Comments on: View the new participants in OnePetro http://www.spe.org/notes/2011/01/view-the-new-participants-in-onepetro/ Updates for SPE members worldwide Thu, 06 Sep 2012 12:40:15 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Patrick Maylor http://www.spe.org/notes/2011/01/view-the-new-participants-in-onepetro/#comment-91 Patrick Maylor Fri, 01 Apr 2011 08:41:47 +0000 http://www.spe.org/notes/?p=2554#comment-91 Firstly congratulations on a comprehensive and clear set of guidelines. Although a late comment, I note that on Page 51 the generalised guidelines give clear and hard guideline on reserves categorisation of results from simulation models based on the availability of performance data. This is of great concern when it comes to the guideline on proved reserves. Based on many years of experience building, history matching and reviewing of simulation model results through time as a reserves manager and an engineer in many different assets, it is very clear to me that in general simulation models however well history matched can be equally optimistic as they can be pessimistic even in fields which are beyond 25% recovery factor. In addition the optimism or pessimism within the model tends to reflect the core competencies and drivers involved. Engineers tend to use a range of estimates of parameters based on the available data, which in most cases means that they can get history matches using combinations of assumptions that revolve around the most likely outcome. As covered in many papers, history matches are non unique and are subject to the accuracy of the input data which in most cases is of a much poorer quality than most will admit. Using what generally becomes a most likely outcome model in proved means that it is equally likely that the forecast of recovery will be exceeded than not, hence making it a proved plus probable outcome and not a proved. I would seriously suggest the removal of the third paragraph applying to the proved category in the guidelines and replace with guidance on the volume of data will assist in reduction in the uncertainty range but also recognising that significant uncertainty can and most likely will remain in late field life. Firstly congratulations on a comprehensive and clear set of guidelines. Although a late comment, I note that on Page 51 the generalised guidelines give clear and hard guideline on reserves categorisation of results from simulation models based on the availability of performance data. This is of great concern when it comes to the guideline on proved reserves. Based on many years of experience building, history matching and reviewing of simulation model results through time as a reserves manager and an engineer in many different assets, it is very clear to me that in general simulation models however well history matched can be equally optimistic as they can be pessimistic even in fields which are beyond 25% recovery factor. In addition the optimism or pessimism within the model tends to reflect the core competencies and drivers involved.

Engineers tend to use a range of estimates of parameters based on the available data, which in most cases means that they can get history matches using combinations of assumptions that revolve around the most likely outcome. As covered in many papers, history matches are non unique and are subject to the accuracy of the input data which in most cases is of a much poorer quality than most will admit. Using what generally becomes a most likely outcome model in proved means that it is equally likely that the forecast of recovery will be exceeded than not, hence making it a proved plus probable outcome and not a proved. I would seriously suggest the removal of the third paragraph applying to the proved category in the guidelines and replace with guidance on the volume of data will assist in reduction in the uncertainty range but also recognising that significant uncertainty can and most likely will remain in late field life.

]]>