With an increasing interest for extra heavy oil (EHO) extraction, a number of heavy oil fields will initially be produced using a cold production scheme, while at a later stage thermal production in the form of steam injection can be introduced to enhance the recovery. Such a fullfield conversion from cold to hot production has so far only been performed to a limited extent, hence the experience and understanding of the various challenges with such a task is limited. While much work has been carried out on both cold and hot production of extra heavy oil as stand alone processes, little has been published around the procedure of converting from cold to hot production. This conversion contains a number of pitfalls detrimental to a successful field development. In an EHO cold to hot conversion scheme, several choices will have to be made with respect to well positioning, completion options, timing and scale of the conversion. Investment decisions in surface equipment, handling and cost of water and gas for steam generation are also important considerations. All of these designs and decisions affect the success and profitability of the conversion scheme. One of the basic challenges with modeling cold to hot conversion is the difference in scale required. The thermal simulation of a hot production scheme requires a very fine grid to capture the fine scale processes, while for full field simulation one needs a much coarser grid to be able to obtain a manageable model both in size and simulation time. The high number of producers and injectors required on a fullfield scale adds to the complexity of the problem, making recovery, production profile and optimal number of wells highly uncertain. A simplified translation to fullfield conditions of these parameters from sector models or spreadsheets, can introduce erroneous results and conclusions. To assure the best field performance over its lifetime, these issues need to be considered already in the initial field design. Based on the work performed, several interesting observations have been made regarding how to optimize field performance.

The Griffin Area Development in the Barrow Sub-basin offshore Western Australia consists of three oil fields; Griffin, Chinook and Scindian, tied into the Griffin Venture Floating Production Storage and Offloading (FPSO) facility. First production commenced in January 1994, and the field is currently producing 12,700 m3/d (80,000 stb/d) of oil from up to nine available wells. In addition, associated gas is being piped onshore to provide up to 1.1 × 106 m3/d (40 million scf/d) of gas and 62.8 t/d of LPG. Oil is found in the Lower Cretaceous sediments of the Zeepaard and overlying Birdrong Formations, with the Birdrong Formation at Chinook and Scindian having gas caps. The Zeepaard reservoirs have excellent reservoir properties and are expected to yield a high recovery through six vertical wells. The Birdrong Formation is unconformably deposited above the Zeepaard Formation, and is a lithologically complex mixture of sands with intermixed clays and shales and is generally of much poorer quality than the Zeepaard Formation. The Birdrong Formation has provided some of the main challenges in the development, resulting in the implementation of three sinuous horizontal wells to access the better quality layers of the formation. Permanent downhole gauges have been used to assist in the determination of pressure support both in the Birdrong and Zeepaard Formations. This paper outlines the key factors which shaped the Griffin Area Development Plan, details the implementation of the subsurface aspects of the Phase I development, and presents key aspects of reservoir performance to date.