Odekon M. Encyclopedia of paleoclimatology and ancient environments

Подождите немного. Документ загружается.

VOLCANIC ERUPTIONS AND CLIMATE CHANGE

Introduction

Volcanoes erupt fragmented mantle material into the upper

atmosphere. Only the lightest particles remain suspended, how-

ever. These tiny supplements to the ordinary air, water vapor,

and dust can strongly affect the chemical, radiative, and dyna-

mical properties of the atmosphere. Critical to assessing the

magnitude of these effects is knowledge of the sizes and chemi-

cal compositions of the volcanic particles, the total mass of par-

ticles injected, the vertical distribution of the injected particles,

the geographical location of the volcano, and the prior state of

the atmosphere. Detailed chronologies of modern volcanic

eruptions and of their spreading particulate clouds (called “dust

veils” or “dry fogs”) have been constructed in order to associate

known volcanic eruptions with changes of long-term weather

(“climate”). The volcano-climate connection can be studied

either empirically, by using actual climate observations, or the-

oretically, by computing a general circulation model (GCM) of

the atmosphere. Interest focuses on how global and regional

climates respond to volcanic forcing on various timescales.

Such an understanding of the consequences of historical erup-

tions can help to illuminate both the past and the future of

the atmosphere of our planet, especially under different

assumptions made about the frequency of large volcanic erup-

tions, the amount of solar variability, and the inputs of other

natural and anthropogenic sources of atmospheric pollution.

History

On the day after the great eruption of Mount Vesuvius in the

year 79, Pliny the Younger noted that the Sun appeared “lurid

as if in eclipse,” but this atmospheric effect was probably only

local. More widespread hazes after other eruptions during anti-

quity(e.g.,Etnain44

BC), although noted, were never associated

with volcanoes by anybody at the time. Benjamin Franklin is the

first person known to have inferred a relationship between the

regional state of the atmosphere (North Atlantic dry haze in

1783 and a cold winter in 1784) and a large volcanic eruption

(Laki, Iceland, 1783). Chemists at the time identified the dry tro-

pospheric haze as being primarily what we would today call sul-

furic acid (H

). The ancient manifestations of it could now

also be provisionally interpreted. However, it was the eruption

of the Indonesian volcano Krakatau 100 years later that brought

a volcano’s worldwide atmospheric effects to public attention: a

fading and unusual coloration of the Sun, Moon, and stars, and

an enhancement of the red twilight sky glow. A peculiar reddish

ring seen around the Sun was named after its discoverer, Sereno

Bishop, and the darkness of a total eclipse of the Moon in this

period was traced to volcanic particles in the Earth’supper

atmosphere by Camille Fl ammarion. The particles were

thought to be silicic ash and possibly sulfurous gases. The

zonal circulation of the upper atmosp here (named the “strato-

sphere” by Teisserenc de Bort in 1900) was mapped by

visually following the westward drift of the volcanic cloud.

Only in hindsight were the unusual optical aftereffects of the

even greater eruption of Tambora, Indonesia, in 1815 now

associated with that event.

After a series of three large Caribbean eruptions in 1902 and

Alaska’s Katmai (Novarupta) eruption in 1912, C. G. Abbot

and W. J. Humphreys independently concluded that volcanic

eruptions produce temporary climate cooling, at least hemi-

spherically. Many studies since 1913, with much impetus com-

ing from Harry Wexler and H. H. Lamb, have confirmed this

fundamental result. In 1961, C. E. Junge and his coworkers dis-

covered an apparently permanent layer of sulfate particles in

the lower stratosphere, which, in 1973, A. W. Castleman, Jr.,

and his coworkers conclusively demonstrated were volcanic

in origin. Fresh stratospheric particles collected in situ by

S. C. Mossop after the Agung, Indonesia eruption of 1963

had already shown most to be sulfuric acid aerosols. Numerous

studies from the ground, air, and space were pursued after the

eruptions of Mt. St. Helens, United States (1980); El Chichón,

Mexico (1982); and Pinatubo, Philippines (1991). These results

have enormously increased our detailed knowledge of volcanic

aerosol clouds and their effects on climate.

Aerosol chemistry and physics

Although some small silicic ash particles remain suspended in

the stratosphere for several weeks after a powerful volcanic

eruption, the major constituents injected are gaseous: primarily

O, N

,SO

, and H

S, and some HCl and HF. With

both pre-existing and volcanically injected H

O, the sulfur

gases undergo a complicated series of heterogeneous chemical

reactions in the presence of sunlight to form sulfuric acid aero-

sols (75% H

and 25% H

O). These tiny droplets then

grow by coagulation and mutual collisions over a period of

weeks. Typically, their final mass density is 1.65 g cm

3

and

their complex index of refraction in the visible is m =1.43-0i,

indicative of negligible light absorption at short wave-

lengths. Their shape appears roughly spherical and their size

distribution can be described as approximately lognormal with

an effective (area-weighted) radius of r

eff

= 0.3 mm, increasing

with time to 0.5 mm, but this description is very crude and may

be inaccurate for some eruptions and for some layers of the

stratosphere. After the initial formation period, the size distri-

bution continues to evolve owing to horizontal dispersion of

the particles and to gravitational and convective sedimentation

of the larger particles. The role of sulfate aerosols in destroying

ozone (O

) involves heterogeneous chemical reactions with pri-

marily anthropogenic chlorine, and so would have been unim-

portant before the late twentieth century. Volcanic halogens

are mostly scrubbed out of the eruption column before reaching

the stratosphere.

Scattering of radiation by the aerosols leads to an attenua-

tion of the incoming solar beam, whose incident intensity is

. The optical depth of the light-scattering medium, defined

as t = ln(I/I

), can be calculated by multiplying the cross-

sectional scattering coefficient by the particle number density

and integrating the product over the path length of the direct

solar beam. Since t is wavelength-dependent, a standard refer-

ence wavelength is taken to be l = 0.55 mm, which lies in the

visible. Essentially, only the two quantities r

eff

and t

vis

are

needed, in practice, to compute the amount of radiative forcing

by an aerosol cloud. Reflection of some incoming solar radia-

tion back into space by the stratospheric aerosols reduces the

total insolation in the troposphere, which in turn lowers the

global-average surface air temperature. Absorption of long-

wave radiation from the ground, troposphere, and Sun by the

aerosols partially counters this effect, and manages to warm

the stratosphere. However, the net result of the competition

between the aerosols’ albedo and the greenhouse effect for

the troposphere is always a cooling.

976 VOLCANIC ERUPTIONS AND CLIMATE CHANGE

The total mass of the stratospheric aerosol veil, M

, can be

computed from the global-average stratospheric optical depth

. It is given by M

= 150t

megatons (Tg).

Records of volcanic eruptions

Catalogs of known volcanic eruptions during the Holocene

epoch (the past 10,000 years) have been published by the

Smithsonian Institution, making use of either eyewitness reports

or geological studies of pyroclastic deposits around the volcano.

Even earlier eruptions are known in a few cases from conspicu-

ous eruptive deposits. These can be dated radiometrically or stra-

tigraphically. They include the huge continental and oceanic

flood basalt eruptions of the past 300 Myr, of which about a

dozen have been recognized. The flood basalts were primarily

effusive in style and sulfur-rich (like Hawaii’s Kilauea), whereas

most known volcanic eruptions have been of explosive type and

silica-rich (like Indonesia’s Krakatau).

All other methods of detecting volcanic eruptions are indir-

ect. One method uses sulfate acidity and tephra fragments

deposited from the stratosphere onto polar ice fields and incor-

porated in the annual layers of polar ice. Tephra found in sea,

lake, and bog sediment cores (usually sited not far from the

erupting volcano) can also be used. Documentary records of

unusual atmospheric optical effects and even of abnormal

regional cooling have proven especially valuable, because his-

torically derived dates going back to 300

BC are typically accu-

rate as to year and month. Ice-core dates are available for

eruptions as far back as 100,000 years ago, but they begin to

degrade in accuracy before

AD 1000.

