GEMAS: Phosphorus in European agricultural soil - sources versus sinks at the continental-scale - the geological perspective.

Phosphorus (P) is one of the essential elements for life on Earth. As a major nutrient it is needed for healthy growth both in plants and living organisms. Although the abundance of P in the Earth's upper continental crust is relatively high (655 mg/kg), many soil types are poor in available phosphorus. The main natural factors controlling the availability of P in soil are pH, mineralogy, and formation of insoluble complexes with Al and Fe under acidic, and with Ca and Mg under alkaline soil conditions. Superimposed weathering processes and climate contribute strongly to P mobility and availability. Additionally, a large fraction of total soil P is in organic forms, which are not directly available to plants. Phosphorus is a major component in fertilisers and thus a significant source of anthropogenic P in soil and water. In the agricultural soil samples that were collected during the Geochemical Mapping of Agricultural and grazing land Soil (GEMAS) project, the total P concentrations (XRF, median 786 m/kg) are only slightly higher than those extracted by hot aqua regia (AR, median 653 mg/kg), while the median concentration in the weak MMI® cold extraction is as low as 4.1 mg/kg. The AR results show very low P concentrations over the coarse-grained sandy sediments of the last glaciation in central and northern Europe and in calcareous soil. The southern limit of the last glaciation is visible as a concentration break on the geochemical maps. In general, north-eastern and north-western Europe are marked by high P values, probably related to cold and humid climate and enrichment in humus-rich coastal soil. The spatial distribution of P at the continental-scale is dominated by geogenic and climatic factors, and the anthropogenic influence is difficult to assess and quantify.

Bioavailable 87Sr/86Sr in European soils: A baseline for provenancing studies.

We present 87Sr/86Sr isotope ratios for ~1200 selected soil samples, collected by the GEMAS consortium from grazing (Gr) and agricultural (Ap) soils in Europe with the aim to better understand the strontium isotope distribution in the bioavailable fraction of the top-soil and its potential for provenancing applications. Spatial analysis shows that there is a clear distinction between coastal (<100 km) and non-coastal (>100 km) samples in their variance and that this variance is mirrored in the sodium concentration, suggesting an important but highly variable contribution from seaspray. We present two 87Sr/86Sr maps at 25 km × 25 km scale: one based solely on the measured data using a classical kriging approach and one based on a Random Forest model using complementary GEMAS data to predict the strontium isotope composition at the remaining 3000+ GEMAS sampling locations, including appropriate uncertainty assessment. Using a forensic Bayesian likelihood ratio approach, a tool was developed in R to create provenancing likelihood ratio maps. The maps delineate areas of high and low likelihood and allow investigators to direct their resources to areas of interest. For actual forensic case work either the measured or the modelled data can be used as reference data for the overall distribution of 87Sr/86Sr values in Europe.

U-Th signatures of agricultural soil at the European continental scale (GEMAS): Distribution, weathering patterns and processes controlling their concentrations.

Agricultural soil (Ap-horizon, 0-20cm) samples were collected in Europe (33 countries, 5.6millionkm2) as part of the GEMAS (GEochemical Mapping of Agricultural and grazing land Soil) soil-mapping project. The GEMAS survey area includes diverse groups of soil parent materials with varying geological history, a wide range of climate zones, and landscapes. The soil data have been used to provide a general view of U and Th mobility at the continental scale, using aqua regia and MMI® extractions. The U-Th distribution pattern is closely related to the compositional variation of the geological bedrock on which the soil is developed and human impact on the environment has not concealed these genuine geochemical features. Results from both extraction methods (aqua regia and MMI®) used in this study support this general picture. Ternary plots of several soil parameters have been used to evaluate chemical weathering trends. In the aqua regia extraction, some relative Th enrichment-U loss is related to the influence of alkaline and schist bedrocks, due to weathering processes. Whereas U enrichment-Th loss characterizes soils developed on alkaline and mafic bedrock end-members on one hand and calcareous rock, with a concomitant Sc depletion (used as proxy for mafic lithologies), on the other hand. This reflects weathering processes sensu latu, and their role in U retention in related soils. Contrary to that, the large U enrichment relative to Th in the MMI® extraction and the absence of end-member parent material influence explaining the enrichment indicates that lithology is not the cause of such enrichment. Comparison of U and Th to the soil geological parent material evidenced i) higher capability of U to be weathered in soils and higher resistance of Th to weathering processes and its enrichment in soils; and, ii) the MMI® extraction results show a greater affinity of U than Th for the bearing phases like clays and organic matter. The comparison of geological units with U anomalies in agricultural soil at the country scale (France) enables better understanding of U sources in the surficial environment and can be a useful tool in risk assessments.

The concept of compositional data analysis in practice--total major element concentrations in agricultural and grazing land soils of Europe.

Applied geochemistry and environmental sciences invariably deal with compositional data. Classically, the original or log-transformed absolute element concentrations are studied. However, compositional data do not vary independently, and a concentration based approach to data analysis can lead to faulty conclusions. For this reason a better statistical approach was introduced in the 1980s, exclusively based on relative information. Because the difference between the two methods should be most pronounced in large-scale, and therefore highly variable, datasets, here a new dataset of agricultural soils, covering all of Europe (5.6 million km(2)) at an average sampling density of 1 site/2500 km(2), is used to demonstrate and compare both approaches. Absolute element concentrations are certainly of interest in a variety of applications and can be provided in tabulations or concentration maps. Maps for the opened data (ratios to other elements) provide more specific additional information. For compositional data XY plots for raw or log-transformed data should only be used with care in an exploratory data analysis (EDA) sense, to detect unusual data behaviour, candidate subgroups of samples, or to compare pre-defined groups of samples. Correlation analysis and the Euclidean distance are not mathematically meaningful concepts for this data type. Element relationships have to be investigated via a stability measure of the (log-)ratios of elements. Logratios are also the key ingredient for an appropriate multivariate analysis of compositional data.