Subduction to Strike-slip Transitions on Plate Boundaries
June 48, 2000
Paul Mann, Institute for Geophysics, University of Texas at Austin, 4112 Spicewood Springs Road, Austin, TX 78759-8500
Nancy R. Grindlay, Department of Earth Sciences, University of North Carolina at Wilmington, 601 South College Road, Wilmington, NC 28403-3297
James F. Dolan, Department of Earth Sciences, University of Southern California, Los Angeles, CA 90089-0740
Most previous research has looked at plate boundaries as either strike-slip zones or subduction zones and have not emphasized plate boundary segments with hybrid character combining both tectonic styles. The manifestation of both tectonic styles is seen in the following combinations of structures and earthquake events:
- outboard subduction systems coupled to inboard strike-slip faults; plate motion is partitioned in complex ways between the two fault systems and leads to diffuse but active seismogenic zones;
- deep filled or underfilled trenches marked by the world's largest gravity minima and characterized by a combination of strike-slip and thrust faults. There are a large number of tectonic models to explain these deep trenches including: slab rupture, interaction with adjacent slabs, extension, and thrust loading;
- complex subducted slab geometries that include continuous but steep slabs, partially detached slabs, and completely detached slabs;
- complex pattern of historical and instrumentally recorded seismicity that includes great (M > 8) thrust and strike-slip earthquakes emanating from the transition zone.
Such modern zones leave complex geologic records characterized by superimposed tectonic styles with a common progression marked by a subduction boundary, a collisional event, followed by a strike-slip phase. The structures and igneous products of each stage are generally distinctive, although the younger, typically strike-slip phase can obscure relationships formed during the earlier convergent phase.
Within this context, we organized a Penrose Conference to systematically discuss and compare all aspects of subduction to strike-slip transition areas within a modern plate-tectonic framework. Our plan was to assemble a multidisciplinary and international group of geologists and geophysicists to first characterize the geologic and seismic signature of active transition areas around the world and then to compare and contrast these characteristics to ancient transitions described from mountain belts. Our goal was to foster feedback between experts in all lithospheric levels of ancient and active transitions-feedback that might lead to new insights to practical aspects of transition zones, including the seismogenic mechanisms for the large and commonly destructive earthquakes affecting these areas.
The conference was held January 1824, 1999, at the Puerto Plata Village Caribbean Resort in Puerto Plata, Dominican Republic. Puerto Plata is a large and easily accessible tourist center located on the Atlantic coast of the Dominican Republic. The city is approximately located over the seismogenic transition from subduction-dominated tectonics to the east produced by subduction of Atlantic (North America plate) lithosphere beneath the Puerto Rico area and strike-slip tectonics to the west. In 1946, the region was rocked by a M8 earthquake related to a large offshore thrust fault.
There were 72 participants, 44 from the United States, 7 from the Dominican Republic, 5 from Germany, 4 from New Zealand, 3 from the United Kingdom, 3 from France, 3 from Spain, 1 from Venezuela, 1 from Hungary, and 1 from Italy. The participants included 17 graduate students from the United States and Germany and 1 U.S. undergraduate conducting her senior thesis study on this topic.
The 70 presentations were divided into six theme sessions over three days at the conference center. Each session was introduced by two to four, 20-minute-long oral presentations by invited speakers and was followed by 3-minute oral introductions by each of the poster presenters in that session. Poster sessions for each session were held in a single hall and lasted one and a half to two hours. The poster sessions were followed by a one-hour discussion group led by four presenters from that poster session. Discussion centered on questions raised during either the oral or poster sessions related to that theme.
This report highlights some of the main results of the oral, poster, and discussion sessions. Those who are interested in further details will find more information, including the titles of all presentations, at the following Web site: http://www.uncwil.edu/people/grindlayn/Penrose.html
THE FIELD TRIP
A two-day field trip across the northern Dominican Republic followed two days of discussion at the conference center and preceded one day of discussion on the last day of the meeting. The field trip was planned to provide the participants a sampling of well-studied research areas within one day's drive from the conference center. Northern Dominican Republic provides an abundance of scenic landscapes, rapidly changing microclimates and diverse geology largely controlled by the active North AmericaCaribbean plate strike-slip boundary that traverses the area from east to west. Those who are interested in further details on the field trip can download the field trip guide at the following Web site: http://www.ig.utexas.edu/research/projects/caribbean/carib.links.htm.
