LECTURE 2 NOTES - BASIC GEOLOGICAL CONCEPTS (updated 09/09/97)
Instructor: Professor Barbara Romanowicz
Director of Seismological Laboratory
Office Hours: Thursday 2-4 pm , upon appointment only
475 Mc Cone Hall
BASIC GEOLOGICAL CONCEPTS
Lecture notes prepared by Professor Barbara Romanowicz
based on Chapter 2, Press and Siever.
Definitions: Geology, geophysics, seismology.....
Major motivation for the study of geology:
to understand and explain nature in order to gain protection from her
vagaries
Geology: how a planet is born, its course of evolution, how it works
today, based on analysis of exposed rocks at the surface.
Geophysics: use tools of modern physics to build realistic models of
earth processes (thermodynamics, mechanics, magnetism...)
Seismology: a particular branch of geophysics: study earthquakes and
use seismic waves to infer properties of the earthÕs interior.
TIME SCALES
Earthquakes happen on time scales of sec to min (Typically, a magnitude 7
earthquake would last 10-15 sec).
Most geologic processes that shape the surface and the deeper structure
operate on time scales of millions (M) and billions of years.
On time scales of 100-1000 yrs, the earth seems a reasonably stable
foundation (excepting earthquakes and volcanic eruptions).
10**6-10**8 (ten to the six (million) to ten to the eight (hundred million))
years : far less stable
Continents, occeans, mountain chains have moved horizontally and
vertically through large distances.
Only recently have scientists begun to recognize a worldwide pattern in
these movements.
Why do we care about time-scales?
We want to be able to reconstruct what happened when; in particular it
will enable us to project what will happen in the near/far future. E.g.:
When did Appalachian or Rocky mountains form?
When did the Atlantic Ocean open?
How do we estimate slow rates of earth processes?
o Some simple calculations:
1) No rocks older than ~200M years found on the deep sea floor.
200 Myears upper limit for age of ocean basin
10,000 km width of ocean: this gives the distance plates have spread to
form basin. We can therefore infer the spreading rate of the ocean to be
10,000 km/200M years or about 5 cm/yr.
We can then compare with independent measurements elsewhere: e.g. a
survey of the San Andreas fault in the last century indicates
displacements along this fault of about 4-6 cm/yr. Matching up
distinctive geological formations that have been split by the fault, on
the other hand, indicate ~1cm/yr over the past several millions of years
in central California. At this rate, 25 M years ago, San Francisco was at
the latitude of LA.
Therefore, we can infer that the rate of horizontal movements related to
plate tectonics is on the order of several cm/yr
Later more precise: dating of magnetic patterns, sediment samples in
Deep Sea Drilling Project (DSDP)
2) vertical movements: date marine deposits that are now above sea level
mountains raised 3000 m in 15Myrs => 0,2 mm/yr; likewise erosion rates
are on the order of 1mm /yr
3) postglacial uplift
Fennoscandia: 40,000 years ago, it was under 3000 m of ice: ice started
melting, and the region was uplifted 500 m (up to present): hence the
rate of uplift os
about 1cm/yr
200 M yrs to open an ocean
20M yrs to raise a mountain
100 M yrs to erode it to sea level
These are still short time scales: the earth has experienced many cycles
of mountain building and erosion, opening and closing of oceans, in 4
billion yrs since its formation.
How do we date?
Rocks exposed at the surface are the visible records of geological
processes of the past.
1 -Time space relations between rocks of different types serve to build
the geologic time scale , which is used to place geologic events of
earth history in sequence according to relative age.
2 - Spontaneous decay of radioactive atoms in rocks give absolute
ages that date the geological periods and the origin of the Earth.
Three types of rocks exist on the earth's surface:
- sedimentary rocks : formed by deposition at the bottom of the sea
- igneous rocks: erupted in conjunction with volcanic events
- metamorphic rocks : sedimentary rocks that have been pushed back deep
into the crust by tectonic forces and subjected to high pressure and
temperature conditions (they have been ÒcookedÓ at high pressures) which
have changed their structure, before being pushed back up towards the
surface.