Quasi-periodicities in global volcanism occur on various

timescales. A weak annual cycle exists, perhaps coupled to

the hydrological cycle. Two other periods, 11 year and 80 year,

are probably solar in origin, through the influence of solar

variability on the atmosphere. Slower cycles of 10

–10

year

may be due to the influence of climate change on volcanism,

arising from Milankovitch periodicities in the Earth’s orbit that

affect the amount of insolation and hence glaciation and

sea level. Tectonic plate movements may also produce spurts

of volcanism, at intervals of a few Myr. Flood basalt eruptions

apparently follow a roughly 30-Myr cycle of uncertain origin.

Even slower cycles may exist, e.g., a 300-Myr cycle of tecton-

ism and associated volcanism.

Explosive strengths of volcanic eruptions have been esti-

mated from such observable quantities as volume of pyroclastic

ejecta and height of the eruption column. An example is the

Volcanic Explosivity Index (VEI) of Newhall and Self

(1982). The older Dust Veil Index (DVI) of Lamb (1970)

includes solar-beam transmission data and surface air tempera-

ture decreases in addition to the near-source tephra volume.

Robock and Free’s(1995) newer Ice-core Volcanic Index (IVI)

utilizes solely sulfate acidity in polar ice cores. The VEI of

course does not take into account the variable sulfur con-

tent of volcanic tephra, while the DVI is an uneven and

mixed indicator, and the IVI suffers from the special restrictions

of the ice-core method, including the highly uncertain rates of

aerosol deposition and the necessary corrections for non-

stratospheric transport of aerosols and for the volcano’s latitude

(often unknown). Statistically, however, all indices are found to

be correlated with t

whenever t

is available, and therefore

they can be used to estimate t

when this is otherwise unknown.

Table V1 contains the stratospheric aerosol masses esti-

mated for Pinatubo-class and larger volcanic eruptions of the

past 75,000 years. The peak visual optical depth over Europe

is also listed. The table is incomplete before 1883 and becomes

more so the farther back in time one goes, especially for the

Southern Hemisphere. Flood basalt eruptions, which are not

listed, probably produced 10

Tg of aerosols; typical examples

are the Columbia River Basalts (16 Myr ago) and the Deccan

Traps (65 Myr ago). The cumulative effect of smaller, but more

frequent, eruptions could have perturbed the Earth’s radiative

balance during some periods.

Movements of volcanic aerosol veils

Volcanoes are not distributed uniformly, or even randomly,

over the globe. They are found in regions of exceptional geolo-

gic activity, e.g., along the “ring of fire” around the Pacific

Ocean and along the Atlantic Mid-Ocean Ridge. Flood basalt

eruptions occur over mantle plumes and hotspots. For purposes

of climate studies, however, all volcanoes can be divided sim-

ply into extratropical (latitudes above 30



) volcanoes and tropi-

cal (or equatorial) volcanoes.

Stratospheric aerosol veils from tropical volcanoes are dis-

tributed by zonal winds very quickly so that they cover all

longitudes in only 2–3 weeks. Their meridional dispersion is

Table V1 Major aerosol-producing volcanic eruptions of the past 75,000 years

Volcano Latitude Year DVI/ E

max

VEI IVI European t

vis

Aerosol M

(Tg)

Pinatubo, Phillippines 15



N 1991 1,000 6 0.2 30

Agung, Indonesia 8



S 1963 800 4 0.06 0.03 20

Santa Maria, Guatemala 15



N 1902 600 6 0.05 0.2 30

Krakatau, Indonesia 6



S 1883 1,000 6 0.12 0.2 44

Tambora, Indonesia 8



S 1815 3,000 7 0.50 1.3 200

Laki, Iceland 64



N 1783 2,300 4 0.19 2 200

Huaynaputina, Peru 17



S 1600 >1,000 5 100

Unknown Tropical 1258 400

Eldgjá, Iceland 64



N 934 4 130

Unknown Northern? 797 1 40?

Unknown 626 100?

Unknown 536 2.5 300?

Etna, Italy 38



N44BC 3+ 200

Unknown 145

BC 80?

Thera, Greece 36



N 1640? BC 6 200

Toba, Indonesia 3



N 74000? BC 3,000?

VOLCANIC ERUPTIONS AND CLIMATE CHANGE 977

much slower (usually taking a few months), depending on the

season of the year, the phase of the quasi-biennial oscillation

(QBO) of the tropical stratospheric winds, and probably other

factors. For unknown reasons, cross-equatorial spreading is

occasionally inhibited, sometimes remarkably so (e.g., after

El Chichón and especially after Agung). Almost complete halts

of the poleward flow regularly occur at the Hadley cell circula-

tion barrier (near 30



) in the summer and at the polar vortex

edge (near 70



) in the winter. In this cyclical, seasonally depen-

dent way, the equatorial reservoir of aerosols (discovered after

the Agung eruption) replenishes the higher latitudes during the

course of the year. Simultaneously, aerosols are constantly fall-

ing out of the stratosphere, the average time for depletion of

the dust veil by a factor of e being 1 year. Once through the tro-

popause, aerosols are swept to the ground in a few days by either

wet or dry deposition. In the wintertime polar vortex, too, they

descend quickly in spite of a lower tropopause near the poles.

High-latitude volcanoes generate stratospheric aerosol veils

that spread equatorward only as far as about 30



. Thus, they

do not produce full hemispheric coverage (this was first rea-

lized after Laki and confirmed after Katmai). Using both Arctic

and Antarctic ice cores, an unidentified volcano that erupted

somewhere on the globe can usually be characterized at least

as being tropical or extratropical, with always the possibility

of error in an Agung-type situation.

Climatic effects (general)

Underneath the ash cloud in the volcano’s vicinity, the surface

air temperature initially rises but then falls precipitously as sun-

light is blocked off. Such an extreme cooling happened notice-

ably after Laki, Tambora, and Krakatau. If the ash cloud either

is very thin or resides at a high altitude, as was the case after

Mt. St. Helens in 1980, the initial warming caused by the blan-

keting effect does not occur.

Feedback from the stratospheric aerosol veil on the global

atmosphere is fairly complex. The simplest element in the feed-

back is a global-average cooling – as theoretically expected

and observationally confirmed. Finer observational analysis,

based mostly on compositing the effects of several modern

eruptions, reveals more structural detail, both temporal and spa-

tial. A prompt cooling (at least in the hemisphere of origin)

lags the eruption by about 1 month and lasts 3–6 months.

A rapid recovery is then followed by a prolonged cooling for

1–5 years, beginning with a comparable temperature drop in

the year of the first post-eruption winter (but sometimes later).

This first winter can be very cold for extratropical eruptions,

but for tropical ones there occurs a slight winter warming over

Northern Hemisphere continents at middle and high latitudes.

This is caused by a faster zonal advection of warm (and wet)

marine air that is associated with a strengthening of the strato-

spheric meridional temperature gradient and stratospheric polar

vortex. The more rapid response of the land than the sea to

radiative forcing is another factor that can explain the smaller

winter warming in the land-starved Southern Hemisphere.

If the surface air temperature is arbitrarily assumed to have

some power-law dependence on the transmitted intensity of

the solar beam, say T / I

, then DT / t for t << 1. In practice,

the drop in hemispheric-average temperature amounts very

roughly to DT (



C) = 1.5 t

for eruptions up to Pinatubo-class.

However, average temperature is given by DT / t

1/2 for

significantly larger eruptions. The cooling is usually accompa-

nied by increased cloudiness and precipitation.

