Calogero Schillaci

Digital soil mapping (DSM) produces spatially continuous soil information for land-use planning, carbon accounting, and Earth system analysis, resting on the SCORPAN framework and its use of remote-sensing and terrain covariates as indicators of soil-forming processes. Similar approaches estimate ecological and hydrological conditions from vegetation, moisture, and climate data. Two assumptions receive little scrutiny: that soil-covariate relationships stay constant in space and time, and that covariates act as independent predictors. In reality vegetation, soil, and water form a coupled system in which each influences the others, and the strength and even the direction of these interactions vary across landscapes and shift over time. We argue that these dynamic relationships are an important but under-recognised source of structural uncertainty, distinct from the sampling, measurement, and model-fitting error usually partitioned in DSM. Drawing on wetlands, drylands, sloping farmland, burned watersheds, drought-affected forests, and thawing permafrost, we show how the feedbacks break the assumptions behind DSM, why the same problem affects land-surface, ecological, and hydrological models, and how mapping might be made feedback-aware.

Digital soil mapping (DSM) produces spatially continuous soil information for land-use planning, carbon accounting, and Earth system analysis, resting on the SCORPAN framework and its use of remote-sensing and terrain covariates as indicators of soil-forming processes. Similar approaches estimate ecological and hydrological conditions from vegetation, moisture, and climate data. Two assumptions receive little scrutiny: that soil-covariate relationships stay constant in space and time, and that covariates act as independent predictors. In reality vegetation, soil, and water form a coupled system in which each influences the others, and the strength and even the direction of these interactions vary across landscapes and shift over time. We argue that these dynamic relationships are an important but under-recognised source of structural uncertainty, distinct from the sampling, measurement, and model-fitting error usually partitioned in DSM. Drawing on wetlands, drylands, sloping farmland, burned watersheds, drought-affected forests, and thawing permafrost, we show how the feedbacks break the assumptions behind DSM, why the same problem affects land-surface, ecological, and hydrological models, and how mapping might be made feedback-aware.

Digital soil mapping (DSM) produces spatially continuous soil information for land-use planning, carbon accounting, and Earth system analysis, resting on the SCORPAN framework and its use of remote-sensing and terrain covariates as indicators of soil-forming processes. Similar approaches estimate ecological and hydrological conditions from vegetation, moisture, and climate data. Two assumptions receive little scrutiny: that soil-covariate relationships stay constant in space and time, and that covariates act as independent predictors. In reality vegetation, soil, and water form a coupled system in which each influences the others, and the strength and even the direction of these interactions vary across landscapes and shift over time. We argue that these dynamic relationships are an important but under-recognised source of structural uncertainty, distinct from the sampling, measurement, and model-fitting error usually partitioned in DSM. Drawing on wetlands, drylands, sloping farmland, burned watersheds, drought-affected forests, and thawing permafrost, we show how the feedbacks break the assumptions behind DSM, why the same problem affects land-surface, ecological, and hydrological models, and how mapping might be made feedback-aware.

Digital soil mapping (DSM) produces spatially continuous soil information for land-use planning, carbon accounting, and Earth system analysis, resting on the SCORPAN framework and its use of remote-sensing and terrain covariates as indicators of soil-forming processes. Similar approaches estimate ecological and hydrological conditions from vegetation, moisture, and climate data. Two assumptions receive little scrutiny: that soil-covariate relationships stay constant in space and time, and that covariates act as independent predictors. In reality vegetation, soil, and water form a coupled system in which each influences the others, and the strength and even the direction of these interactions vary across landscapes and shift over time. We argue that these dynamic relationships are an important but under-recognised source of structural uncertainty, distinct from the sampling, measurement, and model-fitting error usually partitioned in DSM. Drawing on wetlands, drylands, sloping farmland, burned watersheds, drought-affected forests, and thawing permafrost, we show how the feedbacks break the assumptions behind DSM, why the same problem affects land-surface, ecological, and hydrological models, and how mapping might be made feedback-aware.

Digital soil mapping (DSM) produces spatially continuous soil information for land-use planning, carbon accounting, and Earth system analysis, resting on the SCORPAN framework and its use of remote-sensing and terrain covariates as indicators of soil-forming processes. Similar approaches estimate ecological and hydrological conditions from vegetation, moisture, and climate data. Two assumptions receive little scrutiny: that soil-covariate relationships stay constant in space and time, and that covariates act as independent predictors. In reality vegetation, soil, and water form a coupled system in which each influences the others, and the strength and even the direction of these interactions vary across landscapes and shift over time. We argue that these dynamic relationships are an important but under-recognised source of structural uncertainty, distinct from the sampling, measurement, and model-fitting error usually partitioned in DSM. Drawing on wetlands, drylands, sloping farmland, burned watersheds, drought-affected forests, and thawing permafrost, we show how the feedbacks break the assumptions behind DSM, why the same problem affects land-surface, ecological, and hydrological models, and how mapping might be made feedback-aware.