Stratigraphic time scale
based on outcropping sedimentary rocks, eg. Grand Canyon
horizontal stratification : particles settled from air or water to form
sediments assumption: layer on top deposited after layer on bottom (if
series not deformed and overturned)
=> stratigraphic time scale
The stratigraphic time scale (1669, Nicolaus Sten, Dane in Italy) is
based on the principle of continuity: observation that deposition
proceeds as continuous sheets ending at barrier such as shoreline,
gradually thinning to change to bed of different composition.
If we knew how long each bed takes to deposit and if all time was
accounted for by the sum of beds, we would have a clock.
But some missing time periods: interruption in sedimentation, may be very
long (eg intervals between floods).
Fossils
Some sedimentary rocks (eg limestones, CaCO3, or sandstones, or shales)
contain much fossil animal shells, which can be identified.
We find different assortments of species in different layers.=>
formations: grouping of layers that are the same stratigraphic age
and contain materials that have the same physical appearance and properties.
It turns out that fossil assemblages can be used as fingerprints of
formations. (19th century, William Smith, engineer, mapped geological
formations in England)
1859 Charles Darwin theory of evolution, very important: provided
theoretical framework which gave support to the idea that time-related
changes in fossil species could be used to set up a stratigraphic time scale
evolution + travel 19th century geologists -> mapped the earth's surface
and fitted together the Phanerozoic time scale (last 600 M years)
Names: most often names of places where characteristic formations found:
Jurassic: Jura mountains in France
Carboniferous: coal bearing sedimentary rocks in Europe and N. America.
Absolute time and geological time scale
How many years in stratigraphic time scale?
Debated since the ancient Greeks, who already recognized that sedimentary
rocks were formed by deposition and that fossils were remnants of ancient
life, that rocks were very old and the earth much older.(e.g. formation
of Nile delta must have taken many years, since each flood only deposits
a thin layer of alluvial sediments. After the Greeks, and for a long
time, through middle ages and into 18th century, reference: Genesis (4000
years), much too short for geologists.
Buffon (mid 18th century, french) studied melting and cooling of iron
balls, and computed how long it would take for molten earth (made of
iron) to cool ---> he found 75,000 yrs (we now know this is not correct)
and generated heated debate (too long for fundamentalists, too short for
geologists).
1854 Helmholtz, interested in determining the age of Sun and Earth.
Reasoned that if the sunÕs light came from ordinary combustion, it would
take 1000 years to cool off. Proposed that heat of the sun was due to
gravitational contraction (as particles fall to the center, they release
potential energy in the form of heat). Got 20-40 M years as an estimate.
Still too short for geologists. Some other source of energy had to be
found....
--> 1895, Henri Becquerel discovered radioactivity in Uranium salts,
soon after Marie Curie discovered radium. The stage was set for the use of
clocks that are built into the nuclei of radioactive atoms.
Consequences were drawn by Rutherford (1905) who suggested that
radioactive minerals could be used to date rocks. He found that the earth
is Billions of years old.
In the 1920's and 1930's astronomers and physicists recognized that the
immense energies liberated by nuclear processes are responsible for the
luminosity of the Sun and that heat given off by radioactive decay was
the main explanation for the heat trapped in the earth.
Clocks in rocks: radioactive atoms
Atoms of radioactive elements spontaneously disintegrate to form atoms of
other elements, liberating energy in the process.
This is a dependable means of keeping time, since the average rate of
desintegration is fixed (for an element), it does not change with changes
in physical or chemical conditions.
Once a quantity of a radioactive element is created, somewhere in the
universe, the clock starts to tick, steadily firing off atoms at a
definite rate:
all we need is to count...
