In an earlier paper (Bako and Hunt, 1988), a method for the derivation of the baseflow recession constant (K) using one-way analysis of variance was presented. This paper presents the results of the field application of this method.
The K values obtained by using the numerical equation of Bako and Hunt (1988) were inserted in the exponential recession equation (Barnes, 1939) to generate a series of baseflows. The fit between the model and the historical flows was found to be greater than 99 per cent thus confirming the applicability of the numerical method under field situation. The main advantage of this technique is its amenability for computerized application thus making it relatively faster than any of the existing techniques of fitting the recession equation. For this reason, the subjectivity inherent in most of the existing techniques is eliminated and a measure of procedural consistency can be guaranteed. Consistency is necessary if intercatchment comparison or interpolation of K values is to be meaningful.
KEY WOHDS
Groundwater recharge Baseflow recession Hydrograph separation Streamflow components
RECESSION EQUATIONS
The flow of a river often comprises contribution from different sources, viz direct runoff, bankflow, interflow, and groundwater discharge or baseflow. Which component dominates in a particular situation depends on the climatic, topographic, and geological factors.
Although to varying degrees, all the four sources contribute to streamflow during the wet season. However, in dry periods, the flow is largely supported by contribution from groundwater. An understanding of the mechanism of groundwater depletion in a catchment is essential for effective planning and management especially in regions where the dry period could be prolonged. The depletion of groundwater follows an exponential decay (Barnes, 1939). A suitable equation is of the form where
During the dry period, the flow in the stream is entirely composed of groundwater. Hence the dry season discharge of the draining river can be used in Equation (1) to derive K.
Equation (1) can be rewritten as loge, = loge, + tlogK Thus the value of K can be determined from the slope of the plot of loge, versus t .