9.3 Me~ L'lanical Denudation
Mechanical Denudation Rates in Relation to
Relief and Climate
The river-load method (including the filling of artificial reservoirs) has been applied for small and large
rivers all over the world. There is a great body of
data now available, but detailed information about
the quality of these data and the characteristics of the
corresponding drainage areas is often missing. These
shortcomings, but also the complexity of the proces ses involved in weathering, erosion, and sediment
transport have so far prevented the establishment of
clear rules for the evaluation of sediment yields and
mechanical denudation rates. The following discussion is therefore mainly based on empirical results
rather than a theoretical approach.
As has been pointed out in Section 9.3.1, specific
transport rates of total suspended solids, TSS, or specific sediment yields, SY (tonsIkm 2 /a) of rivers and
the mean denudation rates, D~e (mmlka or m1Ma)
of the their drainage areas are closely related (Fig.
9.7). Most of the available data are solely based on
measurements of the suspended load, TSS, but the
error caused by neglecting the bedload is normally
small except for rivers leaving high mountain ranges.
Results from Major River Systems
Major world rivers usually drain regions of different
relief, climate, and rock type (catchment areas of the
order of 10 6 km 2 ). The influence of the factors controlling denudation is therefore not very clear in
these cases. Nevertheless, large river systems show a
distinct increase in their mean denudation rate with
increasing relief (Fig. 9.12). Even the overall relief
ratio, which is easy to determine (compared with the
mean local relief), describes this relationship fairly
weIl (Summerfield and Hulton 1994). The second
factor of some importance for mechanical denudation
is c1imate in terms of mean annual runoff which, on
the other hand, mainly controls chemical denudation
(Fig. 9.12, cf. Sect. 9.2).
Results from Selected Regions and Artificial
Reservoirs
A better evaluation of the factors controlling mechanical denudation can be made for smaller river
systems and the catchment areas of artificial reservoirs. Smaller areas are less heterogenous than larger
ones; they often represent regions in which a certain
relief, c1imate and rock type dominates. Using the
large data base collected by Enos (1991), Milliman
and Syvitski (1992), and data summarized by Einseie
and Hinderer (1997) one can draw the following conclusions (Figs. 9.13aand b):
437
1. Rivers draining high-relief areas may produce specific sediment yields, SY, of suspended matter and a
bedload of some 100 to more than 1000 tlkm 2 /a.
These values correspond to mean denudation rates of
50 to 200, up to 1000 mmlka (1 mmlka = 2.5 t!km 2 /a
for a mean rock density of 2.5 g/cm 3 ). Denudation in
high-relief areas is about two orders of magnitude
greater than that of lowland rivers. Steep valley
slopes, insufficiently protected by vegetation, and
steep stream-channel gradients favor erosion and
sediment transport by running water. Annual or seasonal precipitation and rainstorm events are often
greater in high mountain ranges than in lowlands.
However, one should bear in mind that poorly
lithified sediments deliver a great part of the sediment yield of these regions. Old mountain belts display lower denudation rates than young ones, because their weIl erodible lithologies have been removed by earlit'!' 0rosion (Stallard 1995).
2. Mountain glaciers may enhance the sediment yield
of non-glaciated regions bya factor of 2 to 10. Erosion of wet-based, rapidly moving glaciers in high
mountain ranges produces specific sediment yields in
the order of 500 to more than 5000 t!km 2 /a, often
around 2000 t!km 2 /a and thus denudation rates of
about 200 to 2000 mmlka (summary in Einseie and
Hinderer 1997). However, glacial erosion under large
continental ice sheets is less effective, as for example
in Antarctica (Alley et al. 1988; Henrich 1990). Estimates of the present global flux of glacial detritus to
the oceans vary greatly (Hay 1998).
For the Glacier Bay area of Alaska, an average denudation
rate of 12 m/ka has been suggested (PoweIl and Molnia
1989). The rapid1y moving, cool-temperate glaciers of
southem and southeastem Alaska produced sedimentation
rates on the shelf along the north-east coast of the Gulf of
Alaska of 10-20 mika in the Holocene (Jaeger et al. 1998).
Here, the total annual sediment accumulation on the shelf
is 250 x 10 6 tla. Ifthe source area is assumed to be 5 x 10 4
km 2 in size, then the specific sediment yield of the actively
rising, glaciated Alaskan mountain range is in the order of
5000tlkm 2 /a which is probably one of the highest values
on the present-day globe. On the Alaskan shelf edge and in
deeper water the sedimentation rate amounted to about 0.5
mlka during the past 6 Ma (PoweIl and Molnia 1989).
On the Barents shelf, glacigenic sediments accumulated
at rates between 0.1 and I m/ka; fjord troughs were filled at
rates up to 60 mika (Vorren et al. 1989). In the Ross Sea of
the Antarctic region, however, Pleistocene diamicton accumulated at a much lower rate (around 10 mmlka, Alley et
al. 1988).
In particular, phases of deglaciation can lead to reworking and redistribution of older glacial sediment as observed
in marine sediments offnorthem Norway (Haflidason et al.
1998). Here, about 90% of the sediment mass of the past
15 ka accumulated during deglaciation (15-9 ka B.P.) and
only 10% (hemipelagic sediment) in the Holocene. Likewise, the filling of Alpine lakes by delta prograding is
mainly associated with late Pleistocene deglaciation (cf.
