European Journal of Soil Science最新文献

This study is a proof-of-concept for a practical implementation of the global pedogenon map (GPM) in Europe. A pedogenon is a concept that classifies soil based on similar soil-forming factors at a given reference time rather than on soil properties. Pedogenon classes can potentially be further divided into genosoils and phenosoils, where genosoils represent soils with minimal disturbance and phenosoils reflect soils with greater disturbance due to anthropogenic pressure. This study evaluates two approaches for delineating genosoils and phenosoils across 38 European countries, using the GPM: (1) combining CORINE land cover data with a Human Modification Index (HMI) layer at a 0.2 threshold, and (2) selecting the 5% least disturbed area per pedogenon across the analysed European countries. Soil organic carbon (SOC) values from the LUCAS soil spectral library were used to assess genosoil stability. Results show that while the CORINE+HMI method delineated larger undisturbed areas, using the HMI with 5% least disturbed per pedogenon produced a more spatially balanced genosoil distribution. This second method was a more efficient strategy, as it also showed considerably lower SOC standard deviations in the genosoils, indicating that it identified more stable reference states for monitoring soil capacity and condition in Europe. By accounting for Europe's rich pedodiversity, the GPM provides a consistent basis for detecting changes over time and supports informed land and environmental management.

Visible and near-infrared (Vis–NIR) spectroscopy provides a rapid approach to assess soil properties in situ, reducing the need for labour-intensive analyses of large-scale soil surveys. However, research-grade spectrometers may present financial and logistical challenges for widespread deployment, particularly in field applications. This study evaluated the performance of two emerging low-cost spectrometers from OtO Photonics: a visible-range spectrometer (HummingBird, HB) operating at 350–1020 nm and a NIR-range spectrometer (SideWinder, SW) operating at 900–2500 nm. These spectrometers cost less than half the price of the research-grade Vis–NIR spectrometers. A total of 386 soil samples from Eastern Australia were used in the study to build models for pH, total carbon, clay, sand, and cation exchange capacity. HB, SW, and two Vis–NIR spectrometers (AgriSpec from Malvern Panalytical and PSR+ from Spectral Evolution) were tested in this study. The spectra acquired by different spectrometers were characterized by identical shapes, and the features of organic matter and iron oxides were clear. When constructing models for soil properties, the combination of HB and SW achieved performance comparable to research-grade spectrometers. Together, HB and SW provided the best predictions for clay content (R² = 0.82, RMSE = 6.93%) and sand (R² = 0.72, RMSE = 8.62%). The variable importance in projection scores indicated that the models recognized the same features for HB, SW, and standard Vis–NIR spectrometers. Given their capacity to predict soil properties, the low-cost spectrometers were expected to undertake versatile tasks, such as mapping of soil organic carbon and high-resolution monitoring of soil conditions.

Soil health is critical for sustaining ecosystem functions and addressing environmental challenges. Effective soil health management requires reliable methods for assessing soil properties. Soil spectroscopy may allow resource-effective assessment of soil properties, but more knowledge is needed to transfer knowledge from laboratory-grade spectrometers to in-field data acquisition. This study explores the predictive potential of selected spectral subsets from the full visible and near-infrared (VIS–NIR) range, using various machine learning algorithms (MLAs), as a theoretical exercise to support the design of practical soil sensing tools. Specifically, we evaluated whether narrower spectral ranges can provide predictions comparable to those achieved with the full VIS–NIR spectrum. The ranges are chosen to emulate the spectral coverage and resolution of commercially available sensors which are candidates for widespread and resource-effective data collection. We used the VIS–NIR spectral data (400–2500 nm) alongside laboratory analyses of several soil properties to be predicted from the pan-European LUCAS dataset. We employed four different MLAs for estimating soil properties: support vector regression (SVR), cubist, random forest (RF), and multi-layer perceptron (MLP), which were benchmarked against ordinary least squares regression. Our results showed that spectral subset ranges of 1000–2500 nm and 1350–2500 nm (emulating Trinamix and NeoSpectra sensors, respectively) yielded prediction accuracies similar to the full spectrum. Spectral subsets limited to the visible and early NIR range (350–1000 nm) were less effective. The most informative spectral features were found in wavelengths above approximately 1750 nm. Among MLAs, MLP consistently delivered the best performance, particularly when estimating organic carbon, nitrogen, pH and clay, which were predicted with greater accuracy compared to potassium (K), phosphorus (P) and coarse fragments (CF) which cannot yet be robustly predicted from spectral data alone. This study provides preliminary insight into the spectral regions most relevant for soil property prediction. These findings may inform future development and optimisation of real-world soil sensors. Validation with actual sensor data, both on dried and in situ samples, remains an important next step.

Soil is a vitally important resource supplying numerous ecosystem services such as the provision of food, water purification, nutrient regulation, and biodiversity (Adhikari and Hartemink 2016; Pereira et al. 2018). At the same time soils are increasingly impacted by natural and anthropogenic soil threats such as erosion, contamination and sealing (JRC: Institute for Environment and Sustainability et al. 2015). For example, Sonderegger and Pfister (2021) estimate a long-term global productivity loss of 15% in high-input production areas, of which 43% is due to compaction and 57% due to erosion. The past decades have seen a worldwide increasing pressure on soils due to population growth (Brown 1981), which led to ongoing urbanization (Sun et al. 2023), intensification of agriculture (Kopittke et al. 2019) and a decrease in the maintenance of traditional land use–land management systems, such as extensive ranging or hedgerows and terraces (Plieninger et al. 2006; Arnaiz-Schmitz et al. 2018; Durán et al. 2020). In addition to these developments, climate change, with increases in extreme weather events, both in terms of longer meteorological droughts as well as shorter, more intense rainfall events (Furtak and Wolińska 2023), has led to more frequent and more severe floods and droughts and a worldwide increase in soil degradation (Montanarella et al. 2016). Thus the ecosystem services provided by soils are declining due to anthropogenic activities, while at the same time mankind is dependent on these different ecosystem services.

In view of these challenges, there is a move in research and practice toward more sustainable soil and land management to recover degraded soils and maintain or improve soil health, namely the capacity of soil to function (Lehmann et al. 2020). The ecosystem services supplied by soils are the result of multiple concurrently acting soil functions, which can be supported by sustainable soil and land management using e.g., land management frameworks (Schulte et al. 2014). The United Nations (1992) defines sustainable land management (SLM) as “The use of land resources, including soils, water, animals and plants, for the production of goods to meet changing human needs, while simultaneously ensuring the long-term productive potential of these resources and the maintenance of their environmental functions.” Despite the development of decision support systems at the field scale (Debeljak et al. 2019), quantification of the influence of sustainable land management practices on several soil functions remains challenging. Important steps to improve this quantification are the inclusion of biological soil indicators in soil health assessments (Vazquez et al. 2025), improvement of spatiotemporal measurement me