WHAT'S IN A SPRING HYDROGRAPH?

Many previous studies have used spring hydrographs following rain events to try to determine properties of a karst aquifer. However, little work has been done on the influence of the functional form of the aquifer recharge on the discharge hydrograph. In order to study the manner in which karst conduits modify input flood hydrographs, we develop a simple conceptual model of flood pulse propagation through karst conduits.

First, we identify three fundamental behaviors of karst conduits in response to flood pulses. Conduits can act as full pipes, open channels, or reservoir/constriction systems. Using the equations that govern each of these response types, we analytically derive characteristic response times for each type of conduit flow. We then define a dimensionless number that is the ratio of the conduit response time to the characteristic time of a recharge event. This number can be used to characterize the response regime of a given conduit for a given flood event. Specifically, we demonstrate that using this dimensionless number, the responses of simple karst aquifers can be divided into geometry-dominated and recharge-dominated regimes, in which the discharge hydrograph is either modified or resembles the recharge hydrograph, respectively. Given typical parameters of karst conduits, we conclude that flood pulse modifications are unlikely to occur in simple full pipe or open channel portions of conduit. However, reservoir/constriction systems can significantly modify flood pulses, and the fractured or matrix component of the aquifer is almost certain to modify input pulses.

This study provides important constraints on the amount of information that can be derived via inverse modeling of spring hydrographs, and in particular delineates a regime in which the discharge from the karst spring is nearly identical to the recharge and thus contains little information about the conduit geometry. Further work is underway to extend this approach to more complex networks of karst conduits.