LECTURE 3 NOTES - PLATE TECTONICS (updated 09/22/97)
Instructor: Professor Barbara Romanowicz
Director of Seismological Laboratory
Office Hours: Thursday 2-4 pm , upon appointment only
475 Mc Cone Hall
PLATE TECTONICS
HISTORICAL BACKGROUND ON CONTINENTAL DRIFT
First systematic account 1912 Alfred Wegener in Germany:
"Theory of Continental Drift"
Resemblance of SAM and Africa
occurred to him that might have once formed a single continent which
later split
300Myears ago: supercontinent: "Pangea" Carboniferous-
splits 200-180 Myrs ago (Jurassic)
Africa/S America- Cretaceous 100-50M yrs ago
beginning of quaternary: Greenland/Norway
Himalays born out of collision of India/Asia
Before Wegener: leading theory of mountain formation:
contraction of earth from cooling ---> surface wrinkled.
However problem: observed folding /buckling much more intense than
predicted by wrinkling surface)
Supporting evidence for continental drift
fauna, flora
Distribution of earthworms and garden snails:
exchange across the atlantic--> former connection
Helix Pomatia found only in W Europe and EN America (In Wegener's time)
older families span south Atl.--> south opened first
also mammals
Before Wegener: land bridge theory
oceans formerly bridged by land masses which subsequently became submerged.
evidence for land connections between Africa, Australia, India, Madagascar
and (Antarctica, SAM, Australia, Africa)
No evidence for former land bridges, no remnant of submerged continental
granitic mass on ocean floor (older theories)
Geological structures
S Africa East to West : Permian belt: Cape Mountains (250-200Myrs)
---> Sierras de Buenos Aires
geologic agreements disappear after Cretaceus (100Myrs): Pyrennees, Atlas...
Caledonian (Norway/Scotland) --> Canadian/Appalachian: 350-400 Myrs
Glacial traces
Greenland/Scandinavia joined in early Quaternary
300M years Permo-Carboniferous
extensive glaciation found over every continent in S> Hemisphere even
near equator implies considerable part of Earth's surface under polar climate
In N. Hemisphere no trace of such glaciers, tropical climate
pack neatly glaciated areas into small polar cap
paleoclimatology
paleolatitude from fossils fauna/flora
coal in Antarctica now under severest polar climate-> must have once been
abundant plant life
poles moved?--> leads to contradiction
continents moved (Much more than poles)
Stumbling block: what force could move continents over the mantle
why would drift be confined to such a short period in earth;'s history?
---> convection within the mantle: lack of data in the 1930's
scientists turn away from continental drift...
MAGNETIC CLOCK
2500 yrs ago, Greeks discovered magnetic properties of lodestone
(magnetite, one of iron ores)
1000 yrs later, : Chinese have compass.
14th century brought to Europe and used as navigational aid
1600 William Gilbert (England):
"the whole Earth is a big magnet"
Earth's magnetic field is well described by a bar magnet at center
inclined 11 degrees from geographical axis
angle between geographic meridian and geomagnetic meridian: declination
angle between magnetic field and the surface: inclination
intensity: magnetometer (force exerted by earth's field on standard magnet)
unit : gauss Earth ~0.5 gauss (horseshoe magnet: ~10 gauss)
Permanent magnet model:
fatal defect: heat destroys magnetism, permanent magnetism of materials
lost beyond the temperature of the Ò Curie pointÓ
~500o C for most materials (below 20-30 km in the earth)
--->electric currents: dynamos
fluid core ---> self exciting dynamo
motion in the fluid core: convective (heat from residual radioactivity)
stray magnetic field interacts ---> electric currents ---> magnetic field
there are changes in the earth's magnetic field: declination,inclination,
field strength vary: secular variation (0.1degree per year)
Fossil magnetism
How to find out what magnetic field was Millions of years ago?
As hot magnetizable materials cool below the Curie point: become
magnetized in the direction of surrounding magnetic field: thermoremanent
magnetization
eg. volcanic eruption say 100 Myrs ago, lava solidified: record of
geomagnetic field 100 MYrs ago.
some sedimentary rocks also magnetized: magnetic grains aligned in the
direction of the magnetic field while falling ---> depositional remanent
magnetization
---> paleomagnetism
reconstruct history of the geomagnetic field
o reversals of geomagnetic field
o sea floor as magnetic tape recorder
every 0.5M years the magnetic field changes polarity taking ~1000 yrs to
do it
clearly indicated in fossil magnetic record of layered lava flows --->
magnetic stratigraphy ---> radiometric-magnetic time scale
causes of reversals not yet well explained
sea-floor as tape recorder
airborne magnetometers very sensitive (deveolped in WWII to detect
submarine, then used by oceanographers)
measure 2 things:
main geomagnetic field + local magnetic disturbance = magnetic anomaly
due to magnetized rocks on the sea floor
subtract main field ---> record of magnetized rocks
normal direction: positive anomaly
reverse direction: negative anomaly
amazing patterns: linear over hundreds of miles almost perfect
symmetry/crest of mid-ocean ridge
puzzle until 1963 Vine and Matthews:
evidence in support for sea floor spreading: --> how fast ocean opened up:
precise dating of reversals --> 7M years (radiometric method of dating
lava flows on land loses accuracy beyond that time)
oceans ---> extend to 200 M yrs
once reference established:
measure amgnetic field on sea floor
correlate pattern of reversals with refrence sequence ---> ages assigned
to different regions without examining rock samples!