(the rate is defined as the ratio, over a given time period, of the
number of atoms that have desintegrated to the number of radioactive
atoms left)
parent elements : the original radioactive elements
daughter elements : the product of the disintegration
If we can identify and count daughter element atoms and know the rate of
decay we can work back to the time when there were no daughters but only
parents. Specialists who made this practical: geochronologists
atom: nucleus + cloud of electrons
nucleus : proton (+e) + neutron (neutral)
In a complete atom the # of electrons = # of protons
atomic number = # protons : it is unique to the element: all atoms
of an element have the same atomic number
atomic mass = # protons + neutrons
eg C (carbon) atomic number N= 6; atomic mass is 12 for the most common
stable case, noted: 12C
Different isotopes have different numbers of neutrons:
C: 6,7,8
heavy isotopes have more neutrons than protons (they
have a larger atomic mass)
Isotopes are distinguished by their atomic masses: 12C 13C 14C
12C and 13C are stable: do not spontaneously disintegrate
14C radioactive: decays spontaneously to N (nitrogen): the extra neutron
splits into a proton and an electron. It changes the atomic number (hence
the element) but not the weight (electron is practically weightless)
therefore not the atomic mass.
14C ---> 14N + b
b = (proton+electron)
87Rb -----> 87Sr + b
Rubidium --> Strontium
Different rates for different elements
Rate is stated in terms of the half-life:
time required for one-half of original # of radioactive atoms to decay:
end of one half life: 1/2 left
2nd : 1/4 left
3 rd : 1/8 left
Elements differ by the half lives of their isotopes
14C : half life 5570 yrs
87Rb : 47 Billion yrs
explains why 14C is commonly used to do timekeeping for the last 30,000
years of earth history (now 70,000)
87Rb : choice for dating very old rocks
U is also used
235U ----->207Pb + 7 4He (4He = a=2 protons+2electrons)
238U ----->206Pb + 8 4He half lives of 10**8-10**9 years, suited for
dating the oldest objects in our solar system
also:
232 Th -----> 208 Pb + 6 4He
40K ----->40Ca + b 89% atoms
40K + e ------> 40Ar 11% of atoms (this one used because 40Ar
distinguished easily from Ar formed in other ways)
Reading the clocks
At first: U,Th : require only ordinary chemical analysis for U and Pb,
but cannot distinguish lead originating from U or from Th.
1920-1930: mass spectrometer
produce beam of electrically charged atoms from a sample. Pass through
electric+magnetic fields in such a way that atoms are deflected by an
amount that depends on their masses: isotopes can be separated, nowadays
with great precisions.
other methods: 14C measured directly (# of atoms in particle accelerators)
When were the clocks started?
Once materials in a rock are formed , radioactive elements start ticking
away. We measure the time elapsed since the radioactive parent element
became part of a rock from which daughter elements could not escape
(parent AND daughter trapped in the rock):
--> date of crystallization of rocks. Then by inference, date of
formation of other rocks that bear definite age relation to rock analyzed
eg granite:
we know that the surrounding sedimentary rocks into which granite has
intruded must be older:
--> bracket absolute age of stratigraphic sequence---> date sedimentary
rocks.
possible errors:
must be sure that no removal of daughter element
(solution by groundwater of some Pb formed by U decay for example)
heating or partial melting: geological event can reset clock by allowig
daughter element to escape
K-Ar : a gas can diffuse out of solid
Age of meteorites
Gradually it became apparent that all meteorites are of the same age ~4.5
10**9 yrs, regardless of composition (stony or iron)
this suggests strongly that they originated in bodies of the solar system
that formed at the same time as the earth did
----> age of the earth (provided no geological event came to reset clock)
Radioactive clock: fine watch
Accurate enough to distinguish formations that are only a few meters
thick, less than 1M yrs, and determine relative time to the nearest 10M
years.
combination clock: radioactive + stratigraphy
| |
igneous sedimentary + fossils
rocks rocks
lead to the dating of the phanerozoic time scale:
Arthur Holmes, british geologist, (~1915) very accurate, holds to this day
Phanerozoic time scale : 600 M years
unequal divisions due to accidents of choice predating radioactivity
Paleozoic : 350-450 M years long, names in ÒianÓ (e.g. Permian)
Mesozoic: 150M years long, Triassic, Jurassic and Cretaceous
Cenozoic: 65-70M years long, names in ÒceneÓ (most recent, Holocene)
KTB boundary at 65M years ago: ÒCreataceous Tertiary BoundaryÓ, marks
dinosaur extinction.