Complications confuse this tidy picture, especially for very

small eruptions whose climatic effects are easily masked. The

El Niño-Southern Oscillation (ENSO) phenomenon, a quasi-

cyclical warming of equatorial Pacific waters every 3–7 years,

induces teleconnection effects elsewhere, notably an atmo-

spheric warming over North America. Although its cause is

uncertain, it is very doubtful that a sudden volcanic cooling is

able to trigger an El Niño warming event. Outbreaks of cold

polar air into mid-latitudes, on the other hand, certainly follow

an eruption. The Icelandic pressure low and the northern jet

stream move southward, for example; thus, the North Atlantic

Oscillation (NAO), which is part of the overall Arctic Oscilla-

tion (AO), shows a disturbance of its normal modal pattern.

The patterns of the NAO and SO consist of a cyclical shift in

relative sea-level pressure (and other meteorological variables)

between Iceland and the Azores (NAO) and between the Indian

Ocean and the South Pacific Ocean (SO). Other climate forcing

factors, however, can disturb these patterns. Natural variability

of the climate system, characterized by chaotic, unpredictable

behavior, also plays a role.

Climatic effects (specific eruptions)

Depending on the most characteristic method of dating or mea-

suring the aerosol veil, six different periods in Earth history

can be identified: geochemical, before 50,000

BC; ice-core,

50,000–300

BC; historical, 300 BC–AD 1880; pyrheliometric,

1881–1960; stellar-extinction, 1961–1978; and satellite, 1979–

present. The pre-instrumental years (before 1881) have been

divided into a prehistoric part (before 300

BC), for which indirect

geological approaches are used, and a historical part, employing

visual observations that can be calibrated by using modern tech-

niques. If the total mass of sulfuric acid aerosols has been esti-

mated by using either polar ice-core sulfate acidities or

geochemical analyses of the sulfur content of local volcanic

deposits, then the stratospheric turbidity can be computed from

= M

/150. During the historical period, visual observations

of solar dimming, of red twilight glows (dust-veil reflections),

of Bishop’s ring (aerosol diffraction pattern around the Sun),

and of lunar-eclipse darkening (an inhibited refraction of sunlight

into the shadow cone) can be used to obtain t

directly. The

optical methods and the ice-core method in the historical period

are about equally sensitive; the threshold aerosol mass for detect-

ing a tropical eruption is a Pinatubo-class cloud of 30 Tg.

The largest known aerosol veils during the historical period

occurred in 44

BC and AD 536, 626, 934, 1258, 1783, and 1815

(see Table V1), although the nature of the 536 veil is still con-

troversial. Eruptions producing these veils are known to have

been of explosive type in two cases (Etna and Tambora), and

probably also in the 1258 case, but of effusive type in at least

two other cases (Eldgjá and Laki). The crude climatic data

available for these seven eruptions strongly indicate that their

meteorological consequences followed the patterns for modern

eruptions. However, these older eruptions were large enough

to have had more severe impacts, including greater cold and

raininess, extensive crop damage, famine, and serious disease

outbreaks, as documented in written sources from the Mediter-

ranean and Middle East. For still larger eruptions, the strato-

spheric residence time of climatically significant aerosols may

be prolonged to 6 years or more. If so, tropospheric cooling

could be augmented by positive feedbacks such as a larger

planetary albedo arising from increased cloud, snow, and ice

cover. It is not known, however, whether such large eruptions,

978 VOLCANIC ERUPTIONS AND CLIMATE CHANGE

or even a long series of smaller eruptions, could cause an

abrupt climate change that lasts for decades or longer, although

this is thought to be theoretically possible in a metastable state

of the climate system. Possibly, the long cold period known as

the Little Ice Age (

AD 1400–1900) was volcanically forced, at

least in part, although a protracted solar minimum at this time

is a more likely cause.

Less is known about prehistoric eruptions. The protohistori-

cal eruption of Thera (Santorini), Greece, in the seventeenth

(possibly sixteenth) century

BC may have been of Tambora

class. Toba, Indonesia, erupted about 74,000 years ago in a

far greater paroxysm – a supereruption that exacerbated an

existing climatic downturn and probably caused extensive loss

of life, perhaps related to the human evolutionary “bottleneck”

around this time. Pleistocene glacial fluctuations have also

been attributed to volcanism.

At the top of the ladder are the flood basalt eruptions. These

outflows must have led to massive pollution of air, land, and

water, as well as to local darkness in the atmosphere for months

at a time, regional wildfires, and global cooling for many years.

Laki might represent a recent small-scale analogy. Flood basalt

eruptions appear to have coincided with mass biotic extinc-

tions, especially marine mass extinctions. Their release of

heat-trapping CO

would have been sufficient to counteract,

in part, the cooling brought on by the efficiently back-scattering

sulfate aerosols. Very early in Earth’s history, and during peri-

ods of rapid ocean-floor subduction and accretion, when vol-

canism was more active, outgassing of CO

may even have

been the predominant factor. Probably, outgassing supplied

the atmosphere and oceans with their main constituents after

the end of the massive early bombardment of Earth by comets

and asteroids (3.9 Gyr ago).

Cultural impacts of the greatest volcanic eruptions should

also be mentioned. Poetry has immortalized the atmospheric

aftereffects of Thera, Etna, Laki, Tambora, and Krakatau in

the Bible’s Exodus, Virgil’s Georgics, Cowper’s Task, Byron’s

Darkness, and Tennyson’s St. Telemachus. A few memorable

expressions have endured, such as “the year without a summer”

in 1816. “Tambora skies” and “Krakatau sunsets” are possibly

depicted in some paintings contemporary with those eruptions.

However, the major cultural impacts were probably negative,

reflecting the famines, epidemics, and social disorders that

grew with the climate deterioration.

Richard B. Stothers

Bibliography

Chester, D.K., 1988. Volcanoes and climate: Recent volcanological per-

spectives. Prog. Phys. Geogr., 12,1–35.

Clausen, H.B., Hammer, C.U., Hvidberg, C.S., Dahl-Jensen, D.,

Steffenson, J.P., Kipfstuhl, J., and Legrand, M., 1997. A comparison

of the volcanic records over the past 4000 years from the Greenland

Ice Core Project and Dye 3 Greenland ice cores. J. Geophys. Res.,

102, 26707–26723.

Crowley, T.J., 2000. Causes of climate change over the past 1000 years.

Science, 289, 270–277.

D’Arrigo, R., Jacoby, G., Free, M., and Robock, A., 1999. Northern

Hemisphere temperature variability for the past three centuries:

Tree-ring and model estimates. Clim. Change, 42, 663–675.

D’Arrigo, R., Frank, D., Jacoby, G., and Pederson, N., 2001. Spatial

response to major volcanic events in or about

AD 536, 934 and 1258:

Frost rings and other dendrochronological evidence from Mongolia

and northern Siberia. Clim. Change., 49 , 239–246.

Lacis, A.A., Hansen, J., and Sato, M., 1992. Climate forcing by strato-

spheric aerosols. Geophys. Res. Lett., 19, 1607–1610.

Lamb, H.H., 1970. Volcanic dust in the atmosphere; with a chronology and

assessment of its meteorological significance. Philos. Trans. R. Soc.

Lond. Ser. A, 266, 425–533.

Newhall, C.G., and Self, S., 1982. The Volcanic Explosivity Index (VEI):

An estimate of explosive magnitude for historical volcanism. J. Geo-

phys. Res., 87, 1231–1238.

Rampino, M.R., 2002. Supereruptions as a threat to civilizations on Earth-

like planets. Icarus, 156, 562–569.

Rampino, M.R., Self, S., and Stothers, R.B., 1988. Volcanic winters. Ann.

Rev. Earth Planet. Sci., 16,73–99.

Robock, A., 2000. Volcanic eruptions and climate. Rev. Geophys., 38,

191–219.

Robock, A., and Free, M.P., 1995. Ice cores as an index of global

volcanism from 1850 to the present. J. Geophys. Res., 100,

11549–11567.

Stothers, R.B., 1999. Volcanic dry fogs, climate cooling, and plague pan-

demics in Europe and the Middle East. Clim. Change, 42, 713–723.

Stothers, R.B., 2002. Cloudy and clear stratospheres before

AD 1000

inferred from written sources. J. Geophys. Res., 107 (D23), Paper 4718.

