Well Field Design in a Highly Fractured Aquifer System : A Case Study of the Egypt's Siwa Oasis

In highly fractured aquifer systems, simulation of groundwater flow using the well known classical approaches such as numerical and analytical models is practically impossible in most cases. Recent advances in research in the area of model identification have revealed approaches for inducing models from data, based on learning systems. These approaches can determine the relationships between the input and output variables from data presented to them, without resorting to describing these relationships explicitly in mathematical form. The design of the free flowing groundwater well fields in such aquifer system requires the determination of the free flowing rate and the separation distance between the wells to avoid any undesirable interference. The artificial neural network approach as a model induction from data was used to estimate these two variables in the aquifer system of Egypt's Siwa Oasis, which is characterized by its high fracture intensity and the wide variation of the fracture apertures. Combined with already available understanding of the physical process, the developed model results showed consistency with the observed data to a high degree of confidence. This study is intended to assist the decision makers dealing with groundwater resource management in such complicated system.