Precambrian:
no fossils to rely on
rethinking of how precambrian fits together since radioactivity dating.
much more discontinuous record: can date only episodes of igneous
intrusion, metamorphism: spasmodic events
accuracy still lower than stratigraphic dating: +- 100 M years, lack detail.
ORIGIN OF THE EARTH AND SYSTEM OF PLANETS
Find a model that explains:
-mass, size of planets
-peculiarities of orbits
- relative abundance of elements in planets and Sun
planets: all in same direction around the sun
elliptical orbits (almost circular) almost in the same plane
most moons also revolve in the same direction
all planets except Venus and Uranus rotate in same direction as
revolution around the Sun
each planet is about twice as far as the next inner one from
the Sun (Titus-Bode rule)
2 groups:
Terrestrial: Mercury-Mars: small,rocky,dense 4-5.5 (g/cm^2)
Jovians (giant): Jupiter-Neptune: gaseous 0.7-1.7 (g/cm^2)
Sun has most mass (99.99%), planets most angular momentum
Terrestrial planets 90% Iron, Oxygen, Silicon, Magnesium
Sun 99% Hydrogen, Helium
Nebular hypothesis
1755 Kant: primeval slowly rotating cloud of gas (nebula) which
condensed into discrete globular bodies.
This explains revolution + rotation directions
1796 Laplace ~same, but no mathematical formulation
Moulton (Chicago 19th Century) showed that this violated the fact that
planets have most angular momentum (conservation of angular momentum: Sun
should rotate faster)
Collision hypothesis
Buffon 1749 revived by Chamberlain/Moulton in XIXth century
giant tongues of material torn out of Sun by gravitational attraction of
passing star
---> planetesimals
by collision and graviational attraction larger pieces sweep up smaller
pieces
---> planets
- flaws: hot material (1MoC) would have been dispersed through
space before it could form planetesimals.
Recent theories of origins of earth and planets
Chondrites, clay-like minerals, most solar like chemical composition of
all primitive meteorites
Detection of interstellar matter: 99% gas H,He and 1% dust, similar to
terrestrial planets => Si, Oxides, ice crystals plus organic molecules
Abundance ratios in Sun, chemical fractionation and processes which occures in
Solar Nebula
---> favor primordial cloud that would have collapsed to form Sun and planets
(that cloud is called "Solar Nebula')
Evolution of Earth
Initially homogeneous body ----> differentiated planet
Differentiation: related to formation of atmosphere, oceans, continents,
mountains, volcanoes and magnetic field.
Age of the Earth ~ 4.56 billion years (dated from Allende meteorite,
Pb isotope age)
Unsorted conglomerate of Si compounds, Fe, Mg oxides small amounts of
other chemical elements
3 different effects lead to heating up:
1) each planetesimal: energy of motion converted into heat upon impact
4000 km planetesimal
30 km/s ---> impact generates as much energy as 1 kiloton
nuclear explosion
significant portion of this heat is retained by planet
2) compression also leads to increase in T
(e.g. inflating bicycle tires with pump)
3) Radioactive desintegration
U, Th
K few parts per million
profound effect ---> radioactivity
desintegrate emitting atomic particles absorbed by surrounding matter
---> heat
eg 500 Myrs to brew up a cup of coffee with heat from 1 cm3 granite
several billions of years ---> temperature rise to melting point of granite
4) also: outward flow of heat slowed by low thermal conductivity of rocks.