The first two stops of Day One of the trip, led by Jim Pindell (Tectonic Analysis, Inc.) and Grenville Draper (Florida International University), examined outcrops of arc-related rocks, serpentinite and melange formed during Eocene and older subduction of Atlantic (North America) lithosphere beneath the Caribbean island arc, and their uplift history and exhumation along Neogene strike-slip faults. Both stops stressed the importance of late Neogene strike-slip related deformation, including shale and serpentinite diapirism, in the uplift and reactivation of Paleogene arc and collision-related structures. The second stop of Day One was led by Paul Mann and examined deformed Eocene-Oligocene turbiditic rocks outcropping at the 1000 m topographic high of the Cordillera Septentrional. These rocks unconformably overlie subduction-related rocks and record a transition to strike-slip tectonics within a submarine, California borderlandstype settting. Luis Peña (Consultant, Dominican Republic) led the third stop of Day One to view a quarry exposure of recent faulting of a late Quaternary alluvial fan on the northern edge of the Cibao basin. Strike-slip and normal fault splays across the width of the quarry align with a major fault strand of the North AmericanCaribbean plate boundary that extends across the Cibao Valley and was seen in more detail to the east of this locality on Day Two. At this stop, Johan Erikson (Dartmouth College) summarized evidence for a transpressional origin for the Cibao basin based on gravity and geologic data. The fourth and final stop of Day One was led by Tish Tuttle (University of Maryland), Luis Peña, and Carol Prentice (USGS, Menlo Park) and examined earthquake-induced liquefaction features along riverbank exposures of the Rio Yaqué del Norte in the western Cibao Valley. These sand dikes, sills, and sand blows occur within a late Quaternary fluvial section and are within 10 km of several recently mapped strands of the strike-slip boundary in the area. An ongoing study is relating the ages of these features to historical and prehistorical earthquakes that have affected the area.
Day Two began with stops related to the paleoseismicity and slip rate of the central segment of the Septentrional fault zone, the main strand of the North AmericanCaribbean strike-slip plate boundary. Field trip leaders Carol Prentice, Paul Mann, Luis Peña, and George Burr (University of Arizona, Tucson) showed participants offset terraces and summarized the results of previous fault trenching at Rio Licey. An ~60 m offset in a dated terrace provides a maximum slip rate Hestimate of 11-12 mm/yr. This offset rate is within the range of error (8 ± 3 mm/yr) for slip rates modeled using GPS results from the Dominican Republic. Trench results yield dates for the most recent earthquake between A.D. 780 and 1640 and the penultimate event occurring after A.D. 30. The second stop of Day Two examined a 3542 m offset terrace at the Rio Juan Lopez. Terrace dates constrained a slip rate of 69 mm/yr for this site that is also within the error of slip rates from the Rio Licey site and those modeled using GPS data. The third stop of Day Two visited Matancita, a small coastal community that was inundated by tsunamis triggered by the M8.2 1946 offshore thrust event. Field trip leaders Nancy Grindlay (University of North Carolina, Wilmington) and Jim Dolan (University of Southern California) presented the results of offshore mapping of active structures, including large cusps in the submarine slopes that may mark the scars of past tsunamogenic slumps. The fourth stop of Day Two visited an uplifted staircase of Pleistocene coral reefs that field trip leader Jim Dolan attributed to the localized oblique subduction of an adjacent submarine ridge. The fourth stop also included a summary of recent GPS results from Hispaniola by Eric Calais (University of Nice) and Alberto Lopez Venegas (University of Puerto Rico). At the fifth and final stop of Day Two, Grenville Draper showed the group an outcrop of Cretaceous glaucophane schist within a serpentinite matrix. These rocks were metamorphosed in a Late Cretaceous subduction zone and uplifted by late Neogene strike-slip movements.
