地球統計学手法によるメキシコ，チコンテペック堆積盆油田群のタービダイト油層評価

Abstract

PEMEX and TRC/JNOC commenced the collaborative research project “Geo-statistical modeling of Chicontepec reservoirs: Agua Fria, Coapechaca and Tajin areas” in 1999 for the purpose of the technology development to integrate geological, seismic and reservoir engine-ering information for better field development and reservoir management. This paper presents the procedure of pixel-based hierarchical geostatistical modeling of sandstone distributions in turbidite deposits of the Agua Fria, Coapechaca and Tajin areas in the Chicontepec basin.The Paleogene Chicontepec Basin is located between the Sierra Madre Oriental and the Golden Lane Platform, extending northwest to southeast along the Gulf Coast in Mexico. This basin consists of submarine fan turbidite sandstones derived from the Sierra Madre Oriental. First, geological/sedimentological analyses were conducted based on the concept of sequence stratigraphy for the purpose of stratigraphic division and facies distribution estimation. Four turbidite facies were recognized based on well-log patterns and core descriptions. These estimated facies distribution maps are compared with the results of seismic facies analyses. Second, an architectural framework for each sequence boundary (SB) and sub-internal surfaces of trangressive surface (TS) and minor condensed section (MCS) was modeled by integration of sequence-stratigraphic well-marker data and depth-converted seismic time horizons. The non-parametric Alternating Conditional Expectation (ACE) algorithm was used to develop correlation functions between nine kinds of seismic attribute data and total porosity (PHIT). Finally, three-dimensional distributions of PHIT, shale volume contents (Vsh) and facies distribution were modeled using various geostatistical techniques. A special technique is presented for the integration of seismic and well data, which also simultaneously honors sub-interval-layering information. Comparison of geostatistical facies model with deterministic geological models shows general agreement between the two. However, the integrated geostatistical method generates much more detailed information and is readily applicable to wider reservoir areas.