Earth Observation based multi-scale analysis of crop diversity in the European Union: first insights for agro-environmental policies

TL;DR

This study introduces a high-resolution, EO-derived crop map for the EU (2018) to quantify multi-scale crop diversity using $α$-, $β$-, and $γ$-diversity across nested grids (1–100 km) and NUTS administrative units. It shows pronounced cross-country variation in diversity, with $α$-diversity ranging 2.3–4.4 and $γ$-diversity 3.5–7.5, and identifies a four-quadrant typology of countries based on how γ-diversity changes with scale. The results reveal that $γ$-diversity increases with scale, that higher local diversity ($α$) is tied to smaller farm sizes in certain countries, and that EU crop diversity generally exceeds the USA, with distinct scale-relations (logarithmic for EU, more linear for USA). The work provides a policy-relevant framework for CAP monitoring and regional targeting, highlighting the importance of scale and regional specificity, while outlining future enhancements through time-series, finer grids, and integration with ecosystem services.

Abstract

To understand the resilience of farms and the agricultural sector, as well as the provision of ecosystem services, we need to characterize and quantify crop diversity. Using a 10m resolution satellite-derived product, we created datasets of crop diversity across spatial and administrative scales for 27 EU countries and the UK in 2018. We define local crop diversity, or $α$-diversity, at a 1km scale, corresponding to large or clusters of small-to-medium-sized farms. $α$ crop diversities range from 2.3 to 4.4, with higher levels in systems with many small farms (averaging less than 10 ha). $γ$-diversity, the number and area of crops grown independently of location, increases from 2.85 at 1km to 3.86 at 10km, and levels off at 4.27 at 100km. These levels are higher than those reported in the U.S., possibly due to differences in farm structure and practices. $β$-diversity, the ratio of $γ$ and $α$ diversities, measures the difference between agroecosystems and ranges from 1.2 to 2.3 across EU countries. We classify countries' crop diversities into four groups based on the magnitude and change of $γ$-diversity across scales, with implications for regional to national agro-environmental policy recommendations. Continental Copernicus crop type maps will enable temporal comparisons, and exploring ecosystem co-variates will deepen our understanding of the link between crop diversity and agro-ecosystem services.

Figures (18)

• Figure Fig. 1: Definitions of α-, β- and γ-diversities and visualisation of different spatial scales — administrative regions or grids at different resolutions — adopted in this study. α-diversity: Denotes local diversity and is computed at the finest scale of 1 km x 1 km. γ-diversity: Represents regional diversity, computed at various administrative regional aggregations (from national to sub-national levels) or in coarser resolution grids ranging from 2 km x 2 km to 100 km x 100 km. β-diversity: Defined as the ratio between γ and α, it serves as a measure of inter-ecosystems diversity.
• Figure Fig. 2: Crop diversity is computed at different grid scales ranging from 1 to 100 km as illustrated here for a region in Latvia-Lithuania. For each subfigure, panel A shows the scales 1-km, 2-km, 5-km and 10-km with the crop map in background. Panel B shows the scales 10-km, 20-km, 50-km and 100-km with the crop map in background. Panel C shows the γ-diversity for the respective sample squares shown in A and B. Panel D shows the proportion of crop types for the different scales. These proportions are used to compute the Shannon diversity.
• Figure Fig. 3: Fit of α-diversity ($\alpha_{NUTS_{0}}$) against γ-diversity ($\gamma_{NUTS_{0}}$) at national level (see country abbreviations in \ref{['tab:countrycodes']}).
• Figure Fig. 4: Cumulative distribution of α-diversity (A) and γ-diversities (B) for 1 km to 100 km by grid scale in the EU-28. On figure (B), we see that there is probability of $50\%$ that $\alpha_{1km}=\gamma_{1km}$ takes a value of less or equal to 2.55 and that $\gamma_{100km}$ takes a value of less or equal to 4.3.
• Figure Fig. 5: Maps of the EU-28 with γ-diversity computed for all grid scales (1 km, 2 km, 5 km, 10 km, 50 km and 100 km). Grid cells with less than 1% of cropland area have been filtered out in order to remove the effect of erratic 10 m x 10 m pixels from the EU crop map. This threshold is arbitrary and can be tweaked by any data user (see Section \ref{['sec:code_and_data']}).