Soils are increasingly rediscovered as a vital resource that underpins many natural and societal services. Over more than half a century, agricultural mechanization and a singular focus on plant production, supported by chemical fertilizers, have led to widespread soil degradation. This reductionistic perspective has relied on soil observations focused on physico-chemical properties; properties that can be boosted by chemical additions but ignore the biological and ecological status and functions of the soil. Recognizing the importance of natural soil processes, which have evolved and been fine-tuned over billions of years, a new set of indicators for describing soil health beyond the physico-chemical properties is required. These indicators should preferably be observable and analyzable by farmers, advisors, extension workers and other citizen scientists. Methods that directly or indirectly capture the biological and ecological functions include, for instance i) environmental DNA (eDNA) metabarcoding to characterize the diversity and composition of soil microbial communities, ii) activity rates of key enzymes involved in the main biogeochemical cycles, iii) the ratio of soil fungi to bacteria, an indicator of the extent of disturbance in soil ecosystems, iv) aggregate stability, which is important for soil erosion resistance, and water and nutrient holding capacity, and v) water infiltration capacity as a key measure of the soil water absorption, holding and release potentials. While eDNA requires specialist laboratories and databases, the other methods are currently available for “Do-It-Yourself” (DIY) testing. In this study, as part of the EU-funded project AI4SoilHealth (https://ai4soilhealth.eu) we sampled soils in Greece, Sweden, Finland, Croatia and Denmark. We applied the outlined methods alongside traditional wet chemistry analysis of properties such as carbon, pH and electrical conductivity, and the particle size distribution. These properties were also estimated by leveraging their correlations with diffuse reflectance Near InfraRed (NIR) spectra and applying machine learning models. We are testing both the robustness of the novel methods and their interdependence with more traditional physico-chemical properties and soil spectroscopy. We hypothesize that there is a significant positive correlation between novel indicators (e.g. eDNA richness is correlated to enzymatic activity, which is correlated to aggregate stability, which in turn is correlated to infiltration capacity) and that high scores of the biological and ecological properties are correlated with, for instance, soil carbon content. This study explores the potential of these novel methods for more holistic understanding of soil health.

Most soil properties are continuously varying over different scales in both space and time. An analysis of field sampled soil is the most accurate method for estimating the spatial distribution of soil properties and health indicators at field scale. The field scale spatiotemporal variation in properties, however, requires an effective, well selected and unbiased probability distribution based sampling framework leveraging the in-situ variability of the relevant environmental covariates. Covariates that can be used for modeling the soil properties over the entire study area. Typically such covariates are represented at spatial rasters derived from e.g. topographic data and satellite imagery. In this work, we introduce the probability based balanced stratified sampling algorithm compatible with the proposed European Soil Monitoring law. The algorithm distributes doubly balanced sampling locations over both geographical and feature space constrained by a maximum allowed error – in our case the coefficients of variation. The geographical feature space input data are selected covariates from the Soil Health Data Cube (https://shdc.ai4soilhealth.eu/). We target the covariates that reflect the distribution of soil health indicators tested or developed by the EU funded project AI4SoilHealth (https://ai4soilhealth.eu). To begin this process, a Bethel-inspired optimization approach is applied to stratify the study areas. The strata are then used for computing the approximately equal inclusion probabilities for all units and the number of samples for each strata allocated based on the available auxiliary information. The second stage of the process involves doubly balancing the algorithm, with the aim to select the optimal sampling locations over geographical and feature space with respect to the stratification. In our study, we compare the algorithm with 1) simple random sampling, 2) feature space coverage sampling and, 3) the EU wide Land Use and Cover Area frame Survey (LUCAS) algorithm using legacy data. The advantages of the novel algorithm are demonstrated in the optimized number of samples while preserving the accuracy of target estimates and mapping accuracy. In the first experiment, we subsample the legacy random grid samplings. In the second (numerical) experiment, we use the soil property maps of the Soil Health Data Cube as observed values to design new optimized sampling networks with no gridded location constraints.