Soil profiles are the primary field evidence from which pedogenesis, soil classification and land‐use effects are interpreted, yet their availability remains uneven across climates and land‐cover contexts. Here, we analyse the World Soil Information Service (WoSIS) 2023 snapshot together with Copernicus Global Land Cover 2015 and the global biome delineations to assess the distribution, classification status and temporal coverage of legacy soil‐profile observations. Cropland is the most represented land‐cover class, with 67,370 profiles (29.5%), followed by herbaceous vegetation with 40,277 profiles (17.7%), whereas wetlands, mangroves, tundra and several forested or arid systems remain rather underrepresented. The temporal distribution of profiles with recorded sampling dates increases markedly from the mid‐twentieth century to the 1980s and declines after 1990, suggesting a possible shift away from field‐based profile description and towards mapping, modelling and remotely sensed approaches. The review of forest, volcanic, cold‐region, arid, wetland, cropland, grassland, urban and Mediterranean soils further shows that profile description remains essential for interpreting soil functions, diagnostic horizons, degradation processes and land‐management constraints. This work provides a repeatable workflow for future WoSIS releases and other harmonised soil databases, and highlights the need to recover, standardize and publish legacy soil‐profile information while training a new generation of pedologists able to combine field description with digital soil science.

Land degradation is defined as the persistent reduction in the capacity of land to support humans and other life on Earth. Intensified land use transformation from natural to sealed surfaces and intensive management practices accelerates the pace of land degradation. Several land‐use changes constitute potential drivers of degradation; additional pressures, including intensive tillage, pesticide applications, unsustainable irrigation, and excessive fertilizer use, further compromise soil resilience and undermine conservation efforts. The agricultural sector and the natural resources systems face heightened threats such as loss from erosion, decreasing soil organic carbon stocks, and increasing soil salinization due to poor irrigation water use. In addition to on‐site adverse effects on soil, water, and air resources, these processes exert off‐site negative impacts on socioeconomically vulnerable communities. Over recent decades, climate change, driven by anthropogenic greenhouse gas emissions, has intensified pressures on terrestrial ecosystems through rising temperatures and increased frequency of weather extremes, thereby accelerating land degradation processes. Primary research and systematic evidence synthesis are needed to better address land degradation in response to the combinations of climate change drivers and ongoing human activities. In this editorial, we will provide a comprehensive overview of the anthropogenic impacts on land degradation highlighting the need for integrated approaches that address the social, economic, and environmental dimensions of this critical issue. This editorial draws on a range of studies and examples from around the world, outlining maily on contributions to this special issue “Anthropogenic Impacts on Land Degradation in a Changing World” and other LDD papers, to demonstrate the complexity and urgency of the land degradation issue. We will also emphasize the importance of addressing land degradation in the context of the SDGs, particularly SDG 15.3.1, and the need for a coordinated and integrated approach to addressing this issue.

The soil electrical conductivity (EC) is a measure of the ability of soil to conduct electricity, and it can be used as a proxy for soil salinity. Electrical conductivity is the soil indicator laid down in the just entered into force Soil Monitoring and Resilience Directive (Soil Monitoring Law) for measuring soil salinity degradation process with a 4 dSm ⁻¹ threshold to be met at European level. The Land Use/Cover Area frame statistical Survey Soil (LUCAS) soil module has until now analysed diluted EC using the ISO standard ISO 11265:1994—Soil quality. LUCAS EC measurements were generated under ISO 11265:1994, which has since been superseded by ISO 11265:2025. This standard measures the EC in soil water extract using a ratio 1:5 (EC _1:5 ). Two standards have been included in the SML for determination of soil EC. Diluted EC _1:5 will be determined by using the ISO 11265 and the saturated paste EC will be determined by using the GLOSOLAN standard. In the Soil Monitoring and Resilience Directive (Soil Monitoring Law), a threshold of 4 dS m ⁻¹ from saturated paste EC measurements has been laid down and, therefore, further investigations should be conducted to provide either threshold value for EC _1:5 based on experimental measurements or by providing transfer functions. Additionally, the GLOSOLAN lab standard for EC _e measurements is requesting a large amount of soil samples (between 200 and 400 g per soil measurement) and further investigations should be done for optimization of soil sample to make it suitable and reliable. This work proposes a methodological approach to convert EC _1:5 to EC _e by using LUCAS EC _1:5 topsoil survey measurements using machine learning Pedotransfer Functions (PTFs). In a first step, due to LUCAS soil archive limitation, reduced soil quantity for the EC _e analysis has been tested to provide insights to the reliability of the analytical procedure using different amounts of samples. Model comparison based on R ² , RMSE and MAE indicated that XGBoost outperformed the other methods; RF is a robust model that performs well across all datasets, Quantile Regression Forest and linear regression models, achieving the highest predictive accuracy and lowest error. On the full dataset, XGBoost reached an R ² of 0.92, RMSE of 0.01 dS m ⁻¹ and MAE of 0.11 dS m ⁻¹ , and was therefore selected as the reference approach for the proposed EC _1:5 –EC _e PTF. The full set of functions and new dataset are provided as Supporting Information. This work will allow several stakeholders to use LUCAS EC data for soil assessments.