Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Oil exploration

Every last drop

The price of petrol is going up and new oil discoveries are declining. Can underground fires and hydrocarbon-hungry bacteria keep the oil flowing? Jim Giles finds out.

Field of dreams: new technologies could pull hidden reserves from old wells. Credit: B. ROSS/CORBIS

Cheap oil is on the way out. This has been made abundantly clear in recent months thanks to security problems in Iraq and terrorist attacks in Saudi Arabia — both countries with massive crude-oil reserves. Markets are unsettled and prices at the pump are soaring. Last month the price of crude reached more than US$42 a barrel, its highest for 20 years.

But geologists have known for years that the end of cheap oil is in sight. Since the 1960s, the rate at which new wells have been discovered has been on the wane1. Forty years ago, more than 50 billion barrels of oil could be discovered in a single year. Today, finds of about 10 billion barrels per year are far more common (see graph), many of them from smaller fields. In short, many experts think that all the big gushers — the wells that spew out masses of oil cheaply and easily — have probably been found.


The decline has hit at least one oil company hard. Earlier this year, the Anglo-Dutch oil giant Shell was caught in a media storm after it admitted that its proven reserves — the oil the firm knows it can definitely extract — are 20% less than its previous estimates. Reserve estimates are often optimistic, analysts say, and many wells don't quite produce as much oil as expected. But new wells are usually found, making up the numbers. Analysts suggest that in this case Shell didn't find enough new wells to make up their accounts, drawing attention to the embarassing shortfall.

Shell's misfortunes and the situation in Iraq may change. But these effects are over-lying a trend that cannot be ignored. Oil companies now realize they cannot rely on finding new fields in the long term. Instead, they will need to extract every last drop they can from existing sources.

A slew of speculative techniques, from lighting fires in oil fields to using microbes to help wash out oil reservoirs, are already available to do this — and more are on the way. Some methods have been in development for decades. But only now is the price of oil becoming high enough to make them commercially viable.

“In the United States alone there is 355 billion barrels of oil that is not recoverable using existing techniques,” says Betty Felber, a senior petroleum scientist at the National Energy Technology Laboratory in Pittsburgh, Pennsylvania. This is more than 15 times as much as the proven reserves that we can get at using conventional methods. “More and more people are applying these technologies,” she says.

“We say oil is running out, but 60% of it is left in reservoirs,” adds Malcolm Greaves, a geophysicist at the University of Bath, UK. “What are we going to do? Walk away?”

Greaves is pioneering one radical solution: setting fire to the reservoir. The technique was first trialled, albeit accidentally, in Russian oil fields about 50 years ago. Engineers pumped air into reservoirs in a bid to raise the pressure and force oil out of existing wells, and discovered that the air reacted with and ignited the oil. The combustion, which was limited to a small area of the reservoir, heated the oil, reducing its viscosity and allowing more of it to flow smoothly out of nearby wells.

The method, known as in situ combustion, has since been tested more rigorously, with mixed results. Greaves estimates that about 140 pilot projects were run during the 1980s, when oil prices were also high. Some suffered from ‘blow-backs’ — explosions that travelled back up the well through which the air was injected into the reservoir. In around a third of the experiments, oil flow was not increased as much as anticipated, in part because the injection well was often too far away — typically hundreds of metres — from the producing well.

Fire down below

Sky high: pump prices soar in the United States. Credit: J. GRESS/REUTERS/CORBIS

The technique could be rehabilitated if a large-scale trial set to start this December is successful. Under scrutiny is a variant of in-situ combustion known as toe-to-heel air injection (THAI), developed by Greaves and his colleagues at the University of Bath2. As the wells are being drilled especially for the trial, they are designed to be just a few metres apart — avoiding some of the problems of previous trials. Another new feature of THAI is that the oil is drawn off through a pipe that runs horizontally, rather than vertically, through the reservoir. This means that the fire can move along the pipe, says Greaves, pushing oil out in front of it.

Greaves reckons that the US$30-million trial at the Christina Lake oil field in British Columbia, Canada, could recover around 80% of the billion or so barrels of oil in the field — an ambitious estimate given that no more than 60% of a reservoir's field can normally be extracted. The THAI tests, run by the Canadian company Petrobank, of Calgary, should also prove useful in extracting Christina Lake's more viscous oil, made up of heavier, larger molecules. The combustion will split the oil, allowing the lighter and more valuable components to flow out of the reservoir.

