A5058842572- Soil Geostatistics and Mapping
- Soil erosion and sediment transport
- Soil and Unsaturated Flow
- Soil Moisture and Remote Sensing
- Climate change and permafrost
- Geology and Paleoclimatology Research
- Soil Carbon and Nitrogen Dynamics
- Remote Sensing and LiDAR Applications
- Landslides and related hazards
- Forest ecology and management
- Geochemistry and Geologic Mapping
- Ecology and Vegetation Dynamics Studies
- Hydrology and Watershed Management Studies
- Cryospheric studies and observations
- Hydrology and Sediment Transport Processes
- Remote Sensing in Agriculture
- Land Use and Ecosystem Services
- Environmental and Agricultural Sciences
- Biocrusts and Microbial Ecology
- Plant Water Relations and Carbon Dynamics
- Mineral Processing and Grinding
- Image Processing and 3D Reconstruction
- Spectroscopy and Chemometric Analyses
- Forest Management and Policy
- Aeolian processes and effects
Bern University of Applied Sciences
2024
University of Augsburg
2024
Julius Kühn-Institut
2024
Technical University of Munich
2024
Bavarian State Research Center for Agriculture
2024
University of Tübingen
2013-2022
Bernstein Center for Computational Neuroscience Tübingen
2019-2022
Leipzig University
2017
ISRIC - World Soil Information
2008
Lincoln University
2005
Tree diversity improves forest productivity Experimental studies in grasslands have shown that the loss of species has negative consequences for ecosystem functioning. Is same true forests? Huang et al. report first results from a large biodiversity experiment subtropical China. The study combines many replicates, realistic tree densities, and plot sizes with wide range richness levels. After 8 years experiment, findings suggest strong positive effects on carbon accumulation. Thus, changing...
Understanding the spatial distribution of soil organic carbon (SOC) content over different climatic regions will enhance our knowledge gains and losses due to change. However, little is known about SOC in contrasting arid sub-humid Iran, whose complex SOC–landscape relationships pose a challenge analysis. Machine learning (ML) models with digital mapping framework can solve such relationships. Current research focusses on ensemble ML increase accuracy prediction. The usual method boosting or...
Abstract We investigated the main parameters [e.g. mean annual air temperature , soil temperature, precipitation, moisture (SM), chemistry, and physics] influencing organic carbon (C org ), total nitrogen (N t ) as well plant available min at 47 sites along a 1200 km transect across high‐altitude low‐latitude permafrost region of central‐eastern Tibetan Plateau. This large‐scale survey allows testing hypothesis that beside commonly used ecological variables, diversity pedogenesis is another...
Summary This study introduces a hybrid spatial modelling framework, which accounts for non‐stationarity, autocorrelation and environmental correlation. A set of geographic spatially autocorrelated Euclidean distance fields (EDF) was used to provide additional relevant predictors the covariates commonly mapping. The approach in combination with machine‐learning methods, so we called method (EDM). provides advantages over other prediction methods that integrate dependence state factor models,...
The aim of our research was to understand small-scale effects topography and soil fertility on tree growth in a forest biodiversity ecosystem functioning (BEF) experiment subtropical SE China. Geomorphometric terrain analyses were carried out at spatial resolution 5×5 m. Soil samples different depth increments data height collected from total 566 plots (667 m2 each). soils analyzed for carbon (soil organic [SOC]), nitrogen, acidity, cation exchange capacity (CEC), exchangeable cations base...
Abstract We compared different methods of multi-scale terrain feature construction and their relative effectiveness for digital soil mapping with a Deep Learning algorithm. The most common approach in DSM is to filter attributes based on neighborhood sizes, however results can be difficult interpret because the affected by outliers. Alternatively, one derive decomposed elevation data, but resulting maps have artefacts rendering undesirable. Here, we introduce ‘mixed scaling’ new method that...
Recent advances in wireless sensor technology allow monitoring of soil moisture dynamics with high temporal resolution at varying spatial scales. The objectives this study were to: (i) develop an efficient strategy for the hillslope scale using a network; and (ii) characterize patterns infer hydrological processes controlling such patterns, method analysis that allows identification relevant within large data sets. We combined pedological expertise geophysical measurements methods from...
Abstract. This study investigated the development of biological soil crusts (biocrusts) in an early successional subtropical forest plantation and their impact on erosion. Within a biodiversity ecosystem functioning experiment southeast China (biodiversity (BEF) China), effect these biocrusts sediment delivery runoff was assessed within micro-scale plots under natural rainfall, biocrust cover surveyed over 5-year period. Results showed that occurred widely experimental developed from initial...
Abstract Most common machine learning (ML) algorithms usually work well on balanced training sets, that is, datasets in which all classes are approximately represented equally. Otherwise, the accuracy estimates may be unreliable and with only a few values often misclassified or neglected. This is known as class imbalance problem do not meet this criterion referred to imbalanced data. of soil are, therefore, One our main objectives compare eight resampling strategies have been developed...
We present a new digital terrain analysis framework for soil mapping, referred to as contextual elevation mapping (ConMap). In contrast common regression approaches based on features from analysis, ConMap is not standard attributes, but differences the centre pixel each in circular neighbourhoods only. These are used random forest regressions. applied and validated by predicting topsoil silt content loess region of 1150 km 2 Rhineland‐Palatinate Hesse, Germany. Three hundred forty‐two...
Abstract While functional diversity ( FD ) has been shown to be positively related a number of ecosystem functions including biomass production, it may have much less pronounced effect than that environmental factors or species‐specific properties. Leaf and wood traits can considered particularly relevant tree growth, as they reflect trade‐off between resources invested into growth persistence. Our study focussed on the degree which early forest was driven by , environment (11 variables...
Abstract Most calibration sampling designs for Digital Soil Mapping (DSM) demarcate spatially distinct sample sites. In practical applications major challenges are often limited field accessibility and the question on how to integrate legacy soil samples cope with usually scarce resources laboratory analysis. The study focuses development application of an efficiency improved DSM design that (1) applies optimized set size, (2) compensates accessibility, (3) enables integration samples....
Predicting taxonomic classes can be challenging with dataset subject to substantial irregularities due the involvement of many surveyors. A data pruning approach was used in present study reduce such source errors by exploring whether different methods, which result subsets a major reference soil groups (RSG) - Plinthosols would lead an increase prediction accuracy minor using Random Forest (RF). This method compared random oversampling approach. Four datasets were used, including entire and...