Tropical forest systems: A hydrological approach

This paper briefly examines the importance of considering the rates and magnitudes of water movement in the hillslope-river system of a tropical rainforest catchment. It is proposed that consideration of water movement is a fundamental component in understanding the release and movement of nutrients in this environment. In any such analysis it is essential that the 'opportunity time' or 'residence time' together with the availability of weatherable minerals be considered. Three conditions are suggested to account for the low solute concentrations in stream waters, each, any or all three of which may occur. (1) If there are no soil nutrients of importance then there can be supply neither to the river nor the plants. (2) If the residence time is too short relative to the equilibriation time of the minerals, then weathering and exchange may not occur. (3) If the residence time is too long (because rate of movement is slow), the 'turnover' will be small. In this context the analogy of an overflowing cup is discussed as a possible explanation of low solute concentrations. The results presented in the paper refer to the period 6th - 26thMay 1977, from a small hillslope-river segment at Resewa Ducke, Amazonas. Measurements made included soil tension, piezometric levels, river stage, infìltration rates and wetting front movement. Using Darcy's Law, water fluxes are determined. Draw down characteristics of the piezometers and river stage have been estimated using regressions of the logarithms of both these variables against the logarithm of time. The results suggest that during the period of observation the slope is almost saturated with respect to water. Actual saturation (positive pressures) are observed to occur at the foot of the slope under all conditions and within the slope during the earliest set of observations. Results from the computation of water fluxes indicate little lateral movement, the dominant flow is at or very close to vertical. Analysis of piezometer level and river stage suggests a very close link between the two, with only limited influence from the adjacent hillslope. In conclusion it seems that during the wet season, most of the river flow is generated by rapid rise beneath the floodplain and the slope immediately adjacent to the floodplain as a direct result of rainfall infiltration and that throughflow is unimportant. This is consistent with certain aspects of the cup analogy and goes far to explain the very low solute concentration found in the water of this and similar barrancos.