Oil fields out at sea are vast in size, sometimes more than 100 miles long, albeit they are extremely thin in comparison and are generally made up of the following 3 layers:

Layer 1 - a gas layer on top.

Layer 2 - the oil layer in the middle.

Layer 3 - the water layer at bottom.

In order for an oil company to work efficiently and effectively they require vital statistics on a regular basis. Once an oil field has been discovered work is quickly underway to ascertain, not only how much accessible oil that field possesses but to additionally gauge how much water and gas is also contained within it. If an oil field is considered to hold a sufficient amount of accessible oil then drilling can be performed to enable its extraction. The oil companies require to be kept up to date with all the necessary variables concerning that field. They need to know what the various hydrocarbon characteristics are within the reservoir, such as the water, gas and black oil pressures. As the oil is being extracted over time they also need to know how much reserves of oil are left in the feild. In order to obtain optimum production levels precise evaluations of drilling co-ordinates are of the utmost importance. A lot of money can be wasted by drilling in a less advantageous position within a field. Data produced by seismic radar of the sea-bed are produced and when a potential oil field site is discovered core samples can be extracted. Experiments are conducted on these core samples which help to give the engineers an approximation of the rock's make up. Results from numerous core sample experiments, coupled with siesmic readings give engineers a rough interpretation of the underlying reservoir's structure. This volume of rock is then theorecally broken up into grid-blocks. One method which oil companies use to view these rock structures is by rendering the reservoir's grid-blocks as a 3D model. These cuboidal grid-blocks represent the logical positions of volumes of rock within the oil field, although, in reality, the geological shapes of these cuboids can be quite irregular and may contain faults.

The problem arises as how to best display a 3D model of an oil field at a considerably fine-scale. Firstly computers in general can lack sufficient memory and processing power. These oil fields are extremely large and searching through these large datasets requires high levels of processing power. In addition to this, the file sizes generated from these large datasets can be immense. If when viewing a 3D model of an oil reservoir and one were to zoom all the way down to a single pixel on the monitor screen the pixel produced would be still prove to be far too course a representation of a single grid-block as that single pixel would still represent a vast volume of the oil field, perhaps still several hundred meters across. Some of the models being simulated at present can contain up to one hundred million of these grid-blocks. With advancements in technology, multi-core computers and higher performance graphics cards are beginning to become common place within the businesses and private sector. This is resulting in increased volumes of data which these computers are able to process and has lead to the simulation of much larger 3D models. In the foreseeable future models of up to 100 times larger than those previously modelled could be represented in a 3D visualisation. The problem still remains however, how does one zoom into these blocks to such an extent whereby they can be effectively viewed and evaluated so as to allow oil companies to more precisely pin-point where to drill, find faults in the rock formation and to generate a more exact account of the reserves of oil contained in each of the blocks without having to display all the grid-blocks at once.

The school of Computing Science at Glasgow University has extensive experience in working within the field of computer graphics and 3D models. A technique developed by the department uses a system not totally dissimilar to the Laplacian Pyramid system. This will form the basis for showing vast 3D models at a far more detailed view while the development of an effective coding algorithm will allow for quick uploading times of the model producing the best optimum manoeuvrability on the screen of the model without any delay or buffering times being endured. Fig 1 below shows an illustration of an oil rig at sea.

Click on the following link to see a fly through a 3D animated scene of an oil rig at sea. (4.34MB) 3D fly through animation of an oil rig at sea. - (9.25MB)