Effects of partial and full pipe flow on hydrochemographs of Logsdon river, Mammoth Cave Kentucky USA

Abstract

Summary Logsdon River, within Kentucky’s Mammoth Cave System, drains the 25 km2 Cave City groundwater basin. Temperature, specific conductance, stage, and velocity were measured at the river’s downstream end within the cave with 2-min resolution for one year beginning on May 5, 1995. New interpretation of these data show relationships to the conduit’s geometry and fluid transport behavior. The 37 observed storm hydrographs/chemographs are classified into two groups: open channel flow and pipe flow (conduit filled). In the 16 pipe flow events Logsdon River exceeded the capacity of its conduit, raising the water table above the conduit ceiling with some storm water thus temporarily stored within the aquifer adjacent to the conduit. Pipe flow events were characterized by depression of specific conductance (spC) before peak discharge, a peak in spC at peak discharge, and a second depression in spC after peak discharge. The first spC minimum represents the maximum concentration of passing storm water and the second is interpreted as the result of recent storm water draining back into the conduit, having been temporarily stored in the aquifer adjacent to the conduit. Mean velocity decreases in the transition from partial to full pipe flow due to increased head loss at full pipe flow compared to nearly full pipe flow. Each of the open channel flow cases shows a depression in spC either coincident with, or after, peak discharge. While the hydrochemographs of pipe flow are mainly controlled by the aquifer conduit network geometry, in open channel cases they are mainly controlled by external recharge conditions, including behavior of sinking streams.