Time for a reality checkPhoto: Terje Solbakk

Time for a reality check

Estimating hydrocarbon volumes before drilling is fraught with uncertainty. After mapping 242 discoveries across the Norwegian Continental Shelf, Edmundson and her colleagues have found a novel way to change this.

So, imagine that your hydrocarbon prospect is almost drill-ready; the structural closure is identified, it has been populated with sound, petrological data. The net-to-gross, water saturation, migration story, porosity, and seal properties are all accounted for, this looks promising.

The only thing missing is the hydrocarbon volume estimation. This estimation is of course always tricky. A fill-to-spill situation would be nice, but can we assume this is reasonable?

Here, lessons can be learned from already proven discoveries on the Norwegian Continental Shelf (NCS). Isabel Edmundson and her colleagues at Wintershall Dea and the University of Bergen identified this challenge and dived into it. The results were presented at the Nordic Geological Winter Meeting in Oslo in January this year.

Their aim was to reduce the uncertainty associated with pre-drill volume estimation by better constraining the hydrocarbon column height using empirical data.

Ignoring the importance of the hydrocarbon column height during volume calculations may have significant repercussions. Overestimation of hydrocarbon volumes will of course increase the net-value of a prospect, which may favour the prospect in a drill-decision over alternative, more promising candidates.

In addition, an exploration well may cost tens of millions of dollars. An over-optimistic and unrealistic view on a prospect could result in significant financial loss. The accuracy of your volume estimation is therefore critical.

With seismic data available internally through Wintershall Dea and well data sourced from the Norwegian Petroleum Directorate (NPD), Lars Frette, Emilie Kavli, Sean Mackie and Isabel Edmundson mapped out 242(!) hydrocarbon discoveries on the NCS, down from the Barents Sea (38), through the Norwegian Sea (81) and into the North Sea (123).

Three main parameters were defined from this mapping: depth to top of structure, trap height and hydrocarbon column height. The last two parameters are combined to plot as a trapfill ratio.

The results gave new insights into what to expect from a structural trap in terms of filling degree. From the general patterns that could be determined, one observation came clear: the hydrocarbon explorationist sweet dream of a fill-to-spill discovery in several-hundred-meter high traps is unlikely.

Depth to top of structure given with a, hydrocarbon column height shown with b and trap height with c. Figure from © Edmundson et al. (in preparation), re-printed with kind permission.

Results show two very clear trends. For a discovery with a given trap height, the probability of recording 100% fill increases when the overburden thickness increases. Equally, when the trap height increases for a given overburden thickness, the probability of a 100% trap fill discovery decreases.

This means that the burial depth of your prospect and geometry of the trap can give you a strong indication on what trap fill to expect. Exceptions exist, but there are few.

It is in the nature of an explorationist to be optimistic, but this optimism should be anchored in proper assessments of the available data. When all other data is carefully evaluated in such a way, why should the hydrocarbon column not be treated similarly?

Isabel Edmundson underlined that this numerical approach used in her study should not replace detailed geological evaluation of the prospect and trap specific geology. Rather, it should be integrated.

The NPD has reported similar findings, when comparing pre-drill estimates with discovered oil resources, in their Resource evaluation report of 2018.

The NPD analysis shows that for the period 2007-16 about 58 % of the oil discoveries fell within the uncertainty range in the pre-drill estimate. About 6 % were above and 36 % below this range. The companies overestimated resource expectations by an average factor of 1.4.

The work by Edmundson and her colleagues is currently under review for publication. A pre-print version is available to view here.

Text: TERJE SOLBAKK

X