Zielinski, G.A., 2000. Use of paleo-records in determining variability within

the volcanism-climate system. Quaternary Sci. Rev., 19,417–438.

Cross-references

Atmospheric Evolution, Earth

Bolide Impacts and Climate

Climate Change, Causes

Climate Forcing

Climate Variability and Change, Last 1,000 Years

Flood Basalts, Climatic Implications

Ice cores, Antarctica and Greenland

Paleotemperatures and Proxy Reconstructions

VOLCANIC ERUPTIONS AND CLIMATE CHANGE 979

WEATHERING AND CLIMATE

Weathering is an important biogeochemical process that is both

influenced by and influences climate over the course of Earth

history. Weathering processes control the flux of solutes and

many nutrients to the oceans and the marine and terrestrial bio-

spheres, and the transfer of carbon from the ocean-atmosphere

system to sedimentary rocks. Weathering rates are dependent

on climate, among several factors, and vary widely across the

Earth’s surface. In turn, weathering processes can alter the

atmospheric concentrations of the important greenhouse gas,

carbon dioxide. Weathering processes thus have the potential

to act as a climate feedback, an idea that has been important

in our understanding of the long-term evolution and stabiliza-

tion of the Earth’ s climate for over 150 years.

“Weathering” refers to the chemical alteration of rock and

sedimentary deposits under sub-aerial conditions at the Earth’s

surface. While some analogous processes occur in the oceanic

crust during low temperature alteration by seawater, or during

higher temperature hydrothermal alteration of the continental

or oceanic crust, they fall outside the traditional definition of

weathering. The kinds of geological materials that undergo

weathering may be conveniently classified into silicates, carbo-

nates, and organic matter. The types of weathering reactions,

their products, and their dependence on and implications for

climate differ between the three.

Weathering of both silicate and carbonate minerals can be

described as an acid-base reaction, in which naturally occurring

acids generated by volcanic gas emissions, weathering of sul-

fide minerals, dissolution of CO

in water, or decomposition

of soil organic matter react with minerals phases to consume

acidity and generate dissolved cations and, in most cases, alka-

linity. One important difference between silicate mineral weath-

ering and carbonate mineral weathering is that with silicates, an

additional product of the weathering reaction is often a second-

ary alteration mineral (“clay”), whereas carbonate minerals

generally dissolve completely. In both cases, weathering acts

as a buffer to acidification, but the strength of the buffer

depends on the mineral phase(s) involved.

Of the natural acids, carbonic acid (a “weak” acid, implying

incomplete dissociation of the proton donor) and sulfuric acid

are the most important. Sulfuric acid can be generated by the

oxidation of sulfide minerals, volcanic gases (which commonly

contain sulfur dioxide), or biogenic volatile compounds like

carbonyl sulfide and dimethyl sulfide. These last reactions are

important for atmospheric aerosol formation but are not signif-

icant sources of weathering acidity. Carbonic acid is generated

by the dissolution of CO

in water:

þ H

O ! H

ð1Þ

! H

þ HCO



ð2Þ

Carbonic acid produced by the hydration of CO

(Reaction

(1)) will partially dissociate to produce acidity and bicarbonate

ion (Reaction (2)). At present atmospheric levels (PAL) for

(360 ppmv), the pH of water in equilibrium with the atmo-

sphere is ca. 5.6. pCO

in soils can be much higher due to the

decomposition of soil organic matter, and the pH of soil waters

consequently lower. At pCO

= 10 PAL, pH = 5.1; at 100 PAL,

pH = 4.6

Carbonate rocks (primarily limestones and dolomites) play

an important role in weathering fluxes at the Earth’s surface

because they are relatively soluble under typical conditions

and act as a strong buffer to either natural or anthropogenic

acidity. Natural acids react quickly with carbonate minerals,

neutralizing the acidity and generating alkalinity in the form

of bicarbonate and carbonate ions. An important reaction is:

þ H

O þ CaCO

3 calciteðÞ

! Ca

2þ

þ 2HCO



ð3Þ

in which CO

(from the atmospheric or from the decompo-

sition of organic matter) reacts with calcite and water to pro-

duce dissolved calcium and bicarbonate ions. Under open

system conditions and under present-day pCO

, the pH at equi-

librium with this reaction is near 8.6. Thus the hydrogen ion

activity of the water-carbon dioxide system, with an ideal pH

of ca. 5.6, is reduced by three orders of magnitude, illustrating

the strong buffering effect that carbonate dissolution has on

aqueous solutions. Bicarbonate ion is the dominant form of

dissolved inorganic carbon (DIC) at pHs between 5.7 and 10.2,

which covers the range of most natural waters. Carbonate

minerals are widely distributed, and most rivers have pHs that

reflect their influence. The strong buffering effect ensures that

waters in systems containing carbonate will generally be at

neutral to slightly alkaline pH, which has important effects on

the rates of other reactions involving alumino-silicate minerals

and the solubility of many metals.

The weathering of alumino-silicate minerals is generally a

more complicated process than for carbonates. Many reactions

can be written depending on the phases involved. One impor-

tant example is:

2NaAlSi

8 albiteðÞ

þ 2CO

þ 11H

O ! Al

OHðÞ

4 kaolðÞ

þ 4H

SiO

4aqðÞ

þ 2Na

þ 2HCO



ð4 Þ

which illustrates a weathering reaction involving primary albite

(Na-feldspar) and carbonic acid that yields a hydrated second-

ary clay mineral (kaolinite) which is depleted in cations and

silica relative to the primary mineral. The other products are

dissolved silicic acid, sodium ions, and bicarbonate. This type

of weathering reaction, in which a new secondary phase is

formed, is termed “incongruent” weathering, and is of funda-

mental importance for soil formation. Alumino-silicate mineral

weathering rates depend strongly on pH, temperature, and reac-

tive surface area, and are generally at a minimum at neutral pH.

Thus, the presence of carbonate minerals in a soil may greatly

reduce the rate of silicate mineral dissolution by buffering the

pH to a range in which dissolution rates are slowest. Labora-

tory studies have documented that, under ideal conditions,

reaction rates for alumino-silicate minerals increase by roughly

a factor of three for each 10



C temperature increase (Lasaga

et al., 1994). Field studies have produced much more ambigu-

ous results, probably in large part because of the difficulty in

defining the effective surface area and also because of the inhi-

biting effects of ion sorption on some reactions (Drever and

Clow, 1995).

Old organic carbon, stored in sedimentary rocks as “kero-

gen” also undergoes oxidative weathering reactions. These

reactions are less well understood than those for mineral weath-

ering, as they often involved the production of oxidized func-

tional groups that may later serve as sites for bacterial attack.

A schematic reaction may be written:

O þ O

! CO

þ H

O ð5 Þ

Reaction (5) illustrates that the weathering of organic matter

consumes oxygen and produces carbon dioxide, although as

written it provides little insight into the actual reaction path-

ways. Unlike carbonate and silicate weathering, organic carbon

weathering is oxidative rather than fundamentally an acid-base

reaction, although the production of organic acids (R-COOH !

R-COO



) by partial oxidation may accompany kerogen

degradation (Petch et al., 2000). Organic rich sediments such

as shales and coals often contain pyrite, and the accompanying

oxidation of FeS

results in the formation of sulfuric acid, which

can lead to very low pHs. Sulfuric acid is a far stronger acid

than carbonic acid, and may account for a significant fraction

of the natural acidity even when pyrite is a minor component of

the weathered rock.

In a classic work, Jenny ( 1941) laid out “state factors ” that

affect soil formation, and these generally apply to weathering

in natural systems. They include time, climate, topography

and vegetation. Since most weathering reactions are strongly

rate-limited, weathering can be slow even on a geologic time

scale. Deeply weathered landscapes often have been stable

for long times (10

–10

years). Second, both temperature

and rainfall are strong controls on natural weathering rates.

