The analysis of high resolution seismic data for shallow gas has made significant advances in recent years but there still remain major limitations in the ability to positively identify gas, let alone be able to accurately estimate concentration or pressure. A number of features which are commonly related to the existance of gas have been identified and when two or more such features are present this is commonly taken as being a positive gas indication and a probability of gas presence is often estimated Such probability estimates are largely arbitrary (often based primarily on lateral amplitude variations) and the accuracy of such estimates is dependent on the skill of the analyst and knowledge of local geology. This paper discusses a series of techniques which should lead to more reliable estimates.
There are several ways in which gas identification could be unproved given our current knowledge more careful processing of site survey data including better amplitude control in the shallow section, detailed velocity analyses in the vicinity of well locations and migration of key lines, use of extra analysis techniques such as residual wavefield and forward modelling for Amplitude Versus Offset, use of numerical analysis rather than qualitative estimation
Implementation of some or all of the above techniques enables a better estimate of the probability of gas presence to be made but there still remains uncertainty especially in areas of irregular subsurface relief It is proposed that in order to ‘calibrate’ gas probability estimates more research is needed relating known occurrence of shallow gas to the presence of gas related attributes in seismic data.
Significant advances have been made in recent years In the basic evaluation of marine sites for the presence of gas In the top-hole sections of wells However, there are still limitations to the quantitative analysis of the volume of gas present and even to the certainty of gas charge In this context ‘quantitative gas hazard analysis’ refers to the use of numerical methods to interpret the presence of gas, define its probability and measure its areal extent and thickness
A number of seismic gas indicators have been identified These include amplitude anomalies, flat spots, phase shifts and frequency absorption A positive indication of gas is assumed when two such indicators are observed together and an estimate of gas probability is made Such estimates are often largely arbitrary and are very dependent on the skill of the analyst and the knowledge of local geology.
In order to improve the identification more effort in computer processing of digital data and greater use of numerical measurements are required. In particular the following points should be considered more careful processing of site surveys with special attention to amplitude control in the shallow section, detailed velocity analyses in the vicinity of the proposed well, migration of key lines, use of extra analytical processes such as residual wavefield and Amplitude Versus Offset, forward modelling in instances where the nature of a possible gas indicator is uncertain and the use of numerical techniques (e g calibrated colour amplitude) rather than a qualitative estimate.