With the increasing awareness of environmental issues and the growing concern for the planet’s health, tracking and monitoring the environment has become crucial in the fight to protect our shared natural environment. Over recent years, satellite technology and data have proven to be an effective method to quickly and accurately monitor environmental changes.
From tracking climate patterns to detecting natural disasters and mapping land use changes, satellites have revolutionised how we collect environmental data. Using our expertise in Earth Observation and Remote Sensing at Mallon, we are exploring how readily available satellite data can assist in detecting and tracking environmental issues.
In this blog, we will examine how satellite data can be used to track oil spills out at sea, using a recent spill off the coast of Thailand as an example. Mallon colleague Rafal Marciniak compiled the following case study using data captured from the European Space Agency’s Copernicus Programme and its Sentinel 1 & 2 satellites.
Case Study – Tracking an oil leak in the Benchamas Oil Field
On 15th March 2023, an explosion aboard the Floating Storage and Offloading Unit (FSO) Benchamas 2 caused significant damage to the vessel, resulting in a leak and a risk of sinking. The vessel was located here, off the coast of Thailand. With the boat carrying 400,000 barrels of crude oil, the potential for a significant environmental disaster was very real.
Analysing Sentinel 2 satellite data (in the visible light spectrum) from 6th April and Sentinel 1 data (in the microwave spectrum) from 8th and 9th April showed an oil spill, not from Benchamas 2 but from a platform in the Benchamas oil field. The spill was detected using the Segment Anything Model (SAM) and comparing the time sequence with a wind direction map. SAM is an AI model capable of identifying individual objects within an image even if it has never encountered them. This enables us to search multiple images for a single object, drastically improving the time needed to analyse large datasets over an extended timeframe.
Imagery from Sentinel 2 shows a leak ~3.6 km long from the platform heading eastwards in line with the wind direction.
Imagery from Sentinel 1 captured two days later shows a larger stain ~8.5 km long, heading towards the North–East.
A day later, further imagery captured by Sentinel 1 shows an even larger oil streak ~15.3 km long.
The available satellite imagery shows us that the leak from the Benchamas platform lasted for a minimum of 3 days, causing an oil stain that grew to ~15 km in length. Unfortunately, we could not tell when the leak started or its current state.
What Can We Learn From This?
Firstly, the analysis of imagery around the Benchamas platform following reports of a possible leak highlights the vast potential of Earth Observation to track and monitor the progress of environmental issues.
Added to this is the use of newer technology, such as SAM, that allows us to analyse satellite imagery for specific features over significant time periods. This can dramatically speed up the time taken to investigate and detect potential environmental disasters.
When analysed satellite imagery is combined with other additional datasets, such as wind directions, and the resulting information can be used to help predict the movement of spills. This information can be invaluable to those on the ground looking to contain or clean up any spills by assisting them in preparing appropriate resources and informing potential areas where an oil spill could hit or determining where a leak originated.
Further Information
Please get in touch with us below for further information on the methods used in this case study or to enquire about our Earth Observation services.