Alumino-silicate reaction rates generally follow the Arrhenius

relationship and can be strongly temperature dependent.

Reaction rates can also depend on the degree of undersatura-

tion of the solution in contact with the mineral and thus

increased rainfall and more rapid flushing can enhance reaction

rates and weathering reaction progress. “We t ” systems may

leach out the available carbonate, removing the principal

buffer, allowing pH to decrease and silicate weathering reac-

tion rates to increase (Chadwick and Chorover, 2001). The

decomposition of soil organic matter and production of

organic and carbonic acids also depend positively on both

temperature and (to the limit of saturation) moisture. Thus

weathering rates can be expected to depend significantly on

climate, and the global distribution of highly weathered

soils supports this hypothesis. The removal of soil and rego-

lith by erosion can expose fresh bedrock and enhance the

overall weathering flux, although rapid removal will prevent

weathering reactions from going toward completion. Two

end member cases can be defined: “transport” limited systems,

in which erosion removes regolith before it has a chance

to undergo much chemical attack, and “weathering” limited

systems, in which the development of a thick mantle of weath-

ering products inhibits the introduction of water and acids

to fresh mineral surface, thus slowing the overall weathering

rate (Stallard and Edmond, 1983). A currently important area

of study is the relationship between physical erosion, climate

and weathering fluxes in difference environments, as it has

been difficult to unambiguously point to a simple set of

weathering rate controls across a variety of environments

and spatial scales.

Interpretation of paleoclimate from

paleoweathering studies

Paleosols are preserved weathering profiles found through-

out most of the geological record. Their interpretation is

complicated by a number of factors, including the state of

preservation and the relationship of the regolith to the bed-

rock. For example, many soils have a significant (and in some

cases like loessic soils, dominant) component of allochthonous

transported material in them, transported by wind or other

agents. Calculating the chemical weathering depletion of a soil

by comparing its bulk chemistry to that of the underlying bed-

rock may not be meaningful in such cases, unless considerable

care is taken to account for atmospheric deposition fluxes.

Furthermore, it is often impossible to constrain the time scale

of paleosol formation. However, some useful inferences can

be drawn about paleoclimate from paleosol studies. These

include qualitative data on precipitation and paleovegetation

from isotopic studies of paleosol carbonates (Quade et al.,

1989), and constraints on the composition of the atmosphere

from models that use the differences in diffusivity of

and

to calculate pCO

in the overlying atmosphere

(Cerling, 1992). The “paleosol pCO

barometer” has been

widely applied to Phanerozoic sites, and gives a picture of

the history of pCO

that is in broad agreement with modeling

studies (Mora et al., 1996).

982 WEATHERING AND CLIMATE

Influen ce of weath ering proce sses on

atmosp heric pCO

The three basic weathering processes described above all

involve CO

as a product or reactant. Recognition of the role

of weathering process in controlling CO

, the role of climate

in controlling weathering rates, and the role of CO

as a green-

house gas led to the qualitative notion that a negative feedback

existed in the geologic carbon cycle as early as the mid-

nineteenth century (Ebelmen, 1845; Arrhenius, 1896; Cham-

berlin, 1899). The basic hypothesis is that increasing pCO

whether from volcanic emissions or some other source, would

cause a warming of climate. The warm climate would enhance

weathering reactions by a direct temperature effect, by increas-

ing precipitation on land, or by increasing the activity of plants

that would enhance weathering rates through increased soil

pCO

. The faster weathering reactions would result in en-

hanced consumption of CO

, eventually reducing the green-

house effect and causing climate to cool. Conversely, an

“externally ” forced fall in pCO

would result in cooling and

reduced weathering, gradually allowing natural sources of

to build up in the atmosphere and increase greenhouse

warming. Beginning in the 1980s, these concepts were quanti-

fied with computer models, and the possibility of a viable cli-

mate stabilization with weathering-greenhouse feedback was

demonstrated (Walker et al., 1981; Berner et al., 1983).

The three kinds of weathering processes have different

impacts on the long-term carbon cycle. Carbonate weathering

has no long-term effect – in the oceans, Reaction (1) is exactly

reversible, and the formation of carbonate (biogenic or abio-

genic) sediments in the ocean consumes the Ca

ion and the

bicarbonate and returns the CO

to the ocean atmosphere sys-

tem. In contrast, alumino-silicate weathering can be a net sink

for CO

where Ca and / or Mg silicates are involved. A simple

schematic pair of reactions illustrates the point:

CaSiO

þ 2CO

þ 3H

O ! Ca

2þ

þ H

SiO

þ 2HCO



ð6 Þ

2 þ

þ H

SiO

þ 2HCO



! CaCO

þ SiO

þ 3H

O þ CO

ð7 Þ

Here a c alcium (or equivalently magnesium) silicate under-

goesattackbyCO

and water to for m Ca

ion, bicar bonate

and silicic acid, consuming t w o m oles of CO

per m ole o f

miner al dissolved. In the ocean, the bicar bonate and c alcium

can pr ecipitate to for m sedimentary carbonate, releasing one mole

of CO

back to the ocean- atmosphere system but with a net

removal of one mole of CO

as car bonate. The net reaction is:

CaSiO

þ CO

! CaCO

þ SiO

ð8 Þ

Because weathering of silicates is dependent on climate

(temperature and rainfall), the weathering of Ca-Mg silicates

can act as a feedback mechanism to stabilize atmospheric

pCO

and thus climate. The weathering of K- and Na-silicates

does not interact with the carbon cycle so simply, as K and Na

do not form significant carbonate deposits. A fraction of K

and

released by mineral weathering undergoes ion exchange

with Ca

, and thus can contribute to CO

consumption via this

pathway, but the fraction is uncertain and probably not large,

so weathering of alkali silicate minerals is an inefficient

sink for CO

(France-Lanord and Derry, 1997). The direct

role of Ca-Mg (alkaline earth) silicates in the carbon cycle

and their generally faster weathering rates than K and Na

silicates indicate that weathering of mafic-intermediate rocks

is the critical process by which silicate weathering can act as

a climate feedback (Dessert et al., 2003).

Geochemical models that seek to investigate the global

inter-dependence of climate and weathering have been used

to estimate past atmospheric pCO

, climate and weathering

fluxes. Critical to these models is the formulation of the tem-

perature dependence of weathering rate, typically according to

the Arrhenius relationship:

k ¼ A exp E

=RTðÞð9 Þ

where k is the weathering reaction rate, A is a reaction-specific

constant, R is the Universal Gas Constant, T is temperature

(Kelvins), and E

is the activation energy. This latter term

determines the overall temperature sensitivity of the reaction.

Two difficulties are immediately apparent with this approach.

First, the Arrhenius “law” is most properly applied to a specific

elementary reaction, but in practice the modeled reactions are

composed of several (even many) elementary reactions, and it

is not certain that the Arrhenius relationship holds under these

circumstances, for which it was not originally determined. Sec-

ond, determining an average temperature sensitivity (E

) for

silicate weathering reactions at a large scale is difficult. Most

experimental data point toward values of E

near 60 kJ per

mole for the relevant silicate weathering reactions, which

would imply about a factor of 5–6 increase in reaction rate

between a cool climate (mean annual temperature 5



C) and

a warm climate (25



C). Potentially more important than tem-

perature changes are changes in precipitation that would result

from global climate change. The primary driver for a weather-

ing feedback to climate change may well be via global precipi-

tation (White and Blum, 1995). Under warmer conditions,

more water vapor should cycle through the atmosphere, result-

ing in greater rainfall and runoff. Complicating these relatively

simple physico-chemical factors are the composition of

exposed rock types, and the location of land masses in different

climatic zones (Bluth and Kump, 1994). To illustrate the poten-

tial effects of these two factors, consider a paleogeography in

which highly weatherable mafic-intermediate volcanic rocks

were confined to high, cold and dry latitudes. The combination

of low temperatures and low runoff should produce relatively

low rates of CO

consumption. Alternatively, if these litholo-

gies were primarily exposed in the warm and wet tropics,

Ca-Mg silicate weathering should be enhanced, and CO

consumption maximized.

