World population growth tracks CO2 atmosphere concentration closely.

We live in interesting times. Just four or five months ago I would have been writing about the huge impetus given to the geoscience community by the historically high oil price. I would be drawing analogies with the early ’80s, when one might have waited a year or more to get a seismic crew on location, and even then the contractor would have problems finding enough experienced staff to run it.

How times change! Nothing new for our industry, of course, although admittedly the global financial meltdown provides a new backdrop for this particular situation.

Can we draw guidance from the industry savants — for example, the financial analysts who have their fingers on the pulse of the oil markets and whose rigorous analyses allow them to make authoritative, insightful forecasts? It is common practice to use previous history as a means of forecasting future trends, but this can be seriously misleading in the volatile seismic industry. If you have followed some of their forecasting during the last few turbulent months (well-reported in the trade press), you’ll realize just how much at sea some of these people have been. For example, even after the financial crash of September, forecasts were appearing suggesting an increase in the oil price!

I can’t help recalling the massive increase in the oil price in the early ’70s during a Middle East crisis. This was followed by a significant period of financial market doldrums (there were other factors too, such as the dissolution of the post-war monetary system which regulated exchange rates). Some other major events are also posted in the figure.

Price hike

What drove the oil price upwards so steeply in 2007/2008? Energy demand had been rising fast in the Eastern Hemisphere, and production was struggling to meet it. Despite the demand, production was falling in many basins, putting extra pressure on the traditional high-volume producers. The well-known IEA oil production forecasts had been showing production requirements of around 110 MMbbl/day by 2030, and there was a growing perception that, realistically, these rates were unachievable, despite the forecast being based on quite credible population and GDP growth figures. We’re all familiar with the “hockey stick” shape of the global CO2 atmospheric concentration, but the equally sporty shape of the population growth curve has not yet had the same exposure.

Of course, the markets are complex places — they react psychologically as well as to indices and “hard data.” But as the oil price continued to rise, increased trading volumes due to “hedging” took place, driving the price even higher. Eventually this took the price to the point where the global economy could not survive in its present form. Again there were other factors at work, such as excessive risk-taking in the derivatives markets and “toxic” debt. The markets are smart, and an expanded five-year view of the major stock market indices shows quite clearly that they had begun to react by falling, albeit gradually, well before the accelerated crash in September 2008.

Two probable conclusions can be drawn. First, although many savants agree that the known oil reserves are adequate for several decades, a continuing production growth following the historical 1.8% per annum is probably unsustainable. Second, the global economy can not at present survive intact at a greatly elevated energy cost. The likely result of these two issues is a continued unstable oscillation in both economic activity and oil price, although the periodicity is difficult to predict.

Overall, then, it seems sensible to assume a long-term increased requirement for hydrocarbons, becoming more severe over a period of a few decades. Looked at in perspective, exploration and production activities are still strong. Admittedly, the rig count is dropping and is likely to drop more, but it’s still above 2005 levels. Likewise, the oil price at around US $50 takes us back to about the same year. I seem to recall we were having quite an enjoyable time during that period.

Industry options

What options, then, are open to our industry in these changing times? Since “giant” oil discoveries are becoming more and more rare, the most obvious way to increase reserves and to satisfy greater demand is to increase recovery factors from known reservoirs. The public appears not to have woken up to the fact that the average recovery factor from our oil reservoirs is very poor — around 30 to 40% — and even the best ones are less than double these figures. It’s not currently easy to effect much improvement — reservoir engineers talk of 10% enhancements by using combinations of drilling, in-well techniques, geochemistry, seismic resolution, and time-lapse seismic (4-D).

Since infrastructure such as platforms and pipelines is already in place, additional recovery represents potentially very cheap oil. Consider 5% extra from a 0.5 Bbbl field — that’s $625 million even at $50 oil and a (high) $25 lifting cost. This would pay for several 4-D seismic surveys or a life-of-field seismic (LoFS) system, with lots of profit left over.

Before 4-D had become well-established, the first repeat survey would not usually be triggered until an unexpected fast-falling reservoir production event resulted in a panic attack. Subsequent surveys would then be acquired in order to manage the decline of the reservoir as efficiently as possible and to maximize the recovery.

Overall, it was the gradual recognition and acceptance that “Peak Oil” had occurred in the North Sea around the year 2000 which resulted in a massive expansion of the 4-D technology in that basin. Since the technique has become more routine, most new production projects have a 4-D program built into the reservoir management plan, and often the first repeat is recorded before the onset of production surprises since early warnings have much greater leverage on subsequent field management.

Several reservoirs have now had more than five or six such repeats. On some 4-D surveys this led to an almost inadvertent capture of a region which had only been on depletion or injection for a very short time, for example three months. It was surprising to note that pressure and fluid effects could usually be seen despite the short time intervals, and this was one of several motivations for a LoFS installation on the Valhall field in 2003. The 12th seismic survey is now underway using the Valhall LoFS system since once the cables are laid, the deployment of a simple source vessel is relatively easy and cheap, and it can sail much closer to the platforms.

Furthermore, a seabed detector system allows the acquisition of a much more complete seismic data volume, which improves the seismic data cube in many ways. Firstly, instead of recording the data with single azimuth (as with conventional towed-streamer surveys), the azimuth domain becomes fully populated, allowing a huge increase in seismic quality. Currently this is only achievable on towed surveys by utilizing additional vessels. Secondly, having the detector systems on the seabed allows the use of multisensor acquisition, which again allows a better seismic data volume to be achieved.

It became clear from work by Landrø and subsequently by Calvert and others that geometrical repeatability down to just a few feet was needed to reduce the “noise” resulting from the subtraction of two seismic datasets. Thus, over the last three or four years, the impetus in 4-D technology has been on maximizing the repeatability of these surveys with techniques such as steerable streamers and LoFS systems. The full potential of the 4-D method has not yet been fully realized because of problems of repeatability: Towed-streamer technology rarely allows such tight control of sources and detectors, and even with LoFS systems, which intuitively could offer the best repeatability, subtle effects such as tidal and water temperature variations still result in challenges.

However, if the repeatability of surveys can be improved further, the potential for 4-D is immense. Most of the “noise” in a seismic 3-D cube is really just geology such as scatter and multiples, which are difficult to attenuate (or to image or “invert”) using current seismic technology. Low-noise subtraction of repeat surveys will allow us to concentrate on identifying those differences that result from fluid movement, pressure changes, and geomechanical changes.

Conclusions? It seems possible that even in these chaotic times, we can predict increased spending on time-lapse surveys (probably LoFS on new reservoirs) to give vastly improved visibility on the depleting reservoir, resulting in substantially greater recovery factors.