For fields that already contain lighter oil, Egil Sunde has an alternative3. A marine biologist by training, Sunde has worked for Norwegian producer Statoil for around 20 years. “My idea was to use Nature's ways,” he says. Some microbes feed naturally on hydrocarbons. These bacteria are already used to help strip oil from polluted beaches, he points out, so perhaps they could also help pull it out from underground reservoirs.

Sunde is using microbes that both feed on oil and make it less sticky. Oil is difficult to extract from half-empty fields, in part because it clings to the pores in the reservoir rocks. Water can be pumped in to help push out the oil, but eventually it will simply flow over the top of this residue.

Sunde's microbes grow at the interface between oil and water in the rock pores. This helps water molecules grab onto the oil and detach it. “This makes the oil move more easily through the pores,” says Sunde, who for commercial reasons won't reveal the species of bacteria that do the trick best.

Statoil has been testing this idea since 1991 and was sufficiently encouraged to start a commercial-scale trial in 2001. Bacteria grown in Sunde's lab are now being pumped, together with nutrients and oxygen, into reservoirs in the Norne field off the coast of Norway. Sunde says it is too early to judge the results of this larger trial, but hopes that it will eventually increase the amount of recoverable oil — estimated to be around 530 million barrels — by about 5%.

The technique will not work in every oil field — the pores in the chalk reservoirs of many Middle East fields are too small for bacteria to pass through easily, for example. But industry observers are cautiously optimistic about the technique's potential in sandstone reservoirs like those in the North Sea. “This has a big future,” insists Sunde. “It could revitalize thousands of fields.”

So far there only a few techniques such as Sunde's and Greaves' are ready for large-scale trials, but other fledgling techniques could mature soon. “They will become appropriate now that the oil price is high,” says David Hughes, principal reservoir engineer at Reservoir Management Limited in Aberdeen, UK.

High pressure

The UK-based energy company BP, for example, is investigating the chemistry that governs the reaction between water, oil and reservoir rock. Researchers there think that decreasing the salinity of the water could prevent oil drops from becoming trapped in the pores4.

Other researchers think that plugging holes in the rock with plastic will make it easier to build up water pressure with injected fluid. Another way to increase the pressure and force out the oil could be to use gas-producing microbes. Even more speculative ideas are being pursued, such as using microwaves to heat up the oil and lower its viscosity5.

But the enthusiasm needs to be taken with a pinch of salt. Oil companies tend to hush up unsuccessful projects, points out Nigel Brealey, also at Reservoir Management Limited. In the early days of in situ combustion projects, he says, the fire sometimes broke through into other wells and caused explosions. “But it wasn't well reported,” adds Brealey, “People don't talk about failures.” Techniques that look good on paper can also fail to make economic sense in the field. Despite years of work, unconventional methods such as microbial extraction produced some 2.8 million barrels of oil per day in 2003 — just 3.5% of the global total.

That percentage will, however, almost certainly rise as the price of oil increases. The techniques will not make oil cheaper, but they will keep help keep it flowing as new discoveries dry up. The days of easy oil extraction may be over. But while there is money in it, techniques to pull the last drop of the black stuff from the ground are going to be in demand.


  1. Insight: Hydrocarbon Reservoirs Nature 426, 317–363 (2003).

  2. Greaves, M., Saghr, A. M., Xia, T. X., Turta A. T. & Ayasse, C. J. Can. Petroleum Tech. 40, 38–47 (2001).

    Google Scholar 

  3. Torsvik, T., Gilje, E. & Sunde, E. Proc. 5th Int. Conf. Microbial Enhanced Oil Recovery and Related Biotechnology for Solving Environmental Problems (US Department of Commerce, Springfield, VA, 1995).

    Google Scholar 

  4. Webb, K. J., Black, C. J. J. & Al-Ajeel, H. Low Salinity Oil Recovery. Society of Petroleum Engineers elibrary. 81460.

  5. Vermeulen, F. & McGee B. J. Can. Petroleum Tech. 39, 25–29 (2000).

    Article  Google Scholar 

Download references

Author information

Author notes

  1. Jim Giles is a reporter for Nature, based in London

    • Jim Giles

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Giles, J. Every last drop. Nature 429, 694–695 (2004).

Download citation

  • Issue Date:

  • DOI:

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing