Effect of Non-Instantaneous Release of Air Pressure on Pneumatic Slug Tests

Pneumatic initiation of slug tests has been used for many years and is an efficient technique, since water is not added or withdrawn from the well and no mechanical slug is required. However, for very permeable aquifers the release of air pressure or vacuum can not be considered instantaneous, as is usually assumed. This paper investigates the response time at the lower end of an air column where the water interface would be by using a pressure transducer, which can measure both positive and negative pressures. As the valve at the upper end of the air column is opened the disturbance propagates through the air column and bounces off the bottom of the air column. The disturbance continues propagating and bouncing off the ends while the air escapes from the valve. The net result is an oscillatory air column response that decays exponentially with time. The step function release that is usually assumed is not obtained in practice. Detailed measurements show that the decay is an exponential function of the square root of time and a decay constant. The period of the oscillation is dependent on the length of the air column, with shorter lengths giving higher frequencies. The decay constant seems to vary inversely with the square root of the initial pressure and directly with the air column length for a given initial pressure. For configurations tested and typical field situations the air column goes to atmospheric pressure within about 2 seconds. This means that for slow responding aquifers the effect is minimal. However, for very highly permeable aquifers where the slug test lasts only a few seconds ( we have tested many in this category), the deviation from step function release of air pressure or vacuum can have a significant effect on the data analysis using a specific model for slug test response. In particular, the aquifer response will be stretched out and the effective initial height of the slug will be less than expected. Where positive and negative pressures of the same magnitude are used to check for a directional dependence of aquifer response, there will be a difference in the response due to the difference in length of the air column, which is not related to any aquifer characteristic. For maximum accuracy in determining hydraulic conductivity of highly permeable aquifers from pneumatic slug tests, this effect of non-instantaneous release of air pressure or vacuum must be considered.