Search

Among the existing geophysical methods, the vertical electrical sounding remains a fast and economical way to detect groundwater resources. However, the interpretation of the vertical electrical sounding data often suffers from non‐uniqueness due to the ill‐posed nature of the inverse problem. In recent years, metaheuristic algorithms have been successfully used for solving ill‐conditioned and ill‐posed problems. This work presents a scheme that uses the continuous ant colony optimization (ACOR) technique to invert vertical electrical sounding data. The ACOR is a global search algorithm that explores and finds the globally optimal solution over a search space by mimicking the behaviour of biological ants. The development of this algorithm was due to the requirement to interpret a set of vertical electrical sounding collected at the region of Hassi R'Mel (Algerian Sahara). The area has a particular geological/geoelectrical structure, which renders the interpretation of vertical electrical sounding challenging as standard inversion approaches tend to fail to recover a reliable resistivity model. The ACOR algorithm was initially tested with synthetic data from models simulating the geological/hydrogeological structure of the studied area. The results verified the robustness and stability of the ACOR algorithm even in the presence of a high level of noise. Furthermore, the tests indicated that the ACOR algorithm performed better when compared to other inversion techniques for this particular geoelectrical structure. Five vertical electrical sounding profiles using a Schlumberger array collected in the region of Hassi R'Mel were inverted using the ACOR algorithm. The models confirmed the presence of the two central aquifer systems and showed the geometry of the aquifer with the most favourable conditions for water accumulations.

This paper describes a seismic object detection method using supervised neural networks to combine seismic attributes and transform them into a single ‘object probability’ attribute. Unlike other widely used methods, the approach incorporates interpreter's knowledge into the well-known process of combination of multiple attributes. In fact, the interpreter decides the anomaly to be addressed and picks examples of anomalies for the neural networks. The approach completely relies on the interpreter to select the input attributes. However, the limitation that this approach has over other approaches is that it does not incorporate machine intelligence to validate attribute selection. Thus, the present study attempts to overcome this limitation and uses neural networks in the process. The integration of a neural network has played a key role in determining the type and number of attributes used in the prediction and, thus, gives the approach more reliability and confidence. Furthermore, with the help of the neural network, an appropriate group of attributes could be successfully determined and they could be combined into one object probability attribute that made it possible to clearly localize and delineate three bright spots associated with shallow gas in the Upper Pliocene-Pleistocene off the Dutch coast.

It is well‐known that when inverting two‐dimensional (2D) electrical resistivity data, a major source of errors is the presence of noise and in particular noise spikes. The popular median filter is often applied to the removal of single spikes. However, when the signal is highly corrupted with successive spikes, the median filter performance is poor. This paper deals with the use of the signal dependent rank‐order mean filter for the detection and removal of noise spikes from highly corrupted 2D electrical resistivity imaging data. In addition to its computational simplicity, this filter is shown to be extremely robust, even in the presence of very strong noise, especially when it is applied recursively. The signal dependent rank‐order mean filter was tested on 2D synthetic resistivity data contaminated by near‐surface inhomogeneities and the results confirmed efficient removal of the disturbances normally associated with near‐surface inhomogeneities. The signal dependent rank‐order mean filter was also applied to field data and demonstrated its ability to significantly improve the accuracy of the inversion process and to produce good visual results in the inverted electrical sections.