Multi-year instrumental records for input, throughflow and output waters of the Lilburn Cave system provide control on denudation rates as they respond to seasonal and spatial variability. Data suggest that maximum denudation is in the late fall and early winter. This is when non-snowmelt discharge is at its maximum. At lower discharge rates the volume of water moving through the cave system is the limiting control on the volume of denudation. During periods of snowmelt the limiting control is the rate at which the calcite dissolves. This is probably the result of water flowing through wider channels during these times. Based on instrumental measurements, there is considerable variation in terms of where denudation occurs inside the cave. The loci of dissolution change from year to year. This is to be expected in the dynamic environment of the cave where materials shift routinely. This variability should be studied over longer periods of time in order to more fully understand its extent. The relatively small area of carbonate exposure relative to the area of the drainage basin gives rise to relatively high denudation rates. The carbonate is being removed at a rate of about 5000 metric tons per year, or at about 830 mm/y. This is about five times the rate reported in the humid karst regions of Malaysia. This information indicates that the relative proportion of carbonate in the drainage basin needs to be considered when trying to estimate denudation in other areas.
Analytical studies for well design adjacent to river banks are the most significant practical task in cases involving the efficiency of riverbank filtration systems. In times when high pollution of river water is joined with increasing water demand, it is necessary to design pumping wells near the river that provide acceptable amounts of river water with minimum contaminant concentrations. This will guarantee the quality and safety of drinking water supplies. This article develops an analytical solution based on the Green's function approach to solve an inverse problem: based on the required level of contaminant concentration and planned pumping time period, the shortest distance to the riverbank that has the maximum percentage of river water is determined. This model is developed in a confined and homogenous aquifer that is partially penetrated by the stream due to the existence of clogging layers. Initially, the analytical results obtained at different pumping times, rates and with different values of initial concentration are checked numerically using the MODFLOW software. Generally, the distance results obtained from the proposed model are acceptable. Then, the model is validated by data related to two pumping wells located at the first riverbank filtration pilot project conducted in Malaysia.