About This Item
Share This Item
In the advanced industrial countries the most favorable, least expensive sites for surface reservoirs are already in use or the land already is preempted for other uses and is unavailable for the storage of water. In addition, there are many flat areas in coastal zones, also underlain by saline aquifers, that are unsuitable for water storage although a surplus of fresh water is available in such areas at certain times of year. The lack of a reliable, year-round supply of water has been a major factor in preventing commercial and residential development in these areas.
The storage of fresh water in slightly saline aquifers has been tried empirically several times with some success. To study the physical process in the laboratory we have constructed and operated several miniaquifers and, simultaneously, have devised some approximate mathematical models. The annual cycle of injection, storage, and withdrawal of the fresh water has been found to be feasible under the idealized assumptions normally found in groundwater hydrology--a horizontal, isotropic, homogeneous aquifer of uniform porosity, transmissivity, and storativity. Laboratory experiments on a single-well system built into a miniaquifer constructed of epoxy-consolidated, uniform blasting sand show that the efficiency of the process, per cycle, increases as the number of cycles increases. Our omputational procedure verifies this and has enabled us to change inexpensively and quickly such parameters as density difference, dispersion coefficient, input rate and period, withdrawal rate, storage period, etc. The studies show that storage of fresh water in an aquifer that contains brine is feasible, if a sufficient number of cycles is considered. The cost, in terms of irretrievable fresh water, is calculable under these conditions.
Additional studies were and are being made on a 9-unit well field. Preliminary results show that although the recovery percentage at the end of the first cycle is smaller than that of a single well operating by itself, by the time the third cycle is reached a multiwell system is more efficient. A greater percentage of the water injected during the third cycle is recovered than is recovered by a single well under the same circumstances.
Most water-bearing formations dip and, in many, a measurable groundwater flow occurs under natural, undisturbed conditions. Each of these circumstances affects the position and configuration of the "bubble" of fresh water. For example, the injected fresh water is lighter than the saline water and should tend to move to the roof of the aquifer and thence updip. This should result in a lower recovery efficiency compared with that from a horizontal aquifer. However, the recovery efficiency depends greatly on the duration of the storage part of the cycle. Single-well experiments in a dipping aquifer verify and quantify this expectation. There are indications that it may be possible to overcome the effect of dip and to stabilize the position of the injected fresh water by constructing and perating a system of injection and withdrawal well updip and downdip from the injection well.
This paper is a progress report on work that is well underway but not yet complete. As to the effect of various combinations of dip, movement of native ground water, and density differences, on the recovery efficiency of a multiwell project, we have some qualitative ideas, but at present are trying to devise a quantitative basis for design that will handle all of the variables simultaneously.
End_of_Article - Last_Page 1598------------