Exploring the Links between Bacterial Diversity with Vegetation and Soil Parameters Using Soil Metabarcoding Data and Sentinel-2 Indices

In this paper we explore correlations between soil bacterial diversity and spectral indices derived from satellite remote sensing. We computed alpha and beta diversity indices using metabarcoding data generated from 214 cropland soil samples collected in the context of Eurostat’s 2018 Land Use and Coverage Area frame Survey (LUCAS) Soil module. Subsequently, we derived 13 spectral indices from Sentinel2 images for the sample locations, sticking to the closest available image in relation to the sampling date. We explored the correlations between soil bacterial diversity and spectral indices using univariate Spearman correlations and Mantel tests for beta diversity dissimilarity matrices. We also stratified our analyses in terms of crop type. Significant correlations emerged between alpha and beta diversity and vegetation greenness, soil salinity, moisture, soil texture and chemical content. For beta diversity we found significant correlations only with soil-related Sentinel-2-derived indices. Correlation-based analysis of the commonest crops showed divergent results for bacterial diversity compared to the full cropland sample, and between the crop types.