Continental Drift Effect on Climate and Weathering

The actual site where weathering occurs depends on the local climate. I have already shown how local climate may vary with time due to changes in global climate as a result of changes in greenhouse gas concentration, solar radiation, and so on. However, local climate can also vary because of changes in the position and size of the continents as a result of continental drift. This is a unique feature of the long-term carbon cycle. A simple argument, whereby a land mass is moved from the pole to the equator, illustrates how change in location can lead to great changes in temperature, rainfall, and the rate of CO2 uptake by silicate weathering (Worsley and Kidder, 1991).

One approach to the problem of climate change due to changing pa-leogeography is to assume the same latitudinal climate zones in the past as exist at present. Then, by moving the continents across paleolatitudes, one can sum up the total evaporation and rainfall for each zonal land segment and add those for all segments to obtain total global river discharge. This has been done by Tardy et al. (1989). (A more up-to-date review of the effects of paleogography on chemical weathering is given by Tardy, 1997.)

Although the Tardy (1989) approach is a useful first-order attack on the problem of paleogeography, climate, and weathering, it misses effects due to changes in continental size and topography. Larger continents, especially those with coastal mountains, should experience more extensive monsoonal climate and have very dry interiors (rain shadows) with wet margins (orographic rainfall). This is exemplified especially by the supercontinent Pangea (e.g., Kutzbach and Gallimore, 1989). To approach all aspects of the effect of changing paleogeography on climate and weathering, it is necessary to combine paleogeographic reconstructions (e.g., Scotese and Golonka, 1995) with paleoclimate modeling (e.g., Otto-Bliesner, 1993; Barron et al., 1995; Hyde et al., 1999; Gibbs et al., 2002). For the entire Phanerozoic it would be useful to have quantitative paleoclimatic reconstructions covering most of this period, but so far this has not been done. Attention has been focused on glaciations, the Cretaceous, and the Permian and Triassic, when Pangaea was at its largest size.

Parrish et al. (1982) has attempted to estimate rainfall patterns at various times during the Phanerozoic, based on geological indicators such as coals and evaporites. However, these results cannot be used to estimate mean global runoff over time. Only a few attempts have neen made to calculate runoff (as the difference between precipitation and evporation) over appreciable portions of Phanerozoic time (Otto-Bliesner 1993, 1995; Fawcett and Barron, 1998; Gibbs et al., 1999). Only Otto-Bliesner has considered the entire Phanerozoic. She used GCM modeling and the paleogeographic reconstructions of Scotese and Golonka (1995) to calculate global mean land temperatures and mean runoff for 13 times spanning the Phanerozoic. Unfortunately, because of inadequate knowledge of paleotopography over such a long time, Otto-Bliesner was forced to assume flat, ice-free continents at sea level. Her results, as they affect the rate of chemical weathering of silicates, have been incorporated into GEOCARB modeling (Berner, 1994; Berner and Kothavala, 2001) in terms of the dimensionless parameter:

A plot of fD(t) versus time is shown in figure 2.7. To actually apply fD(t) to the global rate of weathering, two modifications of equation (2.25) are needed. First, the total riverine discharge of water from the continents is what is desired, and this is obtained from runnoff (which is expressed per unit land area) by fAD(t) = fD(t)fA(t) where fA(t) = land area(t)/land area(0) (2.26)

Second, for silicate weathering, to express the effect of dilution of dissolved HCO3- at elevated runoff (see equation 2.20), one should use fD(t)°.65. Thus, the proper term to be applied to the global weathering flux would be

Chemical Weathering of Silicates 35 1.6-,-


Figure 2.7. Plot of fD(t) versus time. The parameter fD(t) represents the effect of changes in paleogeography on global river runoff. Runoff is a major factor affecting silicate weathering. The variable fD(t) is defined as global river runoff at a given past time divided by that at present. High values of fD(t) represent times when a large proportion of land area was in humid belts and low fD(t) values when a large proportion was in dry belts. Data based on the GCM model of Otto-Bliesner (1995).

In considering all factors affecting temperature as they relate to the rate of weathering, one should add the change in land temperature, due to changes in paleogeography, to the effects of the atmospheric greenhouse effect and solar evolution. Thus, equation (2.24) is modified to

T(t) - T(0) = r ln RCO2 - Ws(t /570) + GEOG(t) (2.28)

where GEOG(t) = mean land temperature at a past time minus mean land temperature at present, due solely to changes in paleogeography (in other words, for constant solar radiation and constant atmospheric CO2 concentration). The land temperature data of Otto-Bliesner (1995) can be used for this purpose (Berner and Kothavala, 2001). As pointed out earlier, the value of GEOG(t) should be based only on land where appreciable chemical weathering can take place, which excludes deserts and areas covered by glaciers. So far this has not been done.

Equation (2.28) can be substituted in equation (2.22) to consider all factors that affect global mean temperature as it relates to weathering. This results in a more complete nondimensional expression for the effect of atmospheric CO2 level on the rate of silicate weathering and is represented by fBtCO2 (subscript t stands for temperature). By substituting equation (2.28) in (2.22), we obtain:

fBt(CO2) = (RCO2)zr exp[ZGEOG(t) - ZWs(t/570)] (2.29)

This is the complete expression used in GEOCARB modeling to express the effect of changes in temperature on silicate weathering.

Another aspect of paleogeography, as it has affected weathering in the past, is change in the area of land exposed to weathering, independent of climate. The original approach to carbon cycle modeling was simply that weathering rate is directly proportional to land area (Berner et al., 1983; Fischer, 1983) through the use of equation (2.26). However, this does not distinguish mountainous areas where fresh, easily weathered Ca and Mg silicate minerals are constantly exposed to weathering, from lowland areas where there is an extensive cover of relatively unreactive secondary minerals. Changes in land area over the Phanerozoic have been due largely to the transgression and regression of continental seas, and during sea-level low stands, the extra land area is underlain largely by clay weathering products (shales) and detrital primary minerals (sandstones), both of which are relatively resistant to further chemical weathering. An exception to this generalization is areas at the foot of high mountains, where rapidly eroded, unweathered debris is delivered to nearby lowlands. An outstanding modern example is the Himalayan foothills and the Indo-Gangetic plain, which receive large amounts of freshly eroded material from the Himalayas. Here, due to the presence of the unweathered primary minerals, along with warmth and high rainfall, extensive weathering takes place (e.g., West et al., 2002).

These observations make it difficult to decide whether to include change in total land area (equation 2.24) directly in carbon cycle modeling. Also, there is the problem that changes in large portions of land, present as deserts and areas covered by glaciers, should not be included in fA(t). The approach I have used so far is to exclude total land area from the expression for the rate of CO2 uptake by the weathering of silicates (but not carbonates; see chapter 3) and to focus instead on the importance of continental relief (Berner, 1994; Berner and Kothavala, 2001).

Organic Gardening

Organic Gardening

Gardening is also a great way to provide healthy food for you and your loved ones. When you buy produce from the store, it just isnt the same as presenting a salad to your family that came exclusively from your garden worked by your own two hands.

Get My Free Ebook


  • Alice
    What are the factors affecting weathering?
    8 years ago
  • Doreen Wechsler
    What are the consequences of continental drift on weathering and precipitation?
    2 months ago

Post a comment