European agriculture is under enormous pressure to achieve the level of sustainability envisaged by EU initiatives such as the Green Deal and the Farm to Fork strategy. Meanwhile, the international crisis in Ukraine also reminds us that we need to make our food system more resilient to shocks. Data-driven policy and regulation will be essential to enable a shift in our agricultural systems.
Making our food system more robust and sustainable while improving quality may sound like a contradiction, but the common market can actually be leveraged to achieve this goal through a range of measures that support farmers, keep things affordable for citizens, and improve our food system's sustainability. Across the food production landscape, more than 6.3 million CAP payers each have enormous potential to deliver customized ecological benefits with EU support.
To achieve this, we need a more dynamic regulatory approach. Simply put, we need technology that enables data-driven policy. Data on economics and the environment are already available and sustainability data will be added to future FSDN databases. SAS software can help create a fast and partially automated eligibility process.
Eye in the sky
Classification modeling based on satellite imagery is one approach to establishing a technology-driven regulatory framework. Sentinel II images, funded by the EU, can be used to recognize agricultural outcomes. SAS provides deep learning technology that helps to achieve even greater success in classifying the type of agriculture and farming activity.
We do this by using a recurrent multilayered neural network that reveals more details about the complex spectral, spatial and temporal patterns that distinguish different farm landscapes. Such a model can not only determine agricultural activity on the ground but even improve it further as more data is fed into the prediction system.
It would be a great idea for the European Commission to deploy this technology-inspired "eye in the sky." A data-driven regulatory framework can provide a lightweight and scalable shield for the common market. Take, for example, the Land Parcel Identification System (LPIS) – which codes plots of agricultural land eligible for subsidies. Each image contains knowledge pixels as fine as resolution allows. Algorithms can be trained to recognize these pixels as land-use types and pinpoint specific farming practices by iterative modeling.
The challenge with any incentive scheme is to ensure that it has the best impact per euro invested. Take the example of a large government institution in the U.S. that made national headlines after a high-profile procurement fraud incident. Senior management naturally wanted to investigate how it could have happened and whether it was an indication of a larger problem. SAS consultants, therefore, analyzed all payments, invoices and orders across the enterprise for the past three to four years – including 17,800 suppliers, 25,000 employees and 700,000 payments.
The same threat persists with expanded CAP subsidies and is more difficult to track without technology. The many approaches to promoting sustainable farming require more tracking and feedback than static regulations or manual intensive reporting can deliver. That’s why we need to strengthen technology-based assessments by training computers to quickly see what regulators need to know. Analytical models can be used to verify these myriad effects.
In conclusion, using data and analytics for technology-driven regulation can help us achieve the goal of the Green Deal: ecological regeneration while securing a future for our EU food systems. Europe’s future looks bright when the common market works effectively with the agricultural landscape to deliver the cost-effective benefits that only climate-smart farming can bring. The EU has a unique chance to set a standard for the rest of the world as we move towards an interdependent future.