Soil and Water Research - In Press
Spatial assessment of potential wind-driven soil loss using the Wind Erosion Equation Original Paper
Jana Podhrázská, Josef Kučera, Martin Blecha, Jan Szturc, Hana Středová, Tomáš Středa, Petra Fukalová
Wind erosion represents a locally significant soil degradation process in the Czech Republic, particularly in intensively farmed lowland regions. While areas susceptible to wind erosion have been previously identified, spatially explicit quantification of potential soil loss expressed in t·ha⁻¹·yr⁻¹ at the national scale has so far been lacking. This study presents a comprehensive assessment of potential wind-driven soil loss across the Czech Republic using the Wind Erosion Equation (WEQ). Special attention is given to the soil erodibility index (I), which was derived from extensive laboratory analyses of soil aggregates and evaluated using multiple statistical representations (Q25, median, mean, Q75, and Q90). The resulting variants were used to quantify the sensitivity of modelled soil loss to erodibility assumptions and to compare exceedance of national (9 t·ha⁻¹·yr⁻¹) and European (2 t·ha⁻¹·yr⁻¹) reference limits. Results show substantial spatial variability in index I and associated soil loss estimates. Using the recommended median-based variant, approximately 10% of agricultural land exceeds the European reference limit, while only 0.8% exceeds the national threshold. Higher quantile scenarios (Q75 and Q90) identify erosion hotspots in dry lowland regions and are suitable for preventive planning. The presented outputs provide the first spatially consistent national framework for assessing potential wind erosion losses in the Czech Republic.
Balancing Data Quality in Predictive Geochemical Mapping Using Machine Learning: A Czech Regional Case Study on Topsoil NickelOriginal Paper
Jan Skála, Daniel Žížala, Robert Minařík
Machine learning makes geochemical mapping highly adaptable, as its data-driven nature allows predictions to evolve with new information. In this study, topsoil nickel (Ni) data were compiled from various sources, each with different sampling times and analytical methods. To effectively use such imbalanced data into spatial modelling, it was necessary to test how data uncertainty propagated through the final maps. A comprehensive benchmark of the quantile random forest algorithm was conducted to identify conditions under which the model performs optimally. Predictive maps of topsoil Ni at a 20-metre resolution were subsequently generated and compared using a multi-faceted evaluation strategy. This approach assessed how model adjustments—particularly those addressing uncertainty introduced by regression-based conversion of legacy measurements—affected performance. Extensive benchmarking revealed that while out-of-sample validation showed only modest improvements (e.g., RMSE reduced from 12.6 to 11.2 mg/kg) when modifying training data, covariates, or algorithm parameters, the resulting prediction grids differed substantially. The analysis also demonstrated that output variability across model scenarios occurred at different spatial scales: weighting approaches had localized effects, whereas high variability in input data propagated more broadly across the region.
Discovering Complex Pesticide Pollution in River Water Irrigated Soil/Groundwater Systems: From Targeted Analyses to Non-Targeted Screening and BackOriginal Paper
Alina Sadchenko, Petra Nováková, Aleš Klement, Miroslav Fér, Antonín Nikodem, Vít Kodeš, Radka Kodešová, Roman Grabic
Pesticides and their transformation products are increasingly detected in agricultural soils and surface waters, raising concerns about their persistence, mobility, and ecological impacts. Irrigation with river water contaminated by agricultural runoff represents a significant but understudied pathway contributing to soil pollution. In this study, we investigated pesticide occurrence across soils, irrigation water, and groundwater in three intensively cultivated river basins. Soil samples from vegetable-producing fields exhibited complex contamination profiles, with 12–40 co-occurring compounds and total residues frequently exceeding levels reported for European arable soils. Several pesticides, including pendimethalin, mandipropamid, and azoxystrobin, were found at notably high concentrations, with some soils surpassing 5000 µg/kg. Comparison of detection frequencies across matrices revealed diverse transport and retention behaviour: while certain legacy compounds (e.g., atrazine metabolites) were ubiquitous in both surface and groundwater, others showed strong soil accumulation with limited mobility. Irrigation water was identified as a non-negligible contamination source, particularly for persistent and mobile substances, although direct field applications remained the dominant contributor to peak soil concentrations. By integrating targeted and non-target screening, this study provides the most comprehensive assessment to date of pesticide burdens in river irrigated agricultural soils and highlights the need for improved monitoring strategies in systems where soil and water pollution are tightly interconnected.
Reduction in Hydraulic Conductivity for High Burn Severity Soils Begins After the First Rainfall Event: Results from Laboratory-Scale Rainfall Simulation ExperimentsOriginal Paper
Nana Afua Gyau Frimpong, Jacob Huerta, Ryan Webb
Wildfires can greatly impact the hydraulic properties of soil. This study aims to utilize laboratory experiments to simulate burned soil conditions under a range of slope angles and rainfall intensities to addresses the research question: How does the presence of ash after the first rainfall event impact the hydraulic properties of burned soil in complex terrain? Sandy loam soils for this study were sourced from a mixed conifer forested area in the San Juan Mountains of southern Colorado, USA. Measurements of soil hydraulic properties in experimental microplots were taken 1) before burning, 2) after burning, and 3) 24 hours after a rainfall simulation. A total of 15 experimental microplots with ash applied were run through rainfall simulations at slope angles ranging from 10° to 30°. Results found ash eroded had no significant relation to slope angle, but there was a significant reduction in field saturated hydraulic conductivity of samples with ash after rainfall simulations. To build on the findings presented here, future research should conduct field-based studies across various ecosystems and soil types to observe post-fire soil changes over time.