isochrons: contours of age: time + amount of spreading
PLATE TECTONIC THEORY
Tectonic Plates
Earth has a cool, mechanically strong outermost shell: lithosphere
70-150 km thick, thicker under continents
lithosphere = crust+ uppermost mantle
7 main plates
largest Pacific Plate
small ones: Nazca, Cocos, Juan de Fuca
some only oceanic, some ocean+continent
Assumptions of theory of plate tectonics
1- Generation of new material occurs by seafloor spreading ---> new
oceanic lithosphere generated along mid-ocean ridges
2- New oceanic lithosphere cools down to form part of plate
3- EarthÕs surface area remains constant: sea floor spreading balanced by
consumption of plates elsewhere
4- plates are rigid: they transmit stresses over large distances without
buckling--> relative motion taken up at boundaries.
3 types of plate boundaries
Divergent
plates move away- new material added to lithosphere along mid ocean ridge
system. Only few places where exposed on continents: Afar. Iceland...
eq mag 5-6 typically
Convergent
Plates approach each other
most: oceanic trench, island arc systems of subduction zones
the heavier plate descends into the mantle (oceanic)
can trace down to 700 km (deep earthquakes along Wadati-Benioff zones)
Island arcs=volcanoes, deep sea trenches
deepest: Challenger (discovered 1870): 10,915 meters south of Marianas
when continental material reaches subduction zone, cannot be subducted
efficiently: subduction ceases, continent buckles: mountain belt (Himalayas)
Conservative
plates slide past each other along "transform faults" (SAF)
ridge-ridge most typical: few km to few 100 km.
Movements of plates described quantitatively in terms of rigid body
rotations (earth's surface= spherical shells): rotation axis: passing
through center of earth, intersections at surface called poles of
rotation.
Rotation vectors : length proportional to angular velocity between two
adjacent places.
How do we measure plate movements?
Classical geodetic methods (e.g. 1906 S.Francisco eq. used triangulation)
triangulation (measurements of angles -optical)
trilateration (measurement of distances -laser beams)
levelling : measurement of vertical move,emts
Since the 19th century there are grids of fixed locations across
continents (coordinated at the national level -e.g National Coast and
Geodetic Survey)
This grid allows to refer the measurements to the continental edge -->
use mean sea level as reference.
Current Methods
1- determine local relative motion between 2 plates from strike of active
transform faults= arcs of small circles about the rotation pole which
must be on the great circle orthogonal to small circle: if 2 or more
transform faults: intersection of great circles give position of rotation
poles.
spreading rate at transform fault (magnetic anomalies)---> angular velocity
2- earthquake data--> direction of motion and plane of fault=> direction
of relative motion between plates.
3- satellites: measure instantaneous relative plate motions
satellite laser ranging system:
difference in distance between 2 sites over periods of years
VLBI quasars: signal source,
terrestrial radio telescopes=receivers
difference in distance between 2 telescopes measured over periods of years
GPS: orbit at 20,000 km, revolution twice a day
broadcast signals giving time, position
receiver; determine distance from transit time
location within a few meters
relative location within a network less than 1 cm
relative plate motions can be resolved with much more precision within
the next decade.
Plate boundaries can change with time, form new plates, destruction of
existing plates.
plates can go down subduction zone, eg. Farallon under NA in early Tertiary
direction can change: Pacific Plate.
Absolute Plate motions
Isolated volcanic island chains in oceans, well away from plate boudnaries
chemistry of lavas different from mid ocean and subduction zone volcanoes
active volcano at one end
islands age with distance from active volcano
consistent with hotspot theory
Definition of hot spot: fixed places where melt rises from deep in the mantle
=> volcanic islands formed as plate moved over a hot spot.
Hawaii
Emperor=> change of strike 43 Myears ago
We can obtain absolute motions if we imagine that deep mantle moves much
more slowly than plates: "fixed"
get absolute motions from traces of ocean island chains or continental
volcanoes. On one plate, then propagate using relative velocities.
Works to some extent..
We can also take one plate as a referent plate (fixed one) and calculate
motion of the other plates relative to referent plate.
Triple junctions
3plates meet- different types
example : Mendocino triple junction; transform, transform,trench
slowly migrate towards the north: much of geological history of the area
over the last 30Myrs related to migration of triple junction
Reconstruction of past plate motions ....(Gondwana)