History Matching CSG Production in the Surat Basin

Abstract

Abstract In 2009, QGC (a BG Group business) first planned to produce coal seam gas (CSG) in the Surat Basin as feedstock for a new Liquefied Natural Gas (LNG) plant. Subsurface models and associated field development plans were generated to underpin the investment case for the Queensland Curtis LNG (QCLNG) Project. This paper discusses history matching experience from QGC's Surat Basin reservoir simulation models, the challenges involved, and how these challenges have been overcome. Similar to conventional reservoirs, full-field numerical simulation is necessary to accurately account for well interference and broader reservoir connectivity. Simulationalso fully integrates modelled 3D static property variations and honours the physics of multi-phase flow through porous media. CSG reservoirs possess unique characteristics which differ from conventional reservoirs and tend to increase the challenge of history matching. Some of these challenges include the physics of diffusive flow from the matrix into fractures that combines with relative permeability to create a more complex multi-phase flow problem. Another challenge is the requirement to de-water the coal before gas production commences and the associated stress dependency of coal properties. At the same time, the nature of CSG developments makes conventional history-matching approaches impractical given the large number of wells and the need for quick turnaround and fast-cycle decision making, especially during development ramp-up. These demands are set against a lean business environment where cost efficiency is paramount. The properties of Walloon Subgroup (WSG) coals in the Surat Basin are unique compared to other CSG basin plays worldwide. The WSG is characterised by low to moderate coal rank, highly interbedded seams distributed through an extremely low permeability inter-burden, and highly variable coal permeability throughout the basin. Some of these differentiating properties of the WSG make history matching these reservoirs very challenging. This paper presents guidelines for overcoming these challenges and history-matching CSG production in a dynamic simulation model of the Surat basin WSG. Handling of uncertainty is discussed to consider a range of possible history-matches for 345 wells with 9 years of production history. The history matching guidelines that have been developed are enabling a faster turnaround of model predictions, capturing key uncertainty parameters and informing field development decision-making.