Assessing Biodiversity based on the Spectral Variation Hypothesis at a Natura 2000 site

Grasslands play an important role for biodiversity in Central Europe, but their species diversity is increasingly declining due to land-use intensification. Monitoring their diversity is crucial, especially in the context of the extension of the Natura 2000 ecological network (N2k). However, field sampling methods can be expensive and time-consuming when surveying large areas of grasslands, which limits the feasibility of this traditional approach. To overcome these challenges, remote sensing has emerged as a promising technique for monitoring biodiversity dynamics. By analysing reflectance values from different spectral bands, it is possible to estimate the spectral heterogeneity of an area, which is the variability of reflectance values across different wavelengths. The Spectral Variation Hypothesis (SVH) proposes that the spectral heterogeneity derived from remote sensing data can be used as a proxy for species diversity.

Here, we evaluated the use of the SVH to estimate the biodiversity of grasslands in Germany. Furthermore, we investigated how the spectral variance behaves under different conservation measures, mainly linked to the management of grasslands under the Vertragsnaturschutzprogramm (VNP). We therefore calculated three indices: Rao’s quadratic heterogeneity measure (Rao’s Q) to estimate ecosystem heterogeneity, Shannon’s diversity index (H’) for α-, and Bray Curtis dissimilarity (BC) for β-diversity.

Our results show that anthropogenic landscape features predominantly affect spectral variation, rather than species diversity. The observed variations in index values between different VNP statuses are mainly influenced by the time of mowing. Furthermore, when investigating the correlation between species richness and Rao's Q index, only one VNP status at a single date showed a significant correlation.

We conclude that the SVH has the potential to serve as a proxy for species richness, but more knowledge about the grassland (e.g., species composition) is needed. Further research is required to identify the drivers of spectral variation in grasslands beyond the effects of management practices.