From the archive: Cleaning up an oil slick

The 2006 Israeli offensive in Lebanon led to an enormous oil slick, and prompted The Engineer to take a look at the role technology can play in cleaning up oil spills. By Niall Firth

Although rightly overshadowed by the ensuing humanitarian crisis, one of the lesser-reported effects of the Israeli offensive in Lebanon earlier this year can only be fully appreciated by air. Lapping at the shore, an insidious black scar is creeping along this beautiful stretch of coastline which only a few months earlier had been packed with tourists.

An air strike on the Jiyyeh power station, just over 20km south of Beirut, destroyed its oil tanks resulting in an estimated 35,000 tonnes of viscous oil seeping into the Mediterranean, making it Lebanon’s worst-ever environmental disaster. Threatening local sea life, including turtles, and shattering the already fragile tourism industry, it is estimated that the slick — which has already reached the Syrian coast — may eventually cost more than $100m (£53m) to clean up.

While the legal repercussions of the disaster are ongoing — any funding towards the clean-up process will form part of post-war reparations — it will eventually fall to specialist oil spill engineers to begin reigning in the deadly black ‘gloop’.

As one of the world’s largest oil spill responders, Southampton’s Oil Spill Response (OSR) is likely to be involved in tackling the slick. OSR is a not-for-profit organisation owned by the oil industry, with its 32 members — including major oil companies such as BP and Shell — paying a subscription to be affiliated.

‘There are thousands of different oil types,’ said Dave Salt, OSR’s technical director who oversees the complicated and varied challenges of responding to the world’s worst oil spills. ‘Every one is different, and they all behave differently. We need to know what it is, where it is and where it’s going. Then we can decide what we can do about it. Something like gasoline is a safety problem, while some lighter ones evaporate and disperse quite readily, so that in the teeth of a Shetland gale it is difficult to find evidence of an 85,000-tonne spill. At the other end of the spectrum is a crude oil that is viscous and with a high wax content.’

To decide on how best to tackle the slick, OSR uses software models that analyse databases detailing how different oils react in a variety of weather conditions. While oil moves at nearly 100 per cent of the sea’s current speed, the wind only has about a three per cent influence on its likely future movements. The software looks at historical wind and current data to predict the likely trajectory of the spill. Once the oil type is known — based on information provided by the relevant parties — and the probable effects of the weather and sea temperature on it have been modelled, then a strategy can begin to take shape.

The next step is surveillance, an area which has seen the most technological development in recent years, according to Salt. Remote sensing cameras attached to specially equipped aircraft that fly at little more than 100m measure the infrared image of an oil spill. The black oil absorbs radiated heat and so shows up as warmer than the surrounding water. This means that the thicker the slick, the hotter it becomes and so the stronger the image.

Ultraviolet cameras are also used because when oil is spilt, the wavelets on the sea’s surface are suppressed and so the UV light is reflected differently. The two forms of sensing images are then overlaid to give a more complete picture of the scale of the task ahead, and this information is then fed back into the modelling tool at OSR’s base.

To track spills and to act as a deterrent the government also has its own remote sensing tools. At about 1000m above sea level, aircraft fitted with Side-looking Airborne Radar (SLAR) systems scan vast swathes of the ocean’s surface with tracks extending up to 10 miles either side of the aircraft being scanned for the evidence of spills. The final level of surveillance is to use data from satellites such as LandSat which — while useful for mapping a ‘before’ and ‘after’ image of a spill — are often not current enough to be of any real tactical and strategic use to responders such as OSR, according to Salt.

If the oil is potentially explosive there is nothing to do but to put an exclusion zone around the area and wait for it to disperse. However, for crude oil spills dispersants are applied. The science of dispersants is exact, as the spray nozzles must deploy them in 500-700 micron droplets so they can interact with the oil, break it up and allow it to enter the water column. Dispersants break down the oil into small droplets and then let the ocean currents distribute it over a wide area, where it is gradually bio-reduced.

Although the chemical compound can be applied from the side of a ship, OSR also uses helicopters that carry spray buckets and, most commonly, aircraft that can cover large areas quickly and easily. OSR owns two Hercules aeroplanes, one bought from the US, the other (called Nimbus) specially developed by OSR. Nimbus is a quickly deployable system that can be flown to any part of the world and can carry up to 18 tonnes of dispersant. For smaller spills a converted Cessna 406, developed with remote sensing specialists Atlantic Reconnaissance in Coventry, can carry one tonne of dispersant.

‘If the oil type is right, dispersant is my preferred option,’ said Salt. ‘Aircraft like the Hercules, flying at 161mph and 100ft, can cover 50 acres per minute of sea area. Nothing can beat that. Other technologies like booms and skimmers are far slower.’

In the Prestige disaster in 2002 and the Erika spill in 2000 an extremely viscous oil was so churned up by the sea that it emulsified in the water, creating what Salt described as the ‘chocolate mousse effect’. When the oil is like this there is nothing for it but to resort to ‘skimming’ the oil from the ocean’s surface.

Currently there are four main types of skimmers, the simplest of which are weir skimmers. As oil floats on top of the water, a simple cavity in the centre of the skimmer siphons it off to be pumped into storage. However, a great deal of water also comes with the oil making it an extremely inefficient design.

