Abstract:
Extensive exploitation of groundwater during past two decades, has drawn the attention of groundwater hydrologists towards the problem of maintaining perennial supply of groundwater. Recharge to groundwater on account of percolation from precipitation falling on the outcrop of an aquifer is limited. Since, a stream derives its flow from precipitation over a relatively larger drainage area, large and perennial groundwater supply can better be maintained by inducing percolation from a perennial stream hydraulically connected with the aquifer. With favourable hydrogeological condition and permeable stream-bed. induced infiltration into an aquifer takes place when wells are pumped near the stream. Thus, pumping near effluent reaches of the stream may form an important factor in planning conjunctive use practices. The pumping rate may vary with time from conjunctive use management point of view, hence analytical solutions for drawdown, and rate and volume' of stream depletion for variable/intermittent pumping are required.
Application of the analytical solutions to field problems or for conjunctive use planning, requires that the parameters, (i.e.. transmissivity, storage coefficient, and effective distance to the recharge boundary), are known before hand. These parameters can be determined utilizing the drawdown data obtained during a pump test conducted near the recharge boundary. Methods developed so far, for determination of the parameters either use curve matching, or involve the procedure of finding inflexion point on drawdown vs. time graph. Therefore, use of these methods suffers from errors due to personal judgement.
The present report deals with the development of analytical solutions for unsteady drawdown, and rate and volume of stream depletion for variable rate pumping. A method has been evolved for estimation of the parameters making use of the drawdown data obtained during a pump test_ The method takes into account the variable pumping rate. A derivative-based non-linear optimization technique has been used for the estimation of the parameters. Analytical derivatives w.r.t. the parameters have been used in the optimization. The proposed method do not require use of tables or curve matching as it is based on objective criteria of minimization of sum of the squares of errors between observed and computed drawdowns.
Computer codes in Fortran for both direct problem and inverse problem for variable rate pumping near a recharge boundary, have been developed. The estimated parameters using present methodology are found to be more reliable as compared to that obtained using traditional method.