The weathering of organic matter has the opposite effect on

the atmospheric CO

budget. Oxidation of kerogen releases

. This effect is, in the long term, nearly balanced by the

burial of new organic matter in sediments, but sedimentary

budget and carbon isotope studies indicate that there have been

periods when the oxidation of old sedimentary carbon and bur-

ial of newly reduced carbon are not in balance. For example,

changes in the net storage of organic carbon in sediments

appear to have played a role in creating cool conditions in the

Carboniferous (Canfield and Berner, 1989), but the impact of

climate variability in creating suitable conditions for large bur-

ial fluxes of carbon as coal poses the problem of which came

first. Glacially forced eustatic sea level variations appear to

have been important in burying large quantities of terrestrial

plant material in marginal swamps, while increasing this burial

rate had the effect of further decreasing the quantity of carbon

in the ocean-atmosphere system, which may have led to further

WEATHERING AND CLIMATE 983

cooling. As in other intervals of Earth history, the feedback nat-

ure of the carbon-climate system can make it difficult to deter-

mine ultimate cause-effect relationships.

Mountain building episodes and weathering

The original quantitative models linking weathering, CO

and

climate were essentially driven by the degassing fluxes from

the Earth’s interior, generally scaled to estimates of mid-ocean

ridge crustal production rates. They did not take into account

changes in “weatherability” of the mean continental surface,

either by changes in the average composition of exposed rock

types or by the enhancing effects of uplift and erosion on che-

mical weathering rates. The extent to which uplift affects regio-

nal or global scale weathering rates has been the focus of

considerable debate in the geological literature in recent years

(Raymo and Ruddiman, 1992). Fluxes of sediment are much

higher from orogenic areas, particularly those located in wet

climates (Milliman and Syvitski, 1992). Chemical weathering

fluxes also are higher in streams that drain uplifted regions.

Much of this enhanced flux is apparently due to carbonate dis-

solution, but silicate weathering fluxes have also been shown to

be higher in regions of high relief (Sarin et al., 1989; Jacobson

and Blum, 2003). More difficult is the question of how CO

consumption by weathering depends on uplift history, as both

rock type and climate appear to play important roles as well.

For example, the modern Himalaya do not appear to be a site

of strongly enhanced CO

consumption by weathering, in part

because rapid erosion limits the extent of completion of chemi-

cal alteration reactions, and in part because the alkali silicates

common there are not efficient sinks for CO

(Galy and

France-Lanord, 1999). The creation of large-scale topographic

features may alter climate locally or regionally in ways that

impact weathering. Perturbation of atmospheric circulation pat-

terns by mountain uplift can create regional climates that are

quite different from what would be expected in the absence

of high topography. The current impact of the Tibetan plateau

on monsoon circulation is the outstanding modern example

(Prell et al., 1992). Many mountain ranges across the world

strongly focus and intensify precipitation locally by orographic

effects even if they are too small to drive major circulation pat-

terns like the Indian monsoon. Increased precipitation is often

associated with more rapid and/or complete weathering, but,

conversely, rain shadow effects can result in significant areas

of aridity and low weathering rates. The combination of regio-

nal circulation patterns, rain shadows, and low ocean surface

temperatures results in low precipitation on the high Andean

plateaus and on the west flank of the Andes, and very low

weathering fluxes to the Pacific Ocean. On the east side of

the Andes, high rainfall produces rapid erosion, and most of

the solute fluxes in the large Amazon and Orinoco rivers are

generated there (Stallard and Edmond, 1983). Thus, the effects

of Andean uplift on weathering rates are highly asymmetric.

Mountain building can impact the organic part of the carbon

cycle as well, through enhanced erosion and oxidation of old

sedimentary kerogen, or through enhanced burial of “new”

organic matter. Organic carbon sequestration rates are highest

in areas of high sedimentation rates, and a large fraction of

organic carbon burial takes place along continental shelves

where productivity is high and sedimentation rates fast enough

to ensure rapid burial and isolation. Currently, high sedimenta-

tion rates and a supply of refractory terrestrial organic matter

make the Bengal Fan, the major sedimentary repository of the

Himalaya, the largest site of organic carbon burial globally.

Because carbon-poor rocks are being eroded from the

Himalayas, while carbon-rich sediments are deposited in

the huge submarine fans, Himalayan erosion is a net sink for

the organic carbon sub-cycle (France-Lanord and Derry, 1997).

TheimpactoftheHimalayan(oranyother)carbonsinkoncli-

mate during the last 20 million years is not clear, as available paleo-

proxy data do not indicate any large shift in atmospheric pCO

since the early Miocene, a period during which large scale global

cooling is well documented (Pagani and Freeman, 1999;Pearson

and Palmer , 2000). The overall relationship between the carbon

cycle and climate is not yet well understood, although there are a

number of indications that the carbon cycle does play an important

role in long-term climate change. The impact of weathering pro-

cesses on the long-term carbon cycle, and the potential feedbacks

involved remain one of the more interesting and complex features

of the Earth system. The weathering-climate feedback is an attrac-

tive hypothesis for the stabilization of the Earth’s climate over very

long time scales (10

–10

years). Changing solar output over those

time scales could have forced drastic climatic changes in the

absence of a stabilizing feedback, but for all but a few brief epi-

sodes of Earth history, surface temperatures appear to have been

in the range of stability of liquid water. Gradually increasing solar

luminosity may have been balanced by decreasing pCO

,asweath-

ering reactions gradually increasedthestorageofcarboninsedi-

ments at the expense of the ocean-atmosphere system. Such a

transfer would be important from an atmospheric point of view ,

but quite small in terms of the sedimentary rock reservoir (a relative

change on the order of 10

4

Long term changes in global weathering rates:

Sr isotope evi dence

Available data from various sources, including paleo-sea level

reconstructions, paleobotany, and oxygen isotopes in marine

fossils, indicate that the Earth’s climate has been considerably

warmer at times in the geologic past. These “greenhouse” inter-

vals, such as the Eocene warm period, or the mid-Cretaceous,

should be associated with higher rates of weathering if the

hypothesis of a long-term feedback between climate and weath-

ering is valid. Conversely, “icehouse” intervals, with overall

cool climates (such as the Pleistocene), should have lower

globally averaged weathering rates. Model studies that incorpo-

rate a climate-weathering feedback do support this relationship,

but that is clearly in part a function of the construction of the

model around such a relationship (Berner, 1991). A reasonable

question is whether there is good independent evidence of

changes in global weathering rates that are correlated with cli-

mate on these long time scales. While paleosols do preserve

information on weathering intensity from a particular location,

they cannot be used to reconstruct global weathering rates reli-

ably. Changes in ocean chemistry should result from long-term

changes in weathering fluxes. Because the major element

chemistry of seawater is, at best, imperfectly preserved in old

sediments, attention has been given to isotopic tracers such as

Sr/

Sr. Because strontium is a fairly abundant constituent

of seawater and has a long residence time, it has a uniform iso-

topic composition in the world oceans at any one time. In a

closed system (such as a given mineral phase), the

Sr/

ratio will increase over geologic time because of the decay

Rb to

Sr, with a half life of 42 Ga. Minerals with a high

parent-daughter ratio (high Rb/Sr) will evolve more rapidly

than those with low Rb/Sr.

Analysis of many well-preserved marine carbonate samples

has demonstrated that the

Sr/

Sr ratio of seawater has varied

984 WEATHERING AND CLIMATE

significantly over time (Burke et al., 1982; Veizer et al., 1999).

Marine carbonates have very low levels of

Rb, and so should

not evolve significant

Sr after their formation in the marine

environment. The main causes of variation in the

Sr /

Sr ratio

of seawater (as recorded in marine carbonate sediments of var-

ious ages) must be changes in the fluxes of Sr to the oceans

from sources with different characteristic

Sr /

Sr ratios. The

prime candidates are: (a) changes in the isotopic composition

of Sr in the global river flux, and (b) changes in the relative

contribution of Sr from submarine hydrothermal alteration

and the river flux. The current value of

Sr /

Sr in the oceans

is 0.70917, while the global average river value is near 0.712.

