9.3 Mechanical Denudation
interglacials. "Hard rock" substrate~ show peaks ~f
erosion soon after the onset of glacIal steppe condltions when the forests die and their soils can be easily removed. In the subsequent glacial inten:al ne~
soil formation and thus erosion are limited unhl agam
trees can grow and promote rock weathering and soil
formation.
The Global Mechanical Denudation Rate
Globally, the contemporary total river flux of suspended sediment to the oceans is 15 - 20 x 10 9 tla or
100 to 130 t/km 2 /a (Milliman and Meade 1983;
Tamrazyan 1989; Ludwig et al. 1996). The highest
values of this range come from a very large data base
summarized by Milliman and Syvitski (1992). About
40% of this sediment flux originates from the wet
tropics, 21 % from wet temperate climates, 5% from
boreal climates, and 34% from dry climates (Ludwig
et al. 1996). If 50% of the present-day sediment flux
to the oceans reflects "natural" erosion (e.g. Hay
1998) and the contemporary total soilloss (including
human-induced erosion) is 50-75 x 10 9 tla, it appears
that only c. 15-20% of the eroded soil is delivered to
the oceans (Walling and Webb 1996).
There are significant differences in the specific
sediment yields of the present-day continents (e.g.
Harrison 1994). Asia with its widely extended high
mountain ranges yields 310 tlkm 2 /a, whereas Africa,
Australia, and Europe only deliver 17 to 27 tlkm 2 /a.
The va1ues for North and South America are 88 and
57 tlkm 2 /a, respectively. The natural (pre-European
settlement) sediment yield of Australia is estimated
to have been 2 to 5 times 10wer than the modem one
(Wasson et al. 1996).
A global average specific transport rate of 100
tlkm 2 /a, corresponding to a mechanical denudation
rate of about 40 mm/ka, would lead to an average
sedimentation rate (pore-free) in the world oceans of
about 17 mmIka (also see Sect. 9.4).
9.3.3 General Laws for Mechanical Denudation
Using the empirically gained data discussed above,
several authors have tried to formulate a general law
to express mechanical denudation in relation to relief
and climate. The rock types exposed in the drainage
areas are usually not considered in these attempts.
Taking into account that relief is the main fac~or,
Pinet and Souriau (1988) assurne that the mechalllcal
denudation rate, DR.ne' is largely correlated with the
mean drainage basin elevation (H m in m) above the
base level of erosion (e.g. sea level). They therefore
neglect environmental factors and propose the following empirical formula:
DR.ne = 419 x 10. 3 H m - 0.245
(mm/ka or mlMa)
(1 ka = 1000 years).
441
(9.1)
This simple formula often gives the correct order of magnitude, but sometimes renders results which significantly
differ from those depicted in Figs. 9.13 and 9.14 (see also
Leeder 1991). For example, an elevated plateau which is
not dissected by deep valleys may undergo much le~s denudation than a coastal region with steep slopes, as In South
Africa (Summerfield 1991).
An expression similar to that given above is used by
Pazzaglia and Brandon (1996):
DR.ne = ~ x Hm (mlMa).
(9.2)
Ku is a proportionality constant of denudation (in
short: denudation constant; global average 0.07-0.1
Ma'l). It varies significantly from region to r~gion
(often within a range of two orders of magllltud~)
and therefore has to be calibrated by local mechalllcal erosion data. Possibly there is an exponential relationship between H m and DR.ne, but the deviation
from the linear relationship seerns to be small, at
least at a continental scale (Harrison 1994).
Some improvement in this approach is obtained by
considering loeal relief, R 1 , R 2 , R 3 ... (the differenee
in m between maximum and minimum elevation
within a local area). This method also roughly includes, indirectly, the density of the river net and the
slope gradients which eontrol de!1udation? ra~er than
the absolute elevation of a reglOn. Taking mto account mean local relief, R.n = R1+R 2 +R 3 •• .ln (n =
number of grids) of a drainage basin, Slingerland et
al. (1994) propose the formula (based on Ahnert
1970):
DR.ne = 0.1535 R.n (mmIka or mlMa).
(9.3)
As shown for major rivers, denudation also correlates fairly well with the relief ratio or the overall
slope of the drainage basin (Fig. 9.12, Summerfield
and Hulton 1994).
To introduce the influence of climate, apart from
relief, Ludwig and Probst (1996) have presented the
following empirical equation for sediment yield, SY
(modified):
SY = 0.02 (Q x SP x VP)
(9.4)
where Q is the specific runoff in the river basin
(mmlyear), SP is the average basin slope (in r~dians,
cf. Fig. 9.12), and VP is the index to charactenze the
variability of precipitation over the year: VP = ~
p2monthsIPyear (mmlyear).
