to be a diffuse Eurasia-North America plate boundary (e.g. Churkin, 1972; Chapman and Solomon, 1976). However, a diffuse zone of seismicity is also apparent along the Chersky mountains and on to northeastern Kamchatka and the Aleutian-Kuril junction (Cook et al., 1986; Riegl et al., 1993).
It was recognized in the early development of plate tectonics that plate boundaries in continental areas are substantially wider than those in oceanic plates (e.g. Isacks et al., 1968). The plate boundary in Yakutia is very wide and diffuse. At its widest, the boundary is 600km wide. The diffuse nature of the boundary is more likely to be a property of continental lithosphere than to be due to the slow relative plate velocity. The width of the seismic zone increases markedly between the Nansen ridge and its extension onto the continental shelf in the Laptev Sea. Continental lithosphere is very heterogeneous and has generally undergone a long history of stress- and fracture-producing tectonic activity and is crisscrossed with weak zones highly susceptible to deformation when stressed.
The broad seismic belt in Yakutia lies within the Chersky-Verkhoyansk fold belts, which Churkin (1972) (has) interpreted as a fossil suture marking the early Cretaceous collision of two continental blocks. The current seismicity between Sakhalin and Hokkaido closely follows a Mesozoic plate boundary marking the locus of eastward subduction of one plate beneath another (Den and Hotta, 1973). This is a familiar story in the plate tectonic evolution of the earth's surface: fossil plate boundaries are apparently relatively weak portions of continental blocks and are the preferred sites for the creation of new plate edges.
There are a variety of estimates for the Eurasia-North America pole of rotation, (see for example the summary in Cook et al., 1983). All these estimates are obtained by different methods, including transform orientation and magnetic lineaments in the North Atlantic and Arctic oceans (DeMets et al., 1993??), strikes of faults in the Chersky mountains (Savostin et al., 1982/3), earthquake slip vectors (Cook et al., 1986), and global plate inversions (Minster and Jordan, 1976). Their (?) different methods average the plate motion over different time periods and use information from different geographic areas. It is now believed that the Euler pole for Eurasia-North America has moved around in northeast Russia as one goes backwards in time (David Stone, oral (or written?) communication, 1997), thus the geologic expression of deformation associated with this relative motion is complex.
 

Evidence for a separate Okhotsk microplate
Den and Hotta (1973) proposed the existence of an Okhotsk plate during the Mesozoic and early Cenzoic on the basis of structural trends and orogenic belts in and around the Sea of Okhotsk (though their discussion does not require a distinct Okhotsk plate at present).
Cook et al. (1986) identify a separate Okhotsk plate based on seismicity and focal mechanisms. Based on seismicity and mapped faults, there appears to be a zone of deformation between North America and the seismically inactive Sea of Okhotsk. (complex sentence ->) Combining this information with microseismicity west of the Sea of Okhotsk and the observation of large faults in the Sette Daban range west of the Sea of Okhotsk, indicating a boundary between the Eurasia and Okhotsk in the Sette Daban range, led Riegel et al. (1993) to agree that the Sea of Okhotsk and its environs constitute an independent plate (see Figure 2). Also apparent in the seismicity maps is a segment extending north from the junction of the Kurile-Kamchatka subduction zones before it trends westward to join the Chersky range seismicity. This section may have important implications with respect to the interaction of the Aleutian subduction system with Kamchatka, and also in regard to the apparent sea-floor spreading in the Komandorskii basin (David Stone, oral communication, 1997). The least clear part of the Okhotsk plate is the segment northward from (of?) the north end of Sakhalin island. There is an unclear connection northwards towards the north shore of the Sea of Okhotsk, where the seismicity becomes better defined again trending north toward the Laptev Sea (Figure 1). There is, however, a clear connection westward from the north end of Sakhalin Island through to the Baikal Rift. This unclear margin of the proposed Okhotsk plate may therefore be the result of the extrusion of parts of Asia by the collision of the Indian plate as proposed by Worral et al., (1996), and perhaps the lack of earthquakes in this area is due to high heat flow from crustal thinning.
The Okhotsk plate is being compressed due to the convergence of North America and Eurasia, causing extensive microseismicity. The Okhotsk plate is being extruded to the southeast, in a manner similar to the Anatolian plate between Eurasia and Arabia. DeMets (1993??) , however, demonstrates that slip vectors in the Kuril Trench are best modeled by assuming that the Sea of Okhotsk is part of the North American plate, but this model neither explains the northern zone of epicenters, nor the zone of large thrust earthquakes north of the Kuril-Aleutian junction.
No estimate of relative motion between Okhotsk and North America has been made using direct data. Savostin et al. (1983) compute an Okhotsk-North America pole of rotation and angular relative velocity based on seismic slip vectors, and DeMets (1993??) infers that the rate must be less than 5mm/yr using the Cook et. al. (1986) pole. However, due to the slow relative velocities and the deformation in the marginal areas of the Okhotsk plate it is difficult to estimate relative motions between the North American, Okhotsk, and Eurasian plates using indirect methods.
 

