The humid tropics: weathering and laterites
523
3.7. Weatheringrates
Chemical weathering processes exist in other climatic zones, but in the humid tropics
alteration velocity is much higher (Muxart and Birot, 1977). In this sense, and in addition
to the study of these processes, the quantification of rock weathering is also necessary.
The methods applied for this purpose are varied. Quantitative data can be obtained from
the alteration of building and ornamental stones, whose age of installation is known.
Numerous authors have carried out these types of works under different climates (Ollier,
1984). Approximate data can also be obtained in the laboratory by accelerated aging.
Values recorded in the beginning of the weathering process are the highest. Therefore,
recorded weathering rates will be greater than those existing throughout the whole
alteration, and they have been estimated to be about 200 to 400 times higher than the ones
obtained in the field (Swoboda-Colberg and Drever, 1993). Nevertheless, one of the most
utilized methods is based on the quantification of the solid and solved discharge, measured
in experimental plots of some tens of metres long. The chemical composition of the
weathering residue has also been studied (Nahon, 1991). The experimentation period
is variable and normally fluctuates between 1 and 5 years. In humid tropical savanna zones
of Cameroon a surficial lowering rate has been obtained of 1 to 2 mm/ka (Birot, 1978), and
1.2 mm/ka in Ivory Coast (Roose and Lelong, 1976). In Burkina Faso, Senegal and Ivory
Coast values fluctuate between 1 and 8 mm/ka (compiled by Goudie, 1995). Another
method consists in the study of the bedload, suspended, and dissolved load discharges
from small basins. In the study of three experimental basins in New England in western
Australia, it was found that the dissolved load is ten times greater than the suspended load,
and the bedload is virtually absent (Zakaria, 1977; in Ollier (1984), p. 209). Obtained
denudation values are 14.4 mm/ka for basalts and 4.5 mm/ka for granites (Ollier, 1984).
Data from Rhodesia indicate a rate of surface lowering in granites of 15 mm/ka in rainy
zones and 6 mm/ka in drier zones (Owens and Watson, 1979). Erosion values obtained
on ultrabasic rocks of New Caledonia are about 28 mm/ka (Trescases, 1973, in Muxart
and Birot (1977)). A compilation made by Thomas (1994) on alteration velocities gave an
estimation of 2 to 50 mm/ka. Finally, weathering rates over long periods of time can be
calculated from absolute dating of geologic materials (pyroclasts, lavas, etc.).
Values obtained from experimental methods indicate that most regoliths formed during
the Quaternary because average weathering rate values are about 20 m per million years.
Birot (1978) gave values for the alteration of 1 m granite: 50,000 years in a rainforest, and
70,000 years in a savanna. All of them are similar to that proposed by Thomas (1994).
When considering these long time periods, it must be kept in mind that the climatic
changes can have occurred during the profile development. In dry periods the weathering
rate slows, whereas denudation increases.
3.8. Laterites
Several denotations exist in the literature for the term "laterite," such as "duricrust,"
"plintite," "ferric laterite," "ferricrete," "latosol," and others (Thomas, 1974), and it is
difficult to define. In the 19th century the physical properties, strength, and colour of
laterites were already known. Buchanan (1807) described a material soft enough to be cut
523
3.7. Weatheringrates
Chemical weathering processes exist in other climatic zones, but in the humid tropics
alteration velocity is much higher (Muxart and Birot, 1977). In this sense, and in addition
to the study of these processes, the quantification of rock weathering is also necessary.
The methods applied for this purpose are varied. Quantitative data can be obtained from
the alteration of building and ornamental stones, whose age of installation is known.
Numerous authors have carried out these types of works under different climates (Ollier,
1984). Approximate data can also be obtained in the laboratory by accelerated aging.
Values recorded in the beginning of the weathering process are the highest. Therefore,
recorded weathering rates will be greater than those existing throughout the whole
alteration, and they have been estimated to be about 200 to 400 times higher than the ones
obtained in the field (Swoboda-Colberg and Drever, 1993). Nevertheless, one of the most
utilized methods is based on the quantification of the solid and solved discharge, measured
in experimental plots of some tens of metres long. The chemical composition of the
weathering residue has also been studied (Nahon, 1991). The experimentation period
is variable and normally fluctuates between 1 and 5 years. In humid tropical savanna zones
of Cameroon a surficial lowering rate has been obtained of 1 to 2 mm/ka (Birot, 1978), and
1.2 mm/ka in Ivory Coast (Roose and Lelong, 1976). In Burkina Faso, Senegal and Ivory
Coast values fluctuate between 1 and 8 mm/ka (compiled by Goudie, 1995). Another
method consists in the study of the bedload, suspended, and dissolved load discharges
from small basins. In the study of three experimental basins in New England in western
Australia, it was found that the dissolved load is ten times greater than the suspended load,
and the bedload is virtually absent (Zakaria, 1977; in Ollier (1984), p. 209). Obtained
denudation values are 14.4 mm/ka for basalts and 4.5 mm/ka for granites (Ollier, 1984).
Data from Rhodesia indicate a rate of surface lowering in granites of 15 mm/ka in rainy
zones and 6 mm/ka in drier zones (Owens and Watson, 1979). Erosion values obtained
on ultrabasic rocks of New Caledonia are about 28 mm/ka (Trescases, 1973, in Muxart
and Birot (1977)). A compilation made by Thomas (1994) on alteration velocities gave an
estimation of 2 to 50 mm/ka. Finally, weathering rates over long periods of time can be
calculated from absolute dating of geologic materials (pyroclasts, lavas, etc.).
Values obtained from experimental methods indicate that most regoliths formed during
the Quaternary because average weathering rate values are about 20 m per million years.
Birot (1978) gave values for the alteration of 1 m granite: 50,000 years in a rainforest, and
70,000 years in a savanna. All of them are similar to that proposed by Thomas (1994).
When considering these long time periods, it must be kept in mind that the climatic
changes can have occurred during the profile development. In dry periods the weathering
rate slows, whereas denudation increases.
3.8. Laterites
Several denotations exist in the literature for the term "laterite," such as "duricrust,"
"plintite," "ferric laterite," "ferricrete," "latosol," and others (Thomas, 1974), and it is
difficult to define. In the 19th century the physical properties, strength, and colour of
laterites were already known. Buchanan (1807) described a material soft enough to be cut