HR: 08:00h
AN: H51A-01 INVITED     [PDF]
TI: Overland Flow in Tropical Rainforests: Causes and Consequences
AU: * Elsenbeer, H
EM: helsenb@rz.uni-potsdam.de
AF: Institute of Geoecology University of Potsdam, PO Box 60 15 53, Potsdam, 14415 Germany
AU: Godsey, S
EM: godseyse@email.uc.edu
AF: Dept of Earth and Planetary Science UC Berkeley, 307 McCone Hall, Berkeley, CA 94720-4767
AB: With the expansion of forest hydrology research from mid-latitudinal to tropical areas, a diversity of hydrological process patterns began to emerge that has since been tentatively explained along geographical lines. Our much improved knowledge of runoff generation processes, however, allows us to discard geographical straitjackets and explain the quite diverse hydrological functioning of tropical rainforests world-wide in terms of those soil, meteorological, and geomorphologic factors that control the movement of water on landscapes. An appreciation of these factors, which do not include land cover, is essential for an understanding of the occurrence of overland flow in tropical rainforest, which still seems to be viewed as a contradiction in terms in spite of mounting evidence to the contrary. A perceived link between hydrological processes and land cover and the intuitively appealing, perceived role of forests in the water cycle has resulted in overland flow being viewed as a consequence of human intervention, but not as the natural consequence of a set of environmental factors inherent in many tropical rainforests. This set of environmental factors includes characteristic permeability-depth profiles and rainfall regimes that, in combination, will yield overland flow as a matter of course; in fact, overland flow has emerged as a hydrological process characteristic of vast areas of rain forest-covered tropical soilscapes. These characteristic permeability-depth profiles may, in a simplifying approximation, be reduced to a sequence of two soil layers: a thin (compared to the overall soil depth) surficial layer of high permeability (compared to prevailing rainfall intensities), and a thick subsurface layer of very low permeability. This sequence reflects the decrease of nonstructural, or bioturbation-induced, macroporosity (which is in many places accompanied by a textural change) observed in some widespread tropical soils; it contrasts with rather uniform permeability-depth profiles that reflect nearly depth-independent structural, or mineralogy-controlled, macroporosity observed in other tropical soils. The latter result in a generally vertical movement of soil water, whereas the former tends to generate saturation overland flow. With overland flow established as a characteristic hydrological process of some extensive tropical rainforest ecosystems, we explore its implications for land-use change and nutrient budget studies. Where overland flow is part of undisturbed ecosystem processes, land-use change cannot possibly result in a qualitative change of hydrological processes: it may increase the amount of overland flow, but it cannot cause its occurrence. Although land-use and land cover change of tropical rainforests do affect hydrological processes in many ways, there is no evidence to support generalizations about abrupt shifts in hydrological pathways. Likewise, the effect of land-use change on nutrient losses is likely to be overestimated where the `natural' nutrient losses were underestimated by ignoring overland flow-mediated and hence rainfall event-related catchment outputs. Overland flow must be recognized as an intrinsic, even typical, hydrological process of many tropical rainforests, while the casual assumption of its occurrence as the unique consequence of land-use change is not defensible.
DE: 1860 Runoff and streamflow
SC: Hydrology [H]
MN: 2003 Fall Meeting