Currently accepted scenario: when earth was formed 4.7 Billion years ago,
heat generated by accretion and compression leads to an estimate of the
temperature around 1000 oC. The radioactivity takes over and temperature
rises so that melting point of iron (T~3000 C) is reached 400-800 km
depth after 1 billion years (large fraction of the earth then melts,
since other elements have lower melting points in general)
Iron melts heavier ---> falls towards the center displacing lighter materials
abundant (1/3/mass of earth) => event of catastrophic proportions,
releasing huge amount of gravitational energy, converted to heat.
Differentiation:
molten material lighter, moves upward --> crust
the Earth transforms from homogeneous body to zoned, layered, with dense
iron core, surficial crust made of light materials with lower melting
points (eventually continents)
in between: mantle
escape of gases --> atmosphere and oceans
early continents if they existed --> engulfed: no trace
best estimates: 3.7 to 4.5 billion years ago: oldest rocks we can find
were then formed.
Chemical Zonation
90% of Earth: Fe, O, Si, Mg
Core: 90% Fe, Ni, Co (Cobalt) + some light elements like Oxigen
Mantle+Crust: Silicates (Si,O,Mg,..) (olivine, peridotites...)
comparison crust /core
vertical arrangement not entirely based on relative weights because there
are compounds
distribution goverened by densities, melting points, chemical affinities
of compounds
e.g. feldspar (granites, basalts)
CaAl2Si2O8
KAlSi3O8 easily melted (700-1000oC)
NaAlSi3O8
---> most common minerals in the earth's crust.
mantle: reservoir for magnesium iron silicates (peridotites)
olivine: Mg2SiO4- Fe2SiO4
pyroxene: MgSiO - FeSiO3
perovskite: (Mg,Fe) SiO3
Heavy elements: gold, platinum little affinities ---> core?
U, Th: form oxides, silicates ---> accumulate in the crust, particularly in granites
Differentiation slows the engine,
radioactive fuel ---> crust
heat then conducted through smaller thickness to surface, lost more easily
Convection takes over ----> plate tectonics
INTERNAL HEAT OF THE EARTH
Is the Earth heating or cooling?
Heat flows from the earth: excepting the heat from the Sun, this is the
most important energy source:
2 x 10**20 cal = 10**28 ergs/ yr reaches the surface
this is 3 times as much as we currently use
solar heat much larger (5000 times)
Internal sources of heat:
1)gravitational--->thermal energy (Earth's differentiation)
2)radioactive heating: a layer of granite 20 km thick around the earth
would contain enough radioactive elements to produce the observed heat.
Methods of heat transfer:
1- Conduction
Heat energy in solids= vibration of atoms
Transfer of heat through a material by atomic/molecular interaction
within the material.
Depends on thermal conductivity of material, small for rocks
2 - Convection
Molecules themselves move from one location to another within material:
important in liquids and gases.
Much faster way to transfer heat than conduction
happens naturally when liquid or gas is hotter at the bottom than at the top
convection can occur in solids: heated rock moves slowly upwards because
its density decreases as it expands. The oslid becomes a highly viscous fluid
----> driving force for continental drift
Once this started in the Earth, heat quickly dissipated --> planet
cooled, mantle solidified, core did not (higher melting point plus no
time yet too cool off)
3- Advection
special form of convection
if hot region uplifted by tectonic events, heat physically lifted up with
the rocks: it is advected
4 - Radiation
Direct transfer by electromagnetic radiation (body heated begins to glow)
minor factor within the earth
Conduction not very effective: it would take 5 billion years for heat to
flow by conduction across a 400 km thick rock plate.
Patterns of heat flow from the interior of the earth:
T difference=> gradient=> heat flow from hot to cold. Heat flow depends
on thermal conductivity of rock: Terrestrial heat flow represents the
quantity of thermal energy that the earth is losing (per unit area per
unit time).
heat flow per unit area = rate of T increase with depth (temperature
gradient) times thermal conductivity.
average on the Earth ~ 65 mW/m^2 (65 miliWatts per meter squared) or
1.5x10^-6 (1.5 times 10 to the -6) cal/cm^2/sec (calories per centimeters
squared per seconds)
Measure internal temperature and its increase by drilling holes and
lowering thermometers into them.