In regions ofprecipitation equally distributed over the year,
VP has lower values than for those of seasonal rain. The
interglacials. "Hard rock" substrate~ show peaks ~f
erosion soon after the onset of glacIal steppe condltions when the forests die and their soils can be easily removed. In the subsequent glacial inten:al ne~
soil formation and thus erosion are limited unhl agam
trees can grow and promote rock weathering and soil
formation.
The Global Mechanical Denudation Rate
Globally, the contemporary total river flux of suspended sediment to the oceans is 15 - 20 x 10 9 tla or
100 to 130 t/km 2 /a (Milliman and Meade 1983;
Tamrazyan 1989; Ludwig et al. 1996). The highest
values of this range come from a very large data base
summarized by Milliman and Syvitski (1992). About
40% of this sediment flux originates from the wet
tropics, 21 % from wet temperate climates, 5% from
boreal climates, and 34% from dry climates (Ludwig
et al. 1996). If 50% of the present-day sediment flux
to the oceans reflects "natural" erosion (e.g. Hay
1998) and the contemporary total soilloss (including
human-induced erosion) is 50-75 x 10 9 tla, it appears
that only c. 15-20% of the eroded soil is delivered to
the oceans (Walling and Webb 1996).
There are significant differences in the specific
sediment yields of the present-day continents (e.g.
Harrison 1994). Asia with its widely extended high
mountain ranges yields 310 tlkm 2 /a, whereas Africa,
Australia, and Europe only deliver 17 to 27 tlkm 2 /a.
The va1ues for North and South America are 88 and
57 tlkm 2 /a, respectively. The natural (pre-European
settlement) sediment yield of Australia is estimated
to have been 2 to 5 times 10wer than the modem one
(Wasson et al. 1996).
A global average specific transport rate of 100
tlkm 2 /a, corresponding to a mechanical denudation
rate of about 40 mm/ka, would lead to an average
sedimentation rate (pore-free) in the world oceans of
about 17 mmIka (also see Sect. 9.4).
9.3.3 General Laws for Mechanical Denudation
Using the empirically gained data discussed above,
several authors have tried to formulate a general law
to express mechanical denudation in relation to relief
and climate. The rock types exposed in the drainage
areas are usually not considered in these attempts.
Taking into account that relief is the main fac~or,
Pinet and Souriau (1988) assurne that the mechalllcal
denudation rate, DR.ne' is largely correlated with the
mean drainage basin elevation (H m in m) above the
base level of erosion (e.g. sea level). They therefore
neglect environmental factors and propose the following empirical formula:
DR.ne = 419 x 10. 3 H m - 0.245
(mm/ka or mlMa)
(1 ka = 1000 years).
441
(9.1)
This simple formula often gives the correct order of magnitude, but sometimes renders results which significantly
differ from those depicted in Figs. 9.13 and 9.14 (see also
Leeder 1991). For example, an elevated plateau which is
not dissected by deep valleys may undergo much le~s denudation than a coastal region with steep slopes, as In South
Africa (Summerfield 1991).
An expression similar to that given above is used by
Pazzaglia and Brandon (1996):
DR.ne = ~ x Hm (mlMa).
(9.2)
Ku is a proportionality constant of denudation (in
short: denudation constant; global average 0.07-0.1
Ma'l). It varies significantly from region to r~gion
(often within a range of two orders of magllltud~)
and therefore has to be calibrated by local mechalllcal erosion data. Possibly there is an exponential relationship between H m and DR.ne, but the deviation
from the linear relationship seerns to be small, at
least at a continental scale (Harrison 1994).
Some improvement in this approach is obtained by
considering loeal relief, R 1 , R 2 , R 3 ... (the differenee
in m between maximum and minimum elevation
within a local area). This method also roughly includes, indirectly, the density of the river net and the
slope gradients which eontrol de!1udation? ra~er than
the absolute elevation of a reglOn. Taking mto account mean local relief, R.n = R1+R 2 +R 3 •• .ln (n =
number of grids) of a drainage basin, Slingerland et
al. (1994) propose the formula (based on Ahnert
1970):
DR.ne = 0.1535 R.n (mmIka or mlMa).
(9.3)
As shown for major rivers, denudation also correlates fairly well with the relief ratio or the overall
slope of the drainage basin (Fig. 9.12, Summerfield
and Hulton 1994).
To introduce the influence of climate, apart from
relief, Ludwig and Probst (1996) have presented the
following empirical equation for sediment yield, SY
(modified):
SY = 0.02 (Q x SP x VP)
(9.4)
where Q is the specific runoff in the river basin
(mmlyear), SP is the average basin slope (in r~dians,
cf. Fig. 9.12), and VP is the index to charactenze the
variability of precipitation over the year: VP = ~
p2monthsIPyear (mmlyear).
In regions ofprecipitation equally distributed over the year,
VP has lower values than for those of seasonal rain. The