基于分频融合地震属性的曲流带预测与点坝识别:以渤海湾盆地埕岛油田馆陶组为例*
Prediction of meandering belt and point-bar recognition based on spectral-decomposed and fused seismic attributes: A case study of the Guantao Formation, Chengdao Oilfield,Bohai Bay Basin

埕岛油田馆陶组发育典型的高弯度曲流河沉积。通过分频融合地震属性提取与优选、储层构型模式指导、动态响应特征约束等手段,对研究区海上稀井网条件下曲流带和点坝进行精细刻画。研究表明: (1)分频融合地震属性的储层预测方法提高了曲流带的预测精度,振幅类属性与砂体厚度相关性最好,以最大峰值振幅为最佳,相比原始地震属性,分频融合得到的最大峰值振幅属性更好地刻画了曲流带砂体边界与砂体厚度分布;(2)复合正韵律、砂体厚度大、紧邻废弃河道分布为点坝三大重要识别标志,与此相对应,点坝表现为高最大峰值振幅、高反演属性的地震响应特征,而废弃河道呈弯月状低振幅属性、“顶平底凸”的低反演属性特征;(3)在应用地震属性确定点坝位置的基础上,通过经验公式推算点坝跨度,可认知研究区点坝的规模,有效指导地下点坝识别,动态信息的约束也为点坝识别提供了依据。井震结合的曲流带及点坝识别方法可为相似油田构型分析提供借鉴。
Sedimentary environment of the Guantao Formation in Chengdao Oilfield is meandering river. With the methods including the extraction and optimization of spectral-decomposed and fused seismic attributes,the guidance of reservoir architecture model,and dynamic response characteristics constraints,meandering belts and point bars under the condition of offshore sparse well network are characterized in detail. The result shows that: (1)the reservoir prediction method of spectral-decomposed seismic attributes improves the prediction accuracy of the meandering belt. The correlation between amplitude properties and sand thickness is the best,especially the maximum peak amplitude. Compared to the original seismic attributes,the maximum peak amplitude derived from the spectral-decomposed seismic attributes can characterize the boundary and thickness of the meandering sand-bodies much better. (2)The three important symbols of point bar are composite positive rhythms,thick sand-bodies,and being adjacent to abandoned channels. Corresponding to these features,point bars show the characteristics of seismic response with high maximum peak amplitude and high inversion properties,while abandoned channels show low amplitude attributes and low inversion properties with the characteristics of flat top and convex bottom. (3)Based on location prediction of the point bar by using seismic attributes,the span of the point bar is calculated through empirical formula,the scale of point bar is recognized,and the identification of subsurface point bar is effectively guided. The constraint of dynamic data also provides a basis for point bar identification. The methods proposed in this paper can provide implications to the oilfield with similar sedimentary characteristics and data condition