Improved Curve Fitting Methods for Underdamped Slug Tests

Abstract

Abstract A slug test yielding an oscillating water level response is called “underdamped.” Usually, test data are fitted visually to theoretical solutions. Considering that visual fits are user dependent, this paper provides two best fit user independent methods for analyzing test data. It provides also a velocity graph method that can be used to select only those data that are not influenced by initial dynamic effects such as splashing. Several examples are provided, with emphasis on the utility of the three checks required by ASTM D5785. It is shown that the selected value of the soil modulus or its storativity S does not interfere with the fitting process but does influence the derived k value by about ±30% (the lower the selected S value, the higher the resulting k value). If the elastic S value is confused with a specific yield, the interpretation results do not pass the three checks of ASTM D5785, and, as shown by an example of a few hundred tests, they might yield k values that are wrong by about 300%, with serious consequences for estimates of groundwater velocity and the fate of contaminants.