At surface: T fluctuates constantly: day/night, seasonal variations: due
to heat from Sun, effect becomes negligible at depth > 30 m. In the
oceans, it is easier to measure, a few meters hole is enough because the
ocean water isolates from effects of sun, surface temperature
approximately constant, 1 oC, also no need to drill because there are soft sediments.
Measurements in mines, wells, tunnels: T increases with depth, everywhere
Geothermal gradient: rate at which T increases with depth
non-volcanic regions : 3 oC per 100 m of depth, but highly variable
(1-5 oC/100m)
If assume, gradient continues at same rate: 200,000oC at center of earth.
systematic measurements started 1930's (Bullard).
Interesting results:
Continents: old (3-4 Billion years, granitic crust)
for all continental regions in the world , heat flow more or less the same:
1.5 x 10-6 cal/cm^2/sec=65 mW/m^2 on average (same as above)
We can determine which part of the heat flow originates in the crust, due
to radioactive decay in granite.
By plotting the measured heat flow q (cal/cm2/sec) versus radioactive
heat production in surface rocks (measured in the lab, per unit volume) A
(cal/cm3/sec- volumetric):
q ^ (heat flow on vert. axis)
|
| /
| /
| /
| /
| /
| /
| /
|/
qo/
/|_____________________. A (heat produc on horiz. axis)
0
Straight line doesn't go through zero, so even when A is zero, some q exists.
q = qo + d x A
where d is the thickness of the granitic layer
and qo is the heatflow from beneath the crust
d ~ 10 km throughout continent
Canadian shield : qo = 0.7 microcalories/cm^2/sec
average q = 0.9 microcalories/cm^2/sec --- > 70% originates
from below the crust
Basin and Range Province (young <65 Myears)
qo = 1.4 microcalories/cm^2/sec
q = 2 microcalories/cm^2/sec
also about 70% is a deep source, and more heat supplied from the deep
source than in Canadian shield: rising thermal plume under Basin and Range?
Oceans (young <200 M years; basaltic crust)
Pattern as in continents:
mid-ocean ridges : age <5 Myears q>3 microcal/cm^2/sec
Ocean basins 50-100 Myears old : q ~ 1.4 microcal/cm^2/sec
old ocean platforms > 125 Myears < 1.4 microcal/cm^2/sec
Average ~ 2 microcalories/cm^2/sec or 101 miliWatts/m^2
Global average 87 miliWatts/m^2
Function of age: fits cooling plate model
Fits in a general scheme of cooling of the oceanic lithosphere created at
the ridges (relation with age), accounts fo 60% of heat flow from the mantle
On average about equal heat flow from old continents and from old oceans.
(oldest ~ 50 mW/m^2)
In the continents: heat generated by radioactivity in the crust (granite)
In the oceans: Thin crust, basaltic, expected heat flow to be much less.
This puzzle attracted general attention to the thermal question of the earth.
If radioactivity is the cause, thin crust must imply composition
different beneath oceans and continents, but we know from seismology that
not that different below the crust. The puzzle can be solved by invoking
convection in the mantle and wide distribution of radioactive elements
within the mantle.
Knowledge about temperatures below the crust? Need to be inferred
indirectly, since no direct measurements below about 10 km. Invoke
laboratory measurements of properties of minerals (melting curve in T and
P domain), combine with results from seismology about deep structure.
Anticipating on later lectures:
100 km depth T~1000-1200 oC (near melting point)
>300 km (below the solidus ---> upper limit for temperature
Temperature:
400 km ~1500 oC
700 km phase changes ~1900 oC
core mantle boundary ~3500 oC
inner core ~4300 oC
Mass of the Earth: core -> 32%
lower mantle -> 48%
upper mantle -> 20%
crust -> 0.4%