Combining biodiversity and geodiversity on landscape scale: A novel approach using rare earth elements and spatial distribution models in an agricultural Mediterranean landscape

Landform diversity influences and interacts with both biodiversity and geodiversity and thus, they are key factors in the assessment of landscape resilience. However, research on the spatial relationships between landscape geodiversity and biodiversity is challenging because we are still lacking methods to link abiotic with biotic factors. The goal of this study is to explore and quantitatively assess the spatial relationship between geomorphometric factors and the relative distribution of rare earth elements (REEs) in soils and organism. Therefore, we selected a representative Mediterranean landscape characterized by ancient olive grove cultivations. The results show for different landforms and lithotypes a positive linear correlation in the lanthanum/samarium vs. lanthanum/ytterbium (La/Sm vs. La/Yb) signature between the bioavailable fraction of topsoil and olive drupe. Results of La/Yb vs. La/Sm reported as power function for olive drupes and topsoil follow comparable scaling ranges showing a power law of 0.83 and 0.71 respectively with an R2 0.96 vs. 0.71. A different scaling range behavior from topsoil to the related olive drupe was found for each parent rock material. Results of the Machine Learning (ML) modelling framework showed that the LaN/SmN in topsoil, were substantiality correlated to channel network base level, topographic wetness index, NDWI and valley depth. Under the physiographic environmental variables of the study area, the spatial distribution of LaN/YbN was mainly related to the lithological characteristics. Furthermore, NDVI was the most important variable to predict the fractionation ratio of LaN/YbN in olive drupe and the topographic channel network distance for LaN/SmN in olive drupe. Our findings provide new insights in the spatial distribution of REEs allowing an assessment of bio- and geodiversity of olive groves taking into account biophysical factors. Our research represents a starting point for future applications and modelling techniques to analyze at the catchment-scale the REEs biophysical fluxes and food traceability.