Far more effective is the oleophilic or disc skimmer, which Salt said was up to 90 per cent efficient. The oil sticks to a rotating disc made from aluminium or plastic, but this system is quite limited in the types of oil spill for which it can be used. If the oil is too thin then it will not stick to the disc.

Next in line are vacuum skimmers, which were used during the Sea Empress oil spill off the coast of Wales in 1996. These devices literally suck the oil from the water’s surface but they are easily clogged with debris and are not very efficient when the sea is rough.

Finally there are mechanical skimmers — ideal for retrieving heavy oils — which use metal tooth discs, grab buckets and drum separators to remove the oil then pump it into a tank.

Despite the efficacy of these skimming techniques, the retrieval of oil from the surface of the ocean is laborious, time-consuming and ultimately extremely expensive. Recent technological developments have led to interchangeable systems which can change the way they operate depending upon the weather conditions, as well as more advanced systems that are fitted with thrusters to allow them to operate further from the ship. But skimming remains a very labour-intensive process. ‘It is not that technologically advanced at the moment,’ admitted Salt.

However, a couple of EU-funded projects are underway that could bring a new level of sophistication to oil spill response efforts.

The Ocean Sea Harvester research project, due to finish by the end of next year, is developing an innovative ship design for tackling oil spills. A trimaran, the Sea Harvester will be equipped with both weir and brush skimmers packed away to protect them from the weather until the vessel reaches the scene of the slick. When it arrives there it will be able to open up the two carriage arms and deploy a variety of skimming tools, dependent on the weather, using a conveyor-belt brush skimmer for viscous oils and the weir skimmer for thinner ones.

The ship will also be equipped with containment booms that can be deployed autonomously to enhance the available skimming area and have two remote-controlled skiffs fitted with toxicity and explosives sensors, to determine how volatile the spill might be. There is also the likelihood that the Oil Sea Harvester could store dispersant products on-board, if the oil was likely to respond to them.

The Elimination Units for Marine Oil Pollution (EU-MOPS) project, on the other hand, is looking to revolutionise the way in which oil spills are dealt with, by automating the entire process using a swarm of intelligent robots.

Prof Harilaos Psaraftis from the National Technical University of Athens’school of naval architecture and marine engineering is EU-MOPS’ leader. Now halfway through the three-year project to automate the entire oil-cleaning process, a workable design is beginning to take shape.

The project’s 13 partners include Oxford University, which is developing the propulsion design, and the University of Strathclyde which is looking at the robot’s overall design and hydrodynamics. With their other European partners they plan to unleash a swarm of autonomous oil-cleaning robots on oil slicks that are able to work together as a team to deal with the spill.

Each robot will be no larger than a standard shipping container, with a number of smaller robots for smaller spills. Likely to be powered by a combination of diesel and electric fuel cell motors, the robots could be deployed as part of a rapid response to an oil disaster in swarms of anything up to 50 robots at once.

The key to the technology is that a group of the robots could operate as a swarm. To navigate itself each robot will need to be equipped with sensors that include those for positioning, collision detection and avoidance as well as navigation and communication systems, including GPS. Above and beyond these systems the robots will also come armed with sensors that can detect the presence of oil and investigate its temperature and thickness allowing them to concentrate on operating on the thickest area of the slick.

Researchers at the Fraunhofer Institute in Stuttgart are developing the software that will provide much of the robots’ artificial intelligence, but Psaraftis warned of expectations being too high.

‘Of course they will be deployed with some degree of autonomy but it is not as though you will be able to just push a button on the robot and then go off fishing,’ he said. ‘It is likely that they will still need some degree of operation from the ship.’

Gliding through the water with a catamaran hull the robots will use a combination of feedback from their sensors and fuzzy logic algorithms to determine where to begin working on the oil. However, it is as yet undecided whether the robots will use weir or disc skimming technology or even something like a specially designed oil brush to remove the oil from the water.

Each robot would also be fitted with internal sensors to monitor how much oil it has taken on-board. When it is full the sensor would trigger a command for the robot to return to the mother ship and have its oil pumped out into a storage container on-board. With the largest device capable of operating autonomously for as long as 24 hours, and able to travel at around five knots it is possible that a far larger amount of oil could be dealt with than by traditional skimming techniques.

Engineering consultant BMT has been heavily involved in analysis of the EU-MOPS project. The firm has been running computer simulations based upon the robots’ predicted artificial intelligence and sensor outputs overlaid with weathering data to see how effective the system could be.

According to Tony Morrall, a research manager at BMT, the main application for the robots is likely to be in shallower, more stable waters, particularly in coastal areas, where they could begin operating from the coast-side to prevent as much oil as possible from reaching the shore.

‘One of the problems is that present systems are usually no more than 10 per cent efficient — which really isn’t that much,’ said Morrall. ‘The more oil that can be stopped washing ashore and ruining the environment the better and this project has real potential to even replace dispersant in some cases.’

One major benefit of the project, according to Psaraftis, would be that it could greatly reduce the overall labour costs involved in oil spill cleaning operations, essential when the projected cost of the Lebanese spill is considered. The robots will also be able to go into areas where it is not considered safe to send people, he said.

In Salt’s view, however, this is no time for the industry to rest on its laurels as the challenges for oil spill responders continue to change, just as rapidly as the oil industry itself evolves.

‘Oil in ice is a major issue in the future as that is where it is now being found,’ he said. ‘There are plenty of challenges for clean-up operations when you’re dealing with frozen seas, that’s for sure.’