To assess well efficiency, hydrologists use pumping activities to initiate hydraulic losses. Evaluating these losses helps engineers design feasible and sustainable pumping schemes. Step-drawdown tests are used to estimate the aquifer characteristics and performance of wells under higher pumping rates with controlled flow. While there are many tools to interpret these step-drawdown test results, most use only one measurement, providing limited information.

The authors of a new paper, “Interpretation of Step-Drawdown Tests with the Differential Evolution Approach” in the Journal of Hydrologic Engineering, use a differential evolution algorithm as an alternative option for analyzing step-drawdown test results, evaluating for aquifer and well loss parameters. Emin Çiftçi and A. Ufuk Sahin employed synthetically generated noise-free data sets, compared performance of the DE algorithm with different competitive population-based algorithms, and used additional testing to mimic real pumping conditions. Learn more about this unique use of a differential evolution search method for interpreting step-drawdown test results at The abstract is below.


A step-drawdown test is a common hydrogeological investigation tool employed for identifying the hydraulic characteristics of an aquifer as well as assessing the efficiency of the pumping conditions. Several graphical and optimization-based solution techniques have been devised for analyzing data sets obtained from step-drawdown tests to retrieve aquifer and well loss parameters. This study aimed to introduce the use of a differential evolution (DE) algorithm as an alternative and practical option for interpretation of step-drawdown tests conducted in confined aquifers. The proposed estimation procedure was tested for a large number of synthetically generated noise-free and noisy data sets for evaluating its estimation performance. The DE search method exhibited superior accuracy with considerably higher convergence speed when compared with other competitive and widely used population-based algorithms. Sensitivity analysis was performed to explore the capability of the method in estimating each investigated variable. The DE algorithm was implemented for analyzing a real field data set as well, and it was able to produce parameter estimation results consistent with those reported in previous studies. As demonstrated in this study, the DE search method can be an eligible algorithm for solving inverse problems in the field of hydrogeology, regarding its accuracy, high convergence speed, robustness, and simplicity in coding.

Read the paper in full in the ASCE Library: