Geological Modeling Using Geostatistical Inversion: Late Pliocene Deep-Water Turbidite Fields Offshore Northwestern Myanmar

Kim, Dongil (David) ¹; Yang, Su-Yeong ¹; Kim, Jae Woo ²
(1)Daewoo International Corporation, Seoul, Korea, Republic of. (2) Daewoo International Corporation, Yangon, Myanmar.

Building a realistic geological model in early exploration stage is highly desirable to design an effective appraisal program and development plan that can minimize subsurface uncertainties. One of the most common hurdles to geomodelers in this stage is the limited number of wells and resolution difference between conventional seismic and wells, which will impede establishing reliable relationship between the two and lower the confidence in resulting geological models. To overcome these limitations, we adopted geostatistical inversion (GI) technique and constructed geological models for the late Pliocene deep-water turbidite deposits of three gas fields offshore northwestern Myanmar.

GI is the latest inversion technique which uses Markov-Chain Monte Carlo algorithm together with seismic volume, well logs and geostatistics as main input. GI outputs have at least 4 times higher vertical resolution than seismic as well as multiple realizations that all honor well data being a useful tool for quality control and uncertainty analysis.

There were great benefits in building geological models using GI, especially for property modeling. First, reservoir properties guided by GI showed well defined correlation with well logs. Second, multiple realizations of GI output enabled us to control uncertainty in depositional facies distribution. Third, refined geological trends were obtained from GI through compensation of dimmed seismic amplitude affected by shallow geology.

A great improvement of geological models was achieved using GI in modeling sheet-like sandstone reservoirs thicker than 5 meters. However, it was still insufficient to build realistic models for thinner bedded levee-overbank deposits. Based on these models, we were able to reduce subsurface uncertainties, refine reserve estimation, and optimize production well locations.

AAPG Search and Discovery Article #90135©2011 AAPG International Conference and Exhibition, Milan, Italy, 23-26 October 2011.