Impacts on hydrologie and erosion regimes

Holocene reconstructions of hydrologic and erosion regimes stretching beyond the time-span of documentary records mainly depend on lake sediments or datable, alluvial sequences, supplemented in moist, temperate environments by peat stratigraphic evidence for hydrologic variability (Verschuren and Charman, this volume). In the absence of significant human impacts on catchments, Holocene variability in hydrologic regimes and erosion processes can usually be ascribed to changes in climate (e.g. Knox 2000) or tectonic activity, though the relationships between forcing mechanisms and the resulting responses are seldom simple and direct. Once human activities are superimposed on these relationships, forcing-response linkages often become even more difficult to discern and characterize as more processes interact to introduce additional thresholds and delays in response (e.g. Trimble and Lund 1982; Wasson and Sidorchuk 2000).

Most reconstructions spanning the whole period of human impact on landcover are based on calculations of sediment yield from lake sediment studies. In interpreting sediment yield calculations from lake sediment studies, the effects of catchment size on sensitivity to perturbations, as well as on yield as a percentage of total erosion, must be borne in mind (Dearing and Jones 2003). One implication of the summary in Dearing and Jones is that the most sensitive systems are likely to be small catchments with minimal sediment storage within the drainage net. On the other hand, these may be least representative of processes occurring at a regional scale.

Deforestation and the creation of more open, especially tilled landscapes, has the effect of reducing rainfall interception and evapotranspiration, exposing more of the land surface to rain-splash impact, and increasing hydrologic efficiency within river catchments. These effects tend to increase sediment delivery as well as to increase the volume of material derived from surface soil and slope wash, rather than from within-channel sources, although within-channel instability may also be triggered by land-cover-driven changes in hydrology. These effects are additional to those arising from changed feedback to atmospheric processes considered above. Figure 3.6 shows some European examples of mainly lake-sediment-budget studies illustrating links between sediment yield and human occupation of the catchment. In all cases, there are clear links between the history of human occupation and sediment yield, but they only emerge when the time frame exceeds the shorter time-scales on which individual flood events have been documented. Foster et al. (2003), in stating that "as the time scale of observation becomes shorter, changes in climate and hydro-meteorological conditions become progressively more important" echo earlier observations in a range of environmental contexts (cf. Messerli et al. 2000). Equally, as the time perspective lengthens, the role of human activities in transforming the landscape upon which climate variability acts often becomes more important (Lang et al. 2000; Sidorchuk 2000).

Moving beyond Europe to other long-settled parts of the world, a growing number of Chinese studies are shedding a similar light on the impact of human activities on erosion and sedimentation. Xu (1999) in his study of the Yellow River detects a sharp increase in mean sediment accumulation rate as a result of increased human impacts over the past two to three millennia, as do Saito et al. (2001) in their calculations based on deltaic sedimentation at the mouths of both the Yangtze and Yellow rivers. These inferences are strongly reinforced by He et al. (2006) who detect a sharp increase in erosion rates over the same period on the Loess Plateau of China, culminating in rates some four times those prevailing before significant anthropogenic impact. The impact of these changes on sediment delivery during the extreme flood event of 1998 is evaluated in a simulation study by Xu et al. (2005). Alongside these studies based on major regions and river systems, detailed multi-disciplinary research on individual lake catchments is

Figure 3.6 Long-term changes in sediment yields from lake sediments and alluvial sequences. (a) Frequency of OSL-dated soil-erosion-derived sediments from the loess hills of southern Germany (Lang 2003). (b) Sediment accumulation rates at Holzmaar, western Germany (Zolitschka 1998). (c) Frequency of dated alluvial units in British rivers (Macklin 1999). (d) Changes in the deposition of inorganic mineral matter at Lago di Mezzano, central Italy (Ramrath et al. 2000). (e) Depositional flux of magnetic minerals derived from soil erosion into the mid-Adriatic core RF 93-30 (see Figure 3.4). (f) Alluvial accumulation rates in the Yellow River, China (Xu 1999). Note the close parallels (allowing for some imprecision in chronologies) between (a) and (b) and between (d) and (e). (Reproduced from Oldfield F. (2005) Environmental Change: Key Issues and Alternative Approaches. Cambridge University Press, Cambridge.)

87654321 0 IR-OSL age

86420

86420

Alluvial units

120 -

X

m

u

CM

80

b c

40

>

g

h

0

0.5

X

u

-

Thousand years BP

beginning to shed important light on the links between China's long history of human occupation and the effects on erosion and sediment accumulation rates. In a wide-ranging study of Erhai Lake in south-west China (see summary in Dearing, 2006) detectable human impact on the landscape is shown to stretch back at least 7000 years, with increased topsoil erosion from 2200 years BP in response to the expansion of agriculture by Han peoples. Peak erosion occurred during the 17th to 19th centuries ad. One inference from the study relates to the likely role of irrigation systems from 4300 years BP as moderators of the hydrologic regime within the catchment: evidence for minimum flood peaks declines on millennial time-scales from that date to the present day, despite evidence for a strengthening monsoon system over the past 1500 years.

Evidence for the dramatic impact on erosion by the first European settlers in North America is widespread. The classic study of Davis (1976) at Frains Lake, Michigan is paralleled by others from as far afield as coastal California, where Plater et al. (2006) detect an order of magnitude increase in sediment accumulation rates on the arrival of the first European immigrants. Where it has been possible to discriminate between sediment derived from topsoil and that derived from deeper in the regolith, accelerated loss of topsoil is seen to coincide with the first evidence for deforestation and tillage by European settlers (cf. Oldfield et al. 1985; Yu and Oldfield 1989).

In so far as the results of the many and varied studies of Holocene erosion and sedimentation encourage generalizations, they support the view that the landscape, as transformed by forest clearance and tillage for example, creates a new canvas upon which climate variability and especially hydrologic extremes act. As the land-cover is transformed, the impact of continued variability will change even without a shift in the magnitude/frequency of extreme events. The interplay through time between climate variability and changing land-cover introduces the type of nonstationarity in hydrologic response that negates the assumptions upon which magnitude/frequency projections are based. This is a significant concern for the future (Clarke 2003). Only the empirical evidence from a long-term view of the past can provide well-founded guidance. The need for long-time perspectives in order to understand the nature of human impacts on ecosystems and the hydrologic implications of these impacts is strongly reinforced by Batchelor and Sundblad's (1999) schematic representation of impacts and recovery times (Figure 3.5).

0 0

Post a comment