Modified imaging condition for reverse time migration based on reduction of modelling timeShow others and affiliations
2018 (English)In: Exploration Geophysics, ISSN 0812-3985, E-ISSN 1834-7533, Vol. 49, no 4, p. 494-505Article in journal (Refereed) Published
Abstract [en]
Reverse time migration (RTM) is considered as a high-end imaging algorithm due to its ability to image geologically complex environments. However, this algorithm suffers from very high computational costs and low-frequency artefacts. The former drawback is the result of the intensive computations and huge memory allocation involved in RTM. Wave propagation modelling, as a kernel of RTM, demands intensive computations, and conventional imaging conditions are associated with huge memory allocation. In this paper, a modification of imaging condition is proposed that improves the efficiency of RTM as a reduction of computational cost, memory (RAM) allocation and low-frequency artefacts. The proposed imaging condition is similar to the conventional imaging condition but with the reduction of modelling time to near half the maximum time of recording. As the main idea of the proposed imaging condition, the impact of wave propagation modelling time is investigated on the quality of RTM and illumination of reflectors. The performance of the proposed method is considered using two synthetic models (SEG/EAGE and BP) and a real dataset from an Iranian oilfield in the south of Iran. Results showed that the new imaging condition can properly image the reflectors and enhance the efficiency of RTM. By using the proposed imaging condition, we achieved similar to 25% increase in CPU performance and 50% decrease in the memory allocation. Despite the improvement of the performance, results showed that the proposed imaging condition had no significant effect on the illumination.
Place, publisher, year, edition, pages
CSIRO PUBLISHING , 2018. Vol. 49, no 4, p. 494-505
Keywords [en]
computational performance, imaging condition, pre-stack depth migration, reverse time migration, seismic imaging
National Category
Signal Processing
Identifiers
URN: urn:nbn:se:uu:diva-362837DOI: 10.1071/EG17039ISI: 000441707500006OAI: oai:DiVA.org:uu-362837DiVA, id: diva2:1255516
2018-10-122018-10-122018-10-12Bibliographically approved