Chap. 11.3).
Mechanical Denudation Rates in Relation to
Relief and Climate
The river-load method (including the filling of artificial reservoirs) has been applied for small and large
rivers all over the world. There is a great body of
data now available, but detailed information about
the quality of these data and the characteristics of the
corresponding drainage areas is often missing. These
shortcomings, but also the complexity of the proces ses involved in weathering, erosion, and sediment
transport have so far prevented the establishment of
clear rules for the evaluation of sediment yields and
mechanical denudation rates. The following discussion is therefore mainly based on empirical results
rather than a theoretical approach.
As has been pointed out in Section 9.3.1, specific
transport rates of total suspended solids, TSS, or specific sediment yields, SY (tonsIkm 2 /a) of rivers and
the mean denudation rates, D~e (mmlka or m1Ma)
of the their drainage areas are closely related (Fig.
9.7). Most of the available data are solely based on
measurements of the suspended load, TSS, but the
error caused by neglecting the bedload is normally
small except for rivers leaving high mountain ranges.
Results from Major River Systems
Major world rivers usually drain regions of different
relief, climate, and rock type (catchment areas of the
order of 10 6 km 2 ). The influence of the factors controlling denudation is therefore not very clear in
these cases. Nevertheless, large river systems show a
distinct increase in their mean denudation rate with
increasing relief (Fig. 9.12). Even the overall relief
ratio, which is easy to determine (compared with the
mean local relief), describes this relationship fairly
weIl (Summerfield and Hulton 1994). The second
factor of some importance for mechanical denudation
is c1imate in terms of mean annual runoff which, on
the other hand, mainly controls chemical denudation
(Fig. 9.12, cf. Sect. 9.2).
Results from Selected Regions and Artificial
Reservoirs
A better evaluation of the factors controlling mechanical denudation can be made for smaller river
systems and the catchment areas of artificial reservoirs. Smaller areas are less heterogenous than larger
ones; they often represent regions in which a certain
relief, c1imate and rock type dominates. Using the
large data base collected by Enos (1991), Milliman
and Syvitski (1992), and data summarized by Einseie
and Hinderer (1997) one can draw the following conclusions (Figs. 9.13aand b):
437
1. Rivers draining high-relief areas may produce specific sediment yields, SY, of suspended matter and a
bedload of some 100 to more than 1000 tlkm 2 /a.
These values correspond to mean denudation rates of
50 to 200, up to 1000 mmlka (1 mmlka = 2.5 t!km 2 /a
for a mean rock density of 2.5 g/cm 3 ). Denudation in
high-relief areas is about two orders of magnitude
greater than that of lowland rivers. Steep valley
slopes, insufficiently protected by vegetation, and
steep stream-channel gradients favor erosion and
sediment transport by running water. Annual or seasonal precipitation and rainstorm events are often
greater in high mountain ranges than in lowlands.
However, one should bear in mind that poorly
lithified sediments deliver a great part of the sediment yield of these regions. Old mountain belts display lower denudation rates than young ones, because their weIl erodible lithologies have been removed by earlit'!' 0rosion (Stallard 1995).
2. Mountain glaciers may enhance the sediment yield
of non-glaciated regions bya factor of 2 to 10. Erosion of wet-based, rapidly moving glaciers in high
mountain ranges produces specific sediment yields in
the order of 500 to more than 5000 t!km 2 /a, often
around 2000 t!km 2 /a and thus denudation rates of
about 200 to 2000 mmlka (summary in Einseie and
Hinderer 1997). However, glacial erosion under large
continental ice sheets is less effective, as for example
in Antarctica (Alley et al. 1988; Henrich 1990). Estimates of the present global flux of glacial detritus to
the oceans vary greatly (Hay 1998).
For the Glacier Bay area of Alaska, an average denudation
rate of 12 m/ka has been suggested (PoweIl and Molnia
1989). The rapid1y moving, cool-temperate glaciers of
southem and southeastem Alaska produced sedimentation
rates on the shelf along the north-east coast of the Gulf of
Alaska of 10-20 mika in the Holocene (Jaeger et al. 1998).
Here, the total annual sediment accumulation on the shelf
is 250 x 10 6 tla. Ifthe source area is assumed to be 5 x 10 4
km 2 in size, then the specific sediment yield of the actively
rising, glaciated Alaskan mountain range is in the order of
5000tlkm 2 /a which is probably one of the highest values
on the present-day globe. On the Alaskan shelf edge and in
deeper water the sedimentation rate amounted to about 0.5
mlka during the past 6 Ma (PoweIl and Molnia 1989).
On the Barents shelf, glacigenic sediments accumulated
at rates between 0.1 and I m/ka; fjord troughs were filled at
rates up to 60 mika (Vorren et al. 1989). In the Ross Sea of
the Antarctic region, however, Pleistocene diamicton accumulated at a much lower rate (around 10 mmlka, Alley et
al. 1988).
In particular, phases of deglaciation can lead to reworking and redistribution of older glacial sediment as observed
in marine sediments offnorthem Norway (Haflidason et al.
1998). Here, about 90% of the sediment mass of the past
15 ka accumulated during deglaciation (15-9 ka B.P.) and
only 10% (hemipelagic sediment) in the Holocene. Likewise, the filling of Alpine lakes by delta prograding is
mainly associated with late Pleistocene deglaciation (cf.
Chap. 11.3).