Evidence for a Bering microplate
Based mainly on (these, the aforementioned?) seismic observations, a separate plate or block of varying extent covering the Bering Sea region has been proposed ever since the earliest plate models (e.g. Minster et al., 1974). Minster et al. (1974) proposed a Bering plate, comprising western Alaska, the Bering Sea, and northeast Asia, to explain a systematic misfit of slip vectors from Aleutian and Kuril trench earthquakes to the Pacific-North American rotation pole. Lander et al. (1996) proposed a `Beringia plate' to explain the seismicity of the Koryak highlands, however, no supporting evidence was presented. More recently, Mackey et al. (1997) (have) defined a Bering block moving clockwise with respect to the North American plate, based on microseismicity and focal mechanisms from larger earthquakes (see Figure 2). These authors also back up their model using geological evidence for young crustal extension throughout the Seward Peninsula and continuing into Chukotka, and observe that the many Mesozoic-Cenzoic thrust faults in the Koryak Highlands may be undergoing reactivation, taking up the proposed westward motion of the Bering block. None of the models give any quantitative information on rotation rates for the Bering block relative to either North America or Eurasia. There is still some question as to whether the Bering Sea region is actually a separate plate, or whether in fact it is a deforming part of North America.
 

Outline of Proposed Research

Proposed hypothesis and approach
As outlined in the previous section, there are several proposed tectonic models for the Eurasia-North America plate boundary region. I favor the most recent models (e.g. Riegel et al., 1993; Mackey et al., 1997) as they have both the largest amount of, and the most recent, geological and geophysical information from which to draw their conclusions (see model geometry in Figure 2). While these authors define the plate geometry of their proposed model and indicate a sense of relative motion, no quantitative estimates are given as to the relative velocities involved. These models are concerned with different regions of the plate boundary zone and thus both of these models can be combined to form an overall model for the area. I therefore hypothesize that this combined model is correct in that there is indeed an Okhotsk microplate that is being extruded southeastward due to compression by North America and Eurasia (Riegel et al., 1993), and that a Bering block exists as defined by Mackey et al. (1997) and is rotating. I propose to use additional information from GPS data to test the model geometry. In addition, relative angular velocities of the plates will be determined, an important parameter that has never before been directly measured.
I propose to establish a broad profile of GPS stations across the Bering Sea, from Fairbanks in Interior Alaska, whose velocity is small with respect to stable North America, to Yakutsk in the Sakha republic, which can be thought of as stable Eurasia. Several permanent GPS sites are already in operation in this region, which will be of great benefit beneficial to this project. An additional ten sites are proposed; six on the hypothesized Bering block, two on the Okhotsk plate, and two in the vicinity of the North America-Eurasia plate boundary zone in the Chersky Range (??) (see Figure 3).
Using data from this network, significant constraints on the location and relative velocity of the Eurasian-North American plate boundary will (could?) be made. The existence of any microplates such as the Okhotsk and Bering blocks will be determined along with their velocities relative to the North American and Eurasian plates. The data will not be sufficient to determine the exact boundaries of any blocks, however, it will be possible to test various plate models for the area. Accurate estimates of the Eurasian-North American rotation pole and rate will be measured using GPS geodetic measurements at stations deemed to be within the stable continents. Poles and rates can similarly be calculated for the proposed Bering and Okhotsk plates using GPS data from stations within the proposed boundaries for these blocks. The primary test of a proposed plate model is whether the sense of motion at plate boundaries predicted by the relative angular velocity vector of the adjacent plates is consistent with seismic and tectonic evidence. A second test would be to derive a velocity field for the area from the Euler vectors and the plate geometry for a given model. This velocity field can then be compared with the individual station velocities and the misfit between the observed and calculated velocities will give an idea as to the strength of the plate geometry used.