Global Overview, Classification, and Comparisons of Strike-Slip to Subduction
Transitions on Plate Boundaries
Paul Mann opened the meeting with a global overview and classification of subduction to strike-slip transition areas on active plate boundaries. Because most transition areas form prominent cusps in the trend of the plate boundary, he classified transition areas into three types (see illustrations in abstract by Mann and Frohlich)
"Open corner" transition zones are characterized by plate cusps that describe convex or "open" angles when viewed from the downgoing plate. Examples include the northeastern and southeastern Caribbean, northeastern and southeastern Scotia, central Aleutians, northern and southern Marianas, Philippines, Taiwan, San Cristobal trench of Solomon Islands, southern Vanuatu, northern Tonga, North and South Islands of New Zealand, southern Carpathian Mountains, Aegean, and Sumatra-Andaman. Characteristics of open corner transitions include: (1) the trench and Benioff zone vary smoothly along strike; (2) the trench is anomalously deep69 km; (3) the Benioff zone steepens to vertical as the strike-slip boundary in approached; (4) discontinuous pieces of subducted slabs are present in the area of the oversteepened Benioff zone; (5) strike-slip faults are present in the area of the detached slabs and lengthen in a trenchward direction; and (6) a common tectonic setting of open corner transitions is a closing "loop" or strongly arcuate island arc constrained between two continents; inactive arc segments mark already collided areas now converted to strike-slip faults and active arc segments mark actively subducting areas. "Closed corner" transition zones are characterized by cusps in plate boundaries that describe concave or "closed" angles when viewed from the downgoing plate. Characteristics include: (1) a narrow seafloor bathymetric high or continental block impinges the subduction zone to form the indentation or "closed corner"; (2) the trench and the Benioff zone conform to the shape of the incoming colliding feature; (3) the Benioff zone is locally contorted into a sharp angle but is not necessarily vertical in dip or associated with detached pieces of slab; and (4) strike-slip faults may not be present on both sides of the closed corner. Examples of closed corners and their associated bathymetric highs include: Kamchatka (Hawaii-Emperor seamounts), western Himalayas (promontory of Indian sub-continent), and southeastern Alaska (Yakutat block).
The process responsible for closed and open corners may relate to the subduction or collision of bathymetric highs at subduction zones. Collision of the high forms a prominent closed corner adjacent to the collided area, but continued interaction of the subduction zone with the high can result in regional changes in the shape of the sudduction zone that produce open corners adjacent to the closed corner (e.g., open corner transition of central Aleutians flanks the adjacent closed corner transition of Kamchatka).
A third type of subduction to strike-slip transition is the fault-fault-trench (FFT) or fault-trench-trench (FTT) triple junction type. These junctions can produce abrupt transitions between strike-slip and subduction tectonics as observed in Guatemala (FTT), Panama (FTT), Chile (RTT), Mendocino (FTT), Woodlark (FTT), New Ireland (TTF), and New Guinea (FTT).
Cliff Frohlich summarized the seismic characteristics of the three transition types based on an analysis of the Harvard, EHB, and Abe global earthquake catalogues. Prominent features of open and closed transition zones include the absence or rarity of strike-slip events, even in areas of known strike-slip faults, and a corresponding moment deficit in strike-slip events in comparison to more numerous thrust events. It is unclear if this deficit in strike-slip events is related to a fundamental mechanical process or whether it simply reflects the occurrence of large but infrequent strike-slip events not well represented in the catalogues used for the study. This question recurred throughout the meeting because of obvious seismic risk implications for densely populated transition areas like the Caribbean and the Philippines.
Subduction to Strike-Slip Transitions in Oblique Subduction Settings (Caribbean,
New Zealand, Southeast Alaska, Western Aleutians, Scotia Arc, Sumatra)
This theme focused on defining the geologic and geophysical character of mainly open transition areas in the Caribbean and other areas. Eric Calais (University of Nice) began the session with a summary of the main geologic and geophysical constraints on the directions and rates of the North AmericanCaribbean plate boundary and differing tectonic models that have been derived from these observations. Recent but still preliminary GPS results of Calais and others favor a model with a CaribbeanNorth America motion slightly oblique to the plate boundary trace and partitioned into strike-slip faulting in interior areas and thrust faulting in more outboard areas of Hispaniola and Puerto Rico. Nancy Grindlay summarized evidence from marine surveys for oblique collision between the Bahama Platform and the Greater Antilles arc in Hispaniola and Puerto Rico that appears to have formed a major bend at this transition area. These data indicate that strike-slip faults may extend farther to the east into the subduction area than previously thought and that normal faults in the Mona Passage may segment the colliding area in the Hispaniola area from the already collided area of Puerto Rico. Phil Barnes (NIWA, New Zealand) summarized geologic and geophysical data from two transition areas in New Zealand: a northern transition area from the west-dipping Hikurangi subduction zone to a diffuse Marlborough strike-slip system and a southern transition area from the east-dipping Puysegur-Fiordland subduction zone to a much more localized Alpine strike-slip system. David Scholl (Stanford University) used the central and western Aleutian transition area to illustrate how recently active arc systems deform under the influence of a superimposed strike-slip regime. Posters accompanying this session focused on specific aspects of transition areas in the Caribbean, Alboran, Scotia, southeastern Alaska, Solomon Islands, and western Aleutians. The panel discussion focused on the application of the classification scheme to various settings and possible mechanisms to generate open and closed corners.
