Agricultural landscapes can vary widely, even within a single country. They are affected by unique combinations of local climate, landscape, geology, and cropping patterns, creating complex interactions between agriculture and the environment. In the context of regulations, environmental risks associated with agricultural practices—such as the application of fertilisers and plant protection products to fields—are evaluated based on environmental fate modelling (i.e., the prediction of environmental fate and concentrations in the environment) which relies on a set of standard scenarios. These assumptions, however, might not always match real-world conditions producing overly cautious predictions that create regulatory challenges. For instance, conservative modelling approaches might suggest a region-wide ban on a product due to water contamination risks, overlooking the existence of effective local water protection zones.

To create a more refined risk assessment, it is essential to integrate a wider array of data, such as environmental monitoring data, land use surveys, and hydrological and geological information relevant to a specific region rather than to a zone as a whole. This allows regulatory scenarios to be adapted to real-world scenarios, producing more accurate risk assessments. These data are inherently complex and often widely distributed both spatially and temporally and require tools that will allow the data to be integrated into risk assessment workflows.

Geospatial analysis, facilitated by Geographic Information Systems (GIS), is a valuable toolbox that enables the collection, integration, analysis, and contextualization of complex spatial data as well as modelling of the interactions between objects, and phenomena in space. It also allows communication of the outcomes of the analysis in a clear and concise format. In environmental risk assessments, GIS can be used for modelling, and vulnerability mapping utilising surveys, monitoring data and satellite imagery. This allows the integration of complex spatial and temporal data to create scenarios that are specific to a given region.

In a recent project, we used geospatial analysis to better understand the potential risk of a particular crop protection active substance in drinking water. In this work, a concern had been raised by regulators about the potential for this substance to enter drinking water treatment plants (DWTPs) from surface water sources that were likely to receive runoff from agricultural land. This risk was only manifest in certain DWTPs as the potential risk came not from the substance itself but from a transformation product of the ozonolysis which is a treatment process used in some DWTPs. Consequently, a geospatial assessment was performed to understand the spatial relationship between agricultural areas where this substance was applied and locations of the abstraction areas of the water treatment plant. Publicly available data sets, survey data, and web mapping services were combined to:

• Locate precisely the water treatment plant and define its abstraction points and the water protection zones for the region,
• Identify and map agricultural fields where the substance was used,
• Calculate the shortest distances between fields where the substance was used and the abstraction points for water and the nearest surface water body.

We confirmed that the distance between the closest commercial field that may be treated with the assessed substance and the abstraction area was about 15 km, thus suggesting a low risk of the substance entering the water treatment plant. Moreover, other relevant data such as land-use designations and hydraulic connectivity between the field and the drinking water abstraction point and protection area were included in the analysis. This allowed us to create a more comprehensive picture confirming that even in the event of crop pattern change, the abstraction point would remain protected from residues.

The figure below is a simplified illustration of the key elements central to this case study. It shows:

• The Abstraction Point where water is drawn for treatment
• The surrounding Water Protection Area
• The Nearest Identified Field where the crop of concern is grown

While the actual analysis included more complex data and additional factors such as survey data and modelled land cover data, this illustration provides a clear visual representation of the core components in our study.

Besides the obvious benefits of being a powerful instrument for the refinement of risk assessments, this project demonstrated how geospatial analysis can also aid in coherently communicating complex data. In this case study we were able to combine hydrological, environmental, and civilian infrastructure data in a single graphic, allowing a more intuitive understanding of the methodology and the findings, and facilitating communication and decision-making.

Integrating GIS into environmental fate modelling helps to bridge the gap between regulatory requirements and real-world agricultural practices by providing a robust framework for making informed environmentally responsible decisions. Among other applications, GIS modelling can be applied to predicting the movement and degradation of chemicals in different environmental contexts taking into account runoff patterns, soil erosion and land use. Coupled with machine learning, GIS can also support modelling of cumulative impact of plant protection product use over time considering historical and modelled data on land use patterns, crop rotations, landscape variability and climate conditions. Finally, combining GIS and machine learning can support the creation of mitigation strategies for contaminated sites as well as predict dynamic changes in contamination levels that can be expected after the deployment of mitigatory measures.

Using state-of-the-art Geospatial analytical tools, we offer specialized services in higher-tier plant protection product risk assessment. Through our team’s proficiency with GIS technology and in-depth knowledge of EU regulatory frameworks, we can provide thorough, data-driven pesticide risk evaluations.