Mid-ocean ridge hydrothermal vents are near 0.7034, a value

similar to the upper oceanic crust that they sample. In the sim-

plest case, these input values imply that rivers supply two-third

of the total Sr flux to the oceans, while hydrothermal systems

supply the other third (Palmer and Edmond, 1992). Lower

Sr /

Sr values of the oceans in the past would imply an

increased hydrothermal flux, a reduced river (weathering) flux,

or some combination. However, there are a number of impor-

tant caveats to this simple approach. First, the estimate of the

hydrothermal contribution to ocean chemistry based on Sr iso-

topes is higher than independent estimates from other chemical

or isotopic tracers, and suggests that there are significant

sources of non-radiogenic Sr to the oceans from crustal altera-

tion of the oceanic crust at low temperature, away from ridge

axes (Elderfield and Schultz, 1996). Second, the oceans are

not now at steady state with respect to their Sr budget, and this

disequilibrium results in an over-estimate of the hydrothermal

contribution (Galy et al., 1999). Third, changes in the isotopic

composition of the river flux with time can be large enough

to confound attempts to extract the balance between hydrother-

mal and weathering fluxes through time. For example, Himala-

yan rivers have unusually high Sr concentrations as well as

very high

Sr /

Sr, and the Neogene development of the

Himalaya is at least in part responsible for the high average

value of the global river flux to the present oceans (Edmond,

1992; Krishnaswami et al., 1992; Richter et al., 1992). Further,

the

Sr /

Sr of Himalayan rivers has changed significantly dur-

ing the last 10 Ma (Derry and France-Lanord, 1996; Quade

et al., 1997). The impact of Himalayan tectonic processes on

the

Sr /

Sr of the global river flux illustrates how changes

in the types of rocks weathered at the Earth’ s surface can

change the isotopic composition of one of the major contribu-

tors to the ocean Sr budget, and over time change the

Sr /

of the oceans, without necessarily implying large changes in

globally averaged weathering rates or CO

consumption.

Finally, the fundamental interest in strontium isotope ratios as

a measure of weathering fluxes is that strontium is chemically

similar to calcium. Ideally, changes in the Sr mass balance of

the oceans could be used to estimate changes in the global

weathering calcium flux to the oceans. Carbonate sediments

(limestones and dolomites) should record the

Sr /

Sr of the

ocean at the time they were deposited. Conversely, some silicate

rocks, particularly those with granitic bulk compositions, can

have high Rb / Sr ratios, and over time generate high

Sr /

ratios. The weathering of continental rocks should therefore

yield high Sr isotope ratios in river waters, and the

Sr /

Sr ratio

of seawater should reflect increases in the weathering flux from

old continental material on a million year time scale. Upon

further consideration, an interesting conundrum appears. Granitic

silicates have high Rb / Sr ratios and thus evolve to high

Sr /

values, but they also have relatively low Ca and Mg abundances.

Thus, their potential to act as weathering sinks for CO

via reac-

tions like Reaction (8) is modest. Mafic silicates, including

basalts and intermediate compositions generally have low

Rb/Sr and low

Sr/

Sr ratios, yet these compositions have

abundant Ca and Mg, and can act as strong sinks for CO

(Dessert et al., 2003). The widely held view that increasing

Sr/

Sr in the oceans should correlate with increased CO

con-

sumption via silicate weathering is unlikely to be generally

applicable (Raymo and Ruddiman, 1992).

A comparison of the available data on paleoclimate recon-

structions with the

Sr/

Sr ratio of past oceans suggests that

at least some warm “greenhouse” periods of Earth history are

associated with low

Sr/

Sr. This could reflect increased hydro-

thermal fluxes, but recent independent work strongly suggests

that ocean crustal production rates have varied little since the

Jurassic (Rowley, 2002). Instead, low

Sr/

Sr during much of

the Cretaceous and Paleogene may in part reflect enhanced

alteration of mafic and intermediate silicate rocks, and thus rela-

tively high rates of CO

consumption. The increase in marine

Sr/

Sr since the Eocene is almost certainly in part due to

changes in the types of rocks undergoing weathering, and may

not reflect increased CO

consumption rates. It may reflect

increased weathering rates of granitic materials, but as noted

above, the associated CO

consumption can be modest.

Unfortunately, the question posed at the beginning of this sec-

tion cannot be easily answered at present. While model results

predict enhanced global weathering rates under warm “green-

house” climates, there is no well-understood and fully inde-

pendent data set that verifies this hypothesis. An increasingly

well-documented set of laboratory and field experiments does

support the general outlines of the hypothesis first posed by

Ebelmen over 150 years ago, but we remain far from a com-

plete understanding of weathering-climate relationships over

geologic time.

Louis A. Derry

Bibliography

Arrhenius, S., 1896. On the influence of carbonic acid in the air on the tem-

perature on the ground. Philos. Mag. J. Sci., 5, 237–276.

Berner, R.A., 1991. Atmospheric CO

levels over Phanerozoic time.

Science, 249, 1382–1386.

Berner, R.A., Lasaga, A.C., and Garrels, R.M., 1983. The carbonate-silicate

geochemical cycle and its effect on atmospheric carbon dioxide over the

past 100 million years. Am. J. Sci. , 283, 641–683.

Bluth, G.J.S., and Kump, L.R., 1994. Lithologic and climatologic controls

of river chemistry. Geochim. Cosmochim. Acta, 58, 2341–2359.

Burke, W.H., Denison, R.E., Hetherington , E.A., Keopnic k, R.B., Nelson, H.F.,

and Otto, J.B., 1982. Variation of seawater

Sr/

Sr throughout Phaner-

ozoic time. Geology, 10, 516–519.

Canfield, D.E., and Berner, R.A., 1989. A new model for atmopsheric oxy-

gen over Phanerozoic time. Am. J. Sci., 289, 333–361.

Cerling, E., 1992. Use of carbon isotopes in paleosols as an indicator of the

P(CO

) of the paleoatmosphere. Global Biogeochem. Cycles, 6,

307–314.

Chadwick, O.A., and Chorover, J., 2001. The chemistry of pedogenic

thresholds. Geoderma, 100, 321–353.

Chamberlin, T.C., 1899. An attempt to frame a working hypothesis of the

cause of glacial periods on an atmospheric basis. J. Geol., 7, 545–584.

Derry, L.A., and France-Lanord, C., 1996. Neogene Himalayan weathering

history and river

Sr/

Sr: Impact on the marine Sr record. Earth Pla-

net. Sci. Lett., 142,59–74.

Dessert, C., Dupre, B., Gaillardet, J., Francois, L.M., and Allegre, C.J.,

2003. Basalt weathering laws and the impact of basalt weathering on

the global carbon cycle. Chem. Geol., 202, 257–273.

Drever, J.I., and Clow, D.W., 1995. Weathering rates in catchments. In

White, A.F., and Brantley, S.L. (eds.), Chemical weathering rates of

WEATHERING AND CLIMATE 985

silicate minerals (Reviews in mineralogy, Vol. 31). Washington, D.C.:

Mineralogical Society of America, pp. 463–483.

Ebelmen, J.J., 1845. Sur les produits de la décomposition des espèces

minérales de la famille des silicates. Annales des Mines, 7,3–66.

Edmond, J.M., 1992. Himalayan tectonics, weathering processes, and the

strontium isotope record in marine limestones. Science, 258,

1594–1597.

Elderfield, H., and Schultz, A., 1996. Mid-ocean ridge hydrothermal fluxes

and the chemical composition of the ocean. Annu. Rev. Earth Planet.

Sci., 24, 191–224.

