【摘要】 受限于现有三维地震资料的分辨能力,薄储层、小断裂等小尺度地质目标识别难题始终影响着岩性油气藏勘探开发的精度.基于高分辨地震分频技术的谱反演处理技术为提高薄层识别与预测精度提供了一条新的思路和方法.在原始叠后地震数据基础上开展谱反演处理,可获得全频带地震反射系数序列数据.该数据具有很高的分辨率,完全可以用于微小级别地质体及地质体内部细节的描述与形态刻画.本文对谱反演处理技术的基本原理和实现方法进行了简要阐述,并将其应用到实际生产资料中.实例应用表明,谱反演处理获得的反射系数数据成果分辨率非常高,可清晰识别目标薄储层反射特征及各类层序变化边界,有较强的实用和应用价值.更多还原

【Abstract】 Along with the exploration process of stratigraphic reservoir,the geologic problem of thin-bed has become more and more prominent,and also urgent to get solved.Since most of the technique solutions rely on the seismic data,the final results are always affected by its resolution limitation.The small-scale geological targets including thin-bed and micro-faults are hard to be exactly resolvable using current seismic data.It is necessary to find out some way.The seismic decomposition technique has been considered as an effective way to improve the thin-bed resolvablity on the basis of tuning effects.With the improvement of decomposition processing algorithms,the spectral inversion technique has been introduced into the field by Portniaguine and Castagna,et al.The process differs from other inversions in that it needs no preset geological model and well data,and it is driven by geological rather than mathematical assumptions.The original bandwidth-limited seismic data is set as input,and the output is seismic sparse reflectivity series data,while the wavelet has been removed within the process.The output reflectivity series data has superior resolution.It has been highly recommended to be used in interpretation and prediction of small-scale geological targets.In the paper,the theory and method of spectral inversion technique is briefly introduced,and several field examples have been discussed.Results show that the output reflectivity series data has very high resolution and contains much more geological information than the original input seismic data,so that more reflection detail can be picked up and interpreted geologically.In the examples,the target thin reservoir has been clearly identified,and stratigraphic sequences have exact inter-relationships and obvious changing styles in the reflectivity series data.We suggest apply this technique in characterizing various small-scale targets and looking into the details,such as thin reservoir,small/micro faults,shale gas reservoir,etc.更多还原