A proposal for a statistical framework to define a harmonised monitoring system in Europe, incorporating existing networks.

In Europe, the Soil Monitoring Law (SML) was published on 26 November 2025 and entered into force on 16 December of the same year. Member States have three years from that date to transpose the directive into national law. They must also set up a harmonised monitoring system, with common European descriptors and methodology, in order to assess the health of soils throughout their territory. The directive emphasises the creation of public registers and a risk-based approach, which implies increased collaboration between administrations, scientists and local actors. 
The aim of this presentation is to explore the statistical aspects of the design of the harmonised European network and their implications for Member States. To this end, a statistical framework is proposed to simulate possible configurations of future networks, while taking into account the constraints imposed by SML and pre-existing networks in the territories. This generic approach is discussed and implemented for France, which has existing networks. 
In particular, we test different definitions of the concept of soil unit and soil district. We also propose using existing maps produced by digital mapping methods to feed into the Bethel algorithm, which allows for the optimal allocation of sampling units to strata in stratified random sampling. An important aspect of this work is the integration of prediction uncertainty into sampling unit allocation, thanks to the work carried out as part of the OSPATS method (de Gruijter et al., 2015).

 

de Gruijter, J. J., B. Minasny, and A. B. McBratney. 2015. “Optimizing Stratification and Allocation for Design-Based Estimation of Spatial Means Using Predictions with Error.” Journal of Survey Statistics and Methodology 3 (1): 19–42.