Subduction to Strike-Slip Transition at FTT Triple Junctions (Northern California,
Alan Levander (Rice University) introduced this session on transition zones at FTT triple junctions with results from deep crustal profiling in the Mendocino triple junction of northern California. These data showed the rapid transition from subduction to strike-slip associated with the passage of the triple junction and the presence of magma bodies near the base of the crust. Patricia McCrory (USGS, Menlo Park) reviewed the sedimentary and structural response of the Humboldt basin in response to the migrating Mendocino triple junction. Posters accompanying this session focused on specific aspects of the Mendocino triple junction, including its expression on offshore seismic lines, and the nature of earthquakes in the triple junction region. Panel discussion revolved around the question of whether there were any similarities between triple junctions like the Mendocino and open and closed corner transitions discussed in the first session. Most participants agreed that the triple junctions were abrupt and more localized and represent a different set of tectonic processes than the corner types.
Strain Partitioning in Subduction and Strike-Slip Areas
Strain partitioning, or the manifestation of transpressional plate motion into parallel and coeval systems of strike-slip and thrust faults, has been the topic of earlier Penrose Conferences and was addressed again at this session because of the common occurrence of transpressional suites of structures in all three types of transition area. Martin Reyners (IGNS, New Zealand) opened the session with a summary of earthquake studes of partitioned strike-slip and thrust deformation on the plate interface at the Hikurangi-Marlborough transition zone of the South Island of New Zealand. He showed that the intensity of strike-slip deformation is directly related to degree of plate coupling, with the tightly coupled southern transition area taking up 100% of the margin-parallel motion. Rob McCaffrey (Rensselaer Polytechnic Institute) reviewed the direct mechanical relationships that exist between strike-slip faults in the overriding plate and the shear stresses along the subduction thrust. These relations determine the existence and position of the strike-slip fault in the overriding plate and the regional pattern of slip vectors along the margin. Finite element models and natural examples suggest that strike-slip faults need not form in all zones of oblique subduction. Serge Lallemand reviewed 3-D analog experiments devised to study the various parameters controlling strain partitioning. He concludes that shear can be accommodated either on strike-slip faults in the overriding plate or along the slab interface if a weak zone of the lithosphere is present. The latter case can explain why some oblique subduction areas are not accompanied by strike-slip faults. Posters addressed problems of partitioning in southern Alaska, the Aegean, the Chile triple junction, New Zealand, Ecuador, the Caribbean, the Ryukus, and the western Carpathians. Panel discussion focused on problems of distinguishing collisional effects related to a bathymetric high from transpressional effects related to normal subduction processes.
Controls on Slab Behavior in Subduction to Strike-Slip Transition Areas
The rupture, breakoff, or interaction of opposed slabs at depth in the mantle, all processes observed in transition areas, formed the topic of this session. Jim Dolan provided an overview of evidence for the collision of the opposed North America and Caribbean slabs at depth in the HispaniolaPuerto Rico transition zone. The collision of the shallowly dipping Caribbean slab may force down the steeper dipping North America slab and result in a regional-scale subsidence of the forearc in the Puerto Rico area. Blanka Sperner (University of Karlsruhe) outlined a model for collision and slab behavior at depth in the Alpine-Carpathian system. Mantle flow around the subducting slab reduced the hydrodynamic suction between the overriding and subducting plates and caused the slab to steepen and roll back, as revealed by tomographic studies of the eastern Carpathians. Posters included detailed earthquake and geologic studies of the remnant eastern Carpathian slab and the slab of the northeastern Caribbean beneath Puerto Rico and eastern Hispaniola and beneath the South Island of New Zealand. Other posters dealt with the volcanic expression of ancient transition areas in the Caribbean, Mexico, and Alaska.