France-Lanord, C., and Derry, L.A., 1997. Organic carbon burial forcing of

the carbon cycle from Himalayan erosion. Nature, 390,65–67.

Galy, A., and France-Lanord, C., 1999. Weathering processes in the

Ganges-Brahmaputra basin and the riverine alkalinity budget. Chem.

Geol., 159,31–60.

Galy, A., France-Lanord, C., and Derry, L.A., 1999. The strontium isotopic

budget of Himalayan rivers in Nepal and Bangladesh. Geochim. Cos-

mochim. Acta, 63, 1905–1925.

Jacobson, A.D., and Blum, J.D., 2003. Relationship between mechanical

erosion and atmospheric CO

consumption in the New Zealand South-

ern Alps. Geology, 31, 865–868.

Jenny, H., 1941. Factors of Soil Formation. New York: McGraw-Hill,

281p.

Krishnaswami, S., Trivedi, J.R., Sarin, M.M., Ramesh, R., and Sharma, K.K.,

1992. Strontium isotopes and rubidium in the Ganga-Brahmaputra river

system: Weathering in the Himalaya, fluxes to the Bay of Bengal and

contributions to the evolution of oceanic

Sr/

Sr. Earth Planet. Sci.

Lett., 109, 243–253.

Lasaga, A.C., Soler, J.M., Ganor, J., Burch, T.E., and Nagy, K.N., 1994.

Chemical weathering rate laws and global geochemical cycles. Geo-

chim. Cosmochim. Acta, 58, 2361–2386.

Milliman, J.D., and Syvitski, P.M., 1992. Geomorphic/ tectonic control of

sediment discharge to the ocean: the importance of small mountainous

rivers. J. Geol., 100, 525–544.

Mora, C.I., Driese, S.G., and Colarusso, L.A., 1996. Middle to Late Paleo-

zoic atmospheric CO

levels from soil carbonate and organic matter.

Science, 271, 1105–1107.

Pagani, M., and Freeman, K.H., 1999. Late Miocene atmospheric CO

con-

centrations and the expansion of C4 grasses. Science, 285, 876–879.

Palmer, M.R., and Edmond, J.M., 1992. Controls over the strontium

isotope composition of river water. Geochim. Cosmochim. Acta, 56,

2099–2111.

Pearson, P.N., and Palmer, M.R., 2000. Atmospheric carbon dioxide con-

centrations over the past 60 million years. Nature, 406, 695–699.

Petch, S.T., Berner, R.A., and Eglinton, T.I., 2000. A field study of the che-

mical weathering of ancient sedimentary organic matter. Org.

Geochem., 31, 475–487.

Prell, W.L., Murray, D.W., and Clemens, S.C., 1992. Evolution and varia-

bility of the Indian Ocean summer monsoon: Evidence from the Wes-

tern Arabian sea drilling program. In Duncan, R.A., Rea, D.K., Kidd,

R.B., von Rad, U., and Weissel, J.K. (eds.), Synthesis of Results from

Scientific Drilling in the Indian Ocean, Washington D.C.: American

Geophysical Union, pp. 447–469.

Quade, J., Cerling, T.E., and Bowman, J.R., 1989. Development of Asian

monsoon revealed by marked ecological shift during the latest Miocene

in northern Pakistan. Nature, 342, 163–166.

Quade, J., Roe, L., DeCelles, P.G., and Ojha, T.P., 1997. The late Neogene

Sr/

Sr record of lowland Himalayan rivers. Science, 276,

1828–1831.

Raymo, M.E., and Ruddiman, W.F., 1992. Tectonic forcing of late Ceno-

zoic climate. Nature, 359,117–122.

Richter, F.M., Rowley, D.B., and DePaolo, D.J., 1992. Sr isotope evolution

of seawater: The role of tectonics. Earth Planet. Sci. Lett., 109,11–23.

Rowley, D.B., 2002. Rate of plate creation and destruciton: 180 Ma to pre-

sent. Geol. Soc. Am. Bull., 114, 927–933.

Sarin, M.M., Krishnaswami, S., Dilli, K., Omayajulu, B.l.K., and

Moore, W.S., 1989. Major ion chemistry of the Ganga-Brahmaputra

river system: Weathering processes and fluxes to the Bay of Bengal.

Geochim. Cosmochim. Acta, 53, 997–1009.

Stallard, R.F., and Edmond, J.M., 1983. Geochemistry of the Amazon: 2.

The influence of geology and weathering environment on the dissolved

load. J. Geophys. Res., 88, 9671–9688.

Veizer, J., Ala, D., Azmy, K., Bruckschen, P., Buhl, D., Bruhn, F.,

Carden, G.A.F., Diener, A., Ebneth, S., Godderis, Y., Jasper, T., Korte,

G., Pawellek, F., Podlaha, O.G. and Strauss, H., 1999. Sr-87 /Sr-86,

delta C-13 and delta O-18 evolution of Phanerozoic seawater. Chem.

Geol., 161,59–88.

Walker, J.C.G., Hays, P.B., and Kasting, J.F., 1981. A negative feed back

mechanism for the long-term stabilization of Earth’s surface tempera-

ture. J. Geophys. Res., 86, 9776–9782.

White, A.F., and Blum, A.E., 1995. Effects of climate on chemical weath-

ering in watersheds. Geochim.t Cosmochim. Acta, 59, 1729–1747.

Cross-references

Carbon Cycle

Marine Carbon Geochemistry

Mineral Indicators of Past Climates

Mountain Uplift and Climate Change

Paleosols, Pre-Quaternary

Strontium Isotopes

WISCONSINAN (WEICHSELIAN, WU

RM)

GLACIATION

Introduction

The term “Wisconsinan Glaciation” refers to the last major gla-

cial episode that affected North America, more properly referred

to as the Wisconsinan Stage (Black et al., 1973; Fulton, 1989;

Clark and Lea, 1992)(Figure W1). The term should be restricted

to the deposits and events of the Laurentide Ice Sheet (LIS; see

Laurentide Ice Sheet), which extended from the Canadian Arctic

coast to the northern states in the USA, and across the Dakotas

eastward to New England (Figure W2). In this sense, the term is

not properly used to describe events of the Cordilleran Ice Sheet

(see Cordilleran Ice Sheet), which extended from the Alberta

foothills to the coast of British Columbia. Weichselian and

Würm glaciations refer to the last glacial period in Fennoscandia

and the Alps, respectively, although these terms, in addition

to the “Wisconsinan,” are frequently applied to regions well-

removed from the type areas.

The problem in nomenclature stems from efforts to integrate

the mapping of glacial and other sediments associated with cli-

matic variations into the standard American and International

Stratigraphic Code (Hedberg, 1976). Terms such as Wisconsin

Glaciation or Sangamon Interglaciation are “Geologic-climate

units” whereas Wisconsinan Stage is a time-stratigraphic unit

that requires, by definition, upper and lower isochronous bound-

aries throughout the geographic area to which it is applied (Flint,

1971). By contrast, geologic-climate units are not constrained to

be synchronous, which in reality agrees with the efforts to date

the maximum limit of the LIS during its last major glacial

advance (Dyke and Prest, 1987; Dyke et al., 2002). Such efforts

indicate that the Last Glacial Maximum (LGM; Figure W1)isa

global concept and the timing of the maximum of the late Wis-

consinan Glaciation varies around its margins from around 22 to

possibly as young as 13 cal ka

BP (BP = Before Present with “pre-

sent” equal to

AD 1950). Frequently used terms such as the “late

Wisconsin” generally refer to Marine Isotope Stage (MIS) 2

(Figure W1), but the definitions of middle and early Wisconsin

are more varied in their usage and definitions (Fulton, 1989).

The discussion gets convoluted in terms of definitions and

semantics, leading to a lack of simplicity and uniformity in

usage. For example, as a time-stratigraphic term, the Wisconsi-

nan Stage ends at the Pleistocene/Holocene boundary, which

986 WISCONSINAN (WEICHSELIAN, WU

RM) GLACIATION