Measurement details
Access to eight Trimble SSI GPS receivers is possible (guarantied) through the instrument pool at the Geophysical Institute. Data from modern GPS receivers such as these are of very high quality. The short-term precision of a typical 24-hour survey is 1-2mm (relative), although day-to-day correlations in GPS coordinates degrade the actual measurement precision. Three good yearly surveys are sufficient to estimate relative site velocities with a precision of 1-2mm/yr. Because relative velocities in this area are small, an initial survey of these sites in the first year of the proposal and a repeat survey of the sites after two years is thought to represent a reasonable trade-off between cost and measurement precision; it is more cost-effective to wait for the earth to move than it is to make frequently repeated measurements. Given the proposed rates of motion in this plate boundary region of up to 10 mm/yr (e.g. Riegel et al., 1993), surveying twice over a 3-year time span will give velocities that are significant. The `survey' approach to this project is much more cost-effective than installing permanent GPS monuments at the sites, and the measurement precision in an area with such slow tectonic rates is thought to be equally good.
 The Trimble SSI receivers available for our use can log up to a month of data before being downloaded. Using solar panels in the summer enables a receiver to be run continuously using only a modest-sized (38Ah) battery. The minimum survey time at any one site that should be aimed for is two days, although the longer the survey the better. It is estimated that the GPS fieldwork will require a minimum of two people, although only one of these persons need be experienced in the use of GPS equipment. Summer fieldwork seasons are preferable in order to maximize the possibility of combining this project with other seasonal projects in the area and to avoid the harsh winter conditions in the region.
All GPS data relevant to this project will be analyzed at the Geophysical Institute using the GIPSY/OASIS II software developed by the Jet Propulsion Laboratory in Pasadena. The PI posses extensive experience in processing GPS data. All of the GPS solutions will be combined into a solution for steady-state velocities.

Site selection

Many of the sites proposed have not been visited; however, significant problems are not expected in choosing sites. In both Alaska and northeast Russia it is fairly easy to find sites where a receiver can be left unattended safely for a few days or weeks (personal experience of the PI in Alaska and of David Stone in Russia). Potential sites have been identified based on a number of criteria:
1) the location must be useful for providing information about the tectonics of the area;
2) the bench mark must be in bedrock if possible as GPS measurements in these northern regions are constrained by the problem of ground instability (even without permafrost, ground movements can be as larger than the accuracy of the GPS due to freezing and thawing);
3) several sites near to seismic stations have been identified, which will save on transportation cost as trips can be combined with the seismologists, and often these seismic sites have power available which we can use in place of batteries;
4) use of a helicopter is unavoidable for some sites, however, where possible it is hoped that the helicopter use can be shared with seismologists and geologists to help defray costs.
Although the locations for GPS sites have been identified (see Figure 3), for many of these locations the reconnaissance work to find a suitable site for a GPS bench mark will take on the order of half a day per site. The actual monument installation takes approximately two hours, and setting up the GPS equipment takes only half an hour. Sites on the islands in the Bering Sea, and the site at Nome will likely be set up to record data for three to four weeks. The six remaining sites will be occupied for a minimum of three days each and the schedule for these observations will remain flexible, so that cooperation with other projects may take place and because various obstacles are anticipated due to working in a foreign country such as Russia.
 