Earthquake and Tsunami Hazards in SubductionStrike-Slip Transition Areas
This final theme of the meeting, along with the field trip, addressed earthquake hazards in transition areas. Eli Silver (University of California, Santa Cruz) began the session with an overview of a TFF triple junction north of the island of New Guinea. Rupture of a thrust fault near this triple junction was responsible for the destructive earthquake and tsunami of July 17, 1998. A GPS network is being used to investigate the complexity of plate boundaries and triple junctions. Jim McCalpin (GEO-HAZ Consulting) provided an overview of damage resulting from historic earthquakes in transition zones worldwide. Posters related to this session addressed hazard problems in the Dominican Republic, New Zealand, and Kamchatka. In order to convey hazards-related information to the general public in the Dominican Republic, a special open house was offered at the end of the meeting. The open house was well attended by civil defense, military, political, and interested geologists from the Dominican Republic. The Penrose participants shared their impressions of seismic hazard issues to the visitors based on their experiences in this and other transition areas.
CONCLUSIONS AND FUTURE WORK
The final session attempted to summarize the diverse results of all six sessions by constructing matrix diagrams of the main geologic, geophysical, and earthquake characteristics of the three types of transition area proposed in the first session. Some main conclusions highlighted by the matrix exercise include:
- Open and closed transitions have properties that are quite distinct from the triple junctiontype transitions, and thus should be considered as separate phenomena.
- As a first-order observation, the close association of bathymetric highs and closed corners suggests that closed corners begin to form when highs indent subduction margins. The association of open corners flanking closed corners with bathymetric highs suggests that open corners may form by radial expansion of the arc system adjacent to the area of localized arc indentation.
- Within this overall genetic framework, salient characteristics of closed type transitions include: (1) the presence of an adjacent bathymetric high; (2) transpressional partitioning of deformation into thrust and strike-slip deformation and earthquakes; (3) a high degree of coupling between the overriding and underriding plates consistent with the presence of a subducted, thicker-than-average bathymetric high; and (4) associated deep bathymetric troughs flanking the closed corner. It is unclear if the deep bathymetric troughs associated with closed corners reflect tectonic erosion related to entry of the bathymetric high beneath the forearc or some other tectonic process.
- A salient characteristic of some open corner transitions is detached or vertical subducted slabs. Detached and vertical slabs appear to reflect the termination of the normal subduction process by the entry of a bathymetric high into the subduction zone.
- While these comparisons are crude, they do suggest an evolutionary process initiated by the indentation of a subduction margin by a bathymetric high and the expansion of adjacent open corners flanking the original closed corner. Evolutionary trends are clearest for intra-oceanic subduction systems where bathymetric highs are small in area. Entry of larger bathymetric highs including large areas of continental crust complicate simple models for transition areas.
- There is a paucity of strike-slip and very large earthquakes in the parts of open and closed transitions where plate models predict transform motion. This has important implications for hazard assessment. Is plate motion aseismic in these regions, or do the largest earthquakes simply have very long recurrence times?
- A priority area for future conferences and research is the tomography of crust and mantle structure in the open and closed regions and on the rheology and mechanics of the plate interactions within and adjacent to those areas. One area of particular interest for the study of the northeastern Caribbean and the Solomon Islands is whether the interaction of opposed slabs at depth control deformation in the overlying arc. This information would help clarify the earthquake hazard problem raised in number 6 above.
We thank the management and staff of the Puerto Plata Village Caribbean Resort for providing an ideal venue in an active and subaerial transition area. We also thank Lois Elms of Western Experiences for her organizational expertise, and Christine Herridge de Guerrero and Grenville Draper for much-appreciated help with logistics. We are also grateful to the various field trip leaders, who did an excellent job of organizing their respective stops, assembling a comprehensive field trip guide, and making the necessary arrangements for an informative and multidisciplinary field trip. Special thanks to Michael Rymer (USGS, Menlo Park) for satellite imagery used in the field guide. The conference was supported in part by National Science Foundation grant EAR-9814516, American Chemical Society Petroleum Research Fund grant 33800-SE, NASA, the U.S. Geological Survey, and the Penrose fund of the Geological Society of America.
| Ken Aalto
Thomas H. Anderson
Eusebio Lopera Caballero
Juan Alberto Chalas
Andrew J. Coleman
James F. Dolan
Luis Odenel Gomez
Nancy R. Grindlay
Larry D. Guenther
Christine Herridge de Guerrero
Robert Max Langridge
James P. McCalpin
Giuseppina Mattietti Kysar
Bernard Mercier de Lépinay
Juan Carlos Moya
Jaime Perez de Armas
Uri ten Brink
Jean-Paul van Gestel
Alberto M. Lopez Venegas
Christa G. von Hillebrandt