复杂赋存形态水合物储层地震岩石物理建模与应用

天然气水合物由于其复杂的微观分布特征，现有考虑单一赋存形态的岩石物理模型无法对其进行精确表征.此外，水合物饱和度和赋存形态的耦合影响，导致水合物地层弹性响应机理不明和水合物饱和度预测精度偏低.本文依据不同赋存形态水合物与沉积物颗粒的作用机制，结合Voigt-Reuss-Hill平均、SCA-DEM模型、广义有效介质模型、Wood公式、Gassmann方程、Hill方程和Backus平均等建立了同时含六种赋存形态的水合物地层岩石物理模型.数值模拟研究发现，不同赋存形态水合物地层弹性参数对水合物饱和度敏感性不同，可以为水合物赋存形态识别和饱和度预测提供理论依据.实验测试结果表明，所建模型可以很好地捕捉不同实验方法测量的声学响应并揭示水合物微观赋存形态及其演化机制.Mackenzie三角洲Mallik 2L-38井的应用结果显示，相较于单一赋存形态，考虑多重赋存形态建模可以有效提高水合物饱和度预测精度.最后，利用Nankai海槽AT1-MC井和Hikurangi陆缘U1518B井的纵横波速度和密度数据开展了水合物饱和度和多重赋存形态百分比同步量化表征，结果表明预测的水合物饱和度与电阻率或核磁共振测井解释及岩心分析结果比较吻合，反演的水合物赋存形态也合理揭示了研究区域的水合物形成和分布规律.

Seismic rock-physics modeling of the complex-morphology hydrate reservoirs and applications

Due to the complex hydrate morphologies, many of the existing models considering the single morphology cannot accurately describe the hydrate reservoirs. Moreover, the coupled effect of hydrate saturation and morphology leads to the ambiguity of elastic responses and thus the inaccuracy of hydrate saturation estimate. Concerning the manners of different types of hydrates interplaying with sediment grains, we establish the rock physics model honoring six hydrate morphologies by combining Voigt-Reuss-Hill average, SCA-DEM theory, generalized effective medium model, Wood equation, Gassmann equation, Hill's equation and Backus average. Numerical simulation results suggest that elastic properties of the host sediments containing hydrates of varying morphologies have different sensitivities to hydrate saturation, which provides a theoretical basis for hydrate morphology detection and saturation quantification. The validation of the developed modeling strategy with experimental data demonstrates that our modeling method could not only effectively capture acoustic responses of artificial hydrate-bearing samples forming with different methods, but also reveal hydrate occurrence mechanism and its evolution pattern. The application of Mallik 2L-38 well at the Meckenzie delta shows that rock-physics modeling allowing for multiple morphologies instead of single mode could improve the accuracy of hydrate saturation estimate. Finally, the hydrate saturation and the fractions of multiple morphologies are simultaneously quantified from the sonic and density logs acquired at the well AT1-MC, Nankai Trough and the Hole U1518B, Hikurangi margin. The inversion results exhibit that hydrate saturation estimates are in good agreement with those from resistivity and/or nuclear-magnetic-resonance log, as well as core data. The predicted proportions of multiple hydrate morphologies could provide reasonable interpretations of hydrate formation and distribution for the studied areas.