GPS results
Data from the permanent GPS sites in Alaska are available to the PI. The data from the three permanent GPS sites in Russia are currently unavailable, however it is thought possible that these data may be obtained in the near future (David Stone, personal communication, 1997). Surveys have already been made by the National Geodetic Survey (NGS) at two of the proposed sites, Nome and St. Paul, (see Figure 3), and so after just one year of measurements (one measuring campaign - not a year of measurements)  velocities will be available for these sites. Combining this information with data from the permanent sites in both Alaska and northeast Russia will enable a first-order estimate of tectonic motions to be made after the first year of this proposal.
By examining the final velocity field of all the GPS sites, several things will be apparent: (objectives can be achieved, results can be obtained, can be inferred from the data)
1) plate boundary deformation between North America and Eurasia will be visible and the width of this deformation zone will be observed (can be inferred) ,
2) the velocities of sites on the proposed Bering block will show whether or not the block is moving at a different rate relative to North America and Eurasia, and any rotation of the block should be obvious; if the hypothesized model is correct, extension should be observed in western Alaska and compression in eastern Chukotka,
3) sites on the proposed Okhotsk block should show extrusion to the southeast according to the model being tested, and convergence should be obvious between the Okhotsk block and both Eurasia to the east and North America to the north.
Strain rates and rotation rates are derived directly from the velocity field computed from the GPS measurements. Spatially averaged rates can be computed for any area spanned by the GPS network, using as few as three GPS sites, or for any larger subset of the GPS network. The only assumption made in estimating strain or rotation rates from a series of GPS solutions is that the strain rate is uniform in time and homogeneous over the spatial area that is averaged over. (Compare with results from stress tensor inversion?)
Poles of rotation and angular velocities for the North America-Eurasia, North America-Bering, and North America-Okhotsk plates will be calculated from the GPS data with much more precision than previously possible. These Euler vectors will also be current and not averaged over the past few million years (Minster et al., 1974, used transform faults and magnetic anomalies to determine Euler poles and rotations). Given the density of the proposed network, definition of precise boundaries for the plates will not be possible, especially since it is suspected that these boundaries are broad zones of deformation. However, it will be possible to test various plate configurations given the Euler vectors for each plate. Velocities can be predicted for each site given the Euler vectors and the plate geometry to be tested, and the fit between the observed velocity and that predicted by the plate geometry used will give an idea of how good the model being tested is. If the model hypothesized does not fit the data, other models, such as those mentioned in the Tectonics section of this proposal, can be tested. The final result will be a quantitative tectonic model of this plate boundary area, obtained by combining the motions observed using GPS with knowledge gained from seismic data and what is known about the geologic evolution of eastern Siberia and Alaska.
 

Summary (often first statement)

The aim of this proposal is to better understand the tectonics of the Eurasia-North America plate boundary zone for reasons detailed above.  We propose to install A network of ten GPS stations will be installed to augment the already established permanent GPS stations. These sites will be measured twice over a period of three years, which will enable velocities to be calculated for each of the sites. The GPS data will be used to test proposed tectonic models for this area and quantify these models.
 

If you want to be trendy you mention IARC and the NSF statement of intent - your proposal fits right in! Syun might fund it?
 
  yellow: style??

Reads pretty well - I just had small suggestions that you probably don't like anyhow. You still don't show that you have the expertise in Alaska (refs). I have not read a real proposal were the PI has not a single Refs to his/her own work?NSF requires that you have relevant publications listed in your vitae? after all you have the publications, sort of at least.

Shortening is hard: A little less detail on the operation of the GPS receivers (should be known). To me the general tectonics section is hard to understand, figures would help I guess. Reduce the font size??

 

I am, sure you will get an A. Write 6 month of post-doc funding into the budget when you and David submitt it - then I can work on it?

Ready for a vaccation??