Genetic Algorithms and Gauss Newton Marquardt optimization methods were coupled with the finite element simulation model to predict the zonal hydraulic conductivity and system recharge. These compared favourably with the field obtained values. This technique auto-calibrates groundwater flow and transport models and has considerable importance from the field application point of view.
The hydrodynamic dispersion model was coupled with the ground water flow model to predict the solute concentration values. Soft computing tools were applied to get the concentration predictions which compared favourably with the obtained field values.
Among other things, the work involved carrying out 200 vertical electrical soundings to assess subsurface lithological variations, followed by the drilling of the 24 bore holes to estimate aquifer parameters and velocity vectors within the flow domain. A groundwater quality model was prepared to forecast the distribution of contaminants from the ash pond to the underlying aquifer for a planning period of 25 years. Results indicated that, whereas, due to considerable retardation, the extent of heavy element pollution will be almost negligible, the TDS plume will spread over an area of 3 square kilometers around the fly ash pond. An efficient finite element algorithm was developed to solve the contaminant transport equation.
Genetic Algorithms and Gauss Newton Marquardt optimization methods were coupled with the finite element simulation model to predict the zonal hydraulic conductivity and system recharge. These compared favourably with the field obtained values. This technique auto-calibrates groundwater flow and transport models and has considerable importance from the field application point of view. Study involves examining a strategy for freshwater pumping (location and rate) with a combined system of treated water recharge and saltwater discharge wells which creates a barrier to control seawater intrusion.The numerical model developed is capable of predicting seawater intrusion in coastal aquifers over a wide range of field parameters. The coupled ground water flow and solute transport equations are expressed in terms of freshwater head and solute concentration where, both, are dependent variables. The verified FEM model examines the sensitivity of dispersivity and hydraulic conductivity on seawater intrusion in confined coastal aquifer.Suitable locations for battery of pumping wells are worked out for planned withdrawal of freshwater from the coastal aquifers. Location of pumping wells, their depth and fresh water well head values were worked out which play very important role in safe delivery of freshwater in coastal aquifers. The study finds safe region for pumping wells location. As an extension a three layer synthetic coastal confined aquifer is also considered for evaluation based upon above studies and several safe regions for fresh water intake were identified.
Finite difference forward in time central in space explicit scheme was used to simulate the Blue Lake Aquifer Model. Non linear optimisation package (Stanford University) MINOS was used to obtain the optimal pumping policy to meet the future demand from the aquifer. The results indicated that the system is capable to meet the demands without any adverse effect.
One of the most difficult problems in large system simulation is to find the applicable elemental recharge for heterogeneous anisotropic unconfined aquifers. This technique is found particularly suitable for the very large aquifer systems. It worked out applicable elemental recharge to the aquifer. Their application showed good agreement between the simulated and the observed head distribution in the flow domain.
An efficient interactive finite element algorithm was developed, tested and then used to simulate three distinct sections of the lined Nadiad branch canal involving layered subsurface structure. The digital solutions indicated a net seepage loss of 9600 cubic meter from the chosen 8.6 kms length of the canal. Seepage values were found to be more sensitive to aquifer hydraulic conductivity and adjacent water level compared to top soil and canal liner hydraulic conductivity values. Study concluded that the existing worldwide practice of relying on a single seepage rate from a canal is inadequate and explicitly showed two substantially different seepage rates in a year. The completed study clearly brought out the necessity of dynamic canal seepage assessment in contrast to the existing practice of assuming a fixed seepage rate for a canal. It was indicated that proposed canals seepage losses should be and can be investigated in detail before actual construction work begins which aids considerably in a planned management of the available water resources.
In this study the effect of short period water waves on the longitudinal mixing of pollutants is examined. Experiments were conducted on non wavy and wavy flow and the values of longitudinal dispersion coefficients were found out on measured solute concentrations at various time intervals. Since a river flow normally involves wind generated short period waves, it is essential to consider the wave phenomena in computing the dispersion coefficient. The study has demonstrated that the magnitude of longitudinal dispersion coefficient in wavy and non wavy cases is distinctly different. The negative horizontal velocity of the particle travelling under the crust of the wave is the major factor to decide the magnitude of wave induced longitudinal dispersion coefficient. This component pulls back the particle moving in forward direction, which results in the churning action. An increase in the magnitude of this component causes corresponding increase in the churning of particles and therefore enhances the magnitude of wave induced longitudinal dispersion coefficient. Change in the magnitude of longitudinal dispersion coefficient in wavy environment is clearly depicted in the present work. The rate of increase of the wave induced longitudinal dispersion coefficient is more with increase in amplitude than increase in the wave period. Also the effect of waves on the dispersion of pollutants reduces with increase in mean water depth. From the time concentration break through curves it is observed that larger the values of wave amplitude and wave period, higher is the magnitude of initial concentration gradient due to increasing churning action.