ASE detecting unlawful waste dumping sites in Europe from space

ASE artice pic 2

Air & Space Evidence (ASE) is an academic spin-out company, originating from University College London. It is a world leader in using satellite technologies for environmental crime detection. The leading US business magazine Fast Company named their work as one of the main “world changing ideas” of 2015.  

Using satellite technology to combat waste crime

ASE’s specific focus in the EMERITUS project is to develop and test a detection model which uses satellite technologies and artificial intelligence (AI) to identify unlawful waste sites. There will then be an opportunity for law enforcement bodies to acquire bespoke data about unlawful waste sites in their own countries using this model via the EMERITUS platform. 

This important technological solution is being developed because criminal activity involving waste dumping is a huge problem across Europe, costing countries an estimated €72-90 billion a year. The head of the Environment Agency in England described waste crime as the “new narcotics,” as it can often be as profitable as drug trafficking. One single unlawful site in the UK operated undetected for so long it eventually contained 1.7 million tonnes of waste and potentially avoided €200 million in tax.  

Unlicenced waste dumping is increasingly organised, complex and serious in nature, and despite the dedication and clear efforts of law enforcement agencies to tackle it, it is problematic for them to effectively tackle it because there are certain characteristics inherent to unlawful waste sites that make them difficult to detect on the ground. For this reason many unlawful sites are not being detected or identified quickly enough before they cause significant harm to the environment. Innovative new technological approaches being tested in EMERITUS by ASE should significantly improve the ability to detect and intervene at unlawful waste sites, meaning less risk to the environment and human health, less tax is avoided, and clean-up will be less difficult and expensive.

ASE article pic

ASE’s detection model and future implementation

ASE has developed an award-winning detection model that uses ESA Sentinel-2 satellite data and AI algorithms to find unknown unlawful sites. The first stage of this process is to select an area of interest, such as a region of even a country, and to then run the detection model to produce latitude and longitude coordinates of all potential waste sites in that area. The second stage uses more detailed very high resolution satellite data, where an experienced analyst examines each individual site that the model had found to confirm if that site looks to be an unlawful site or not. Intelligence data without context can result in poor decisions and the inefficient use of resources. In the third stage the data is categorised and the results presented using a risk profile and confidence rating agreed with the law enforcement user.  

In EMERITUS the detection model has been mainly tested and utilised in Greece. The Hellenic Police, who are EMERTUS partners, have conducted many site visits on the ground to provide feedback on the detection model results, whether the sites were already known to them (or if they were new intelligence finds), and the helpfulness of data. The model has been found to be performing extremely well as an intelligence tool, finding significant numbers of unlawful sites that were previously unknown to the authorities.  

The detection model has also been tested to a lesser extent in other European countries, including the UK, Romania, Belgium and Italy, to ensure the model can work in different climate conditions, and to improve the quantity of the training data to make sure it will work more effectively when it is available as a service.  

It is expected that a service based on the detection model outputs will be available for law enforcement users to use on the EMERITUS platform in the second half of 2025. This will make it operationally possible for them to detect and stop unlawful waste sites over very large geographic areas more quickly, and in a more targeted and intelligence-led way.