No abstract

Soil degradation is a major environmental problem in many parts of the world, including Slovenia. The spatially distributed WATEM/SEDEM model can be used to identify the most critical parts of the catchment with regard to soil erosion. Five Slovenian (Central Europe) catchments with inhomogeneous topography, land use, geological conditions, hydro-meteorological properties and sizes (catchment areas between 1 and 2000 km 2 ) were modeled with calibrated parameters, while the WATEM/ SEDEM model was calibrated with an automatic parameter estimation procedure, which is model independent. Both direct and indirect information regarding sediment yields, including turbidity measurements, daily suspended sediment concentration observations and bed load observations, were used for the WATEM/SEDEM model's calibration. A detailed rainfall erosivity (R) factor map, which was constructed from 5-min rainfall data from 31 pluviographic meteorological stations, was used as one of the inputs for the WATEM/SEDEM model. The calculated mean annual soil loss was between 0.3 and 7.4 t/ha/year, and the sediment delivery ratio (SDR) ranged from 0.07 to 0.22 for 5 modeled catchments. The results indicate that the SDR decreases with increasing catchment area; however, the ratio between the average sediment yield and mean soil erosion obviously depends on many other factors, e.g., topography, climatic and geological conditions. The parcel trap efficiency parameter for forests had the greatest influence on the WATEM/SEDEM model's outputs in all five case studies.

Debris flows with different magnitudes can have a large impact on debris fan characteristics such as height or slope. Moreover, knowledge about the impact of random sequences of debris flows of different magnitudes on debris fan properties is sparse in the literature and can be improved using numerical simulations of debris fan formation. Therefore, in this paper we present the results of numerical simulations wherein we investigated the impact of a random sequence of debris flows on torrential fan formation, where the total volume of transported debris was kept constant, but different rheological properties were used. Overall, 62 debris flow events with different magnitudes from 100 m 3 to 20,000 m 3 were selected, and the total volume was approximately 225,000 m 3 . The sequence of these debris flows was randomly generated, and selected debris fan characteristics after the 62 events were compared. For modeling purposes, we applied the Rapid Mass Movement Simulations (RAMMS) software and its debris flow module (RAMMS-DF). The modeling was carried out using (a) real fan topography from an alpine environment (i.e., an actual debris fan in north-west (NW) Slovenia formed by the Suhelj torrent) and (b) an artificial surface with a constant slope. Several RAMMS model parameters were tested. The simulation results confirm that the random sequence of debris flow events has only some minor effects on the fan formation (e.g., slope, maximum height), even when changing debris flow rheological properties in a wide range. After the 62 events, independent of the selected sequence of debris flows, the final fan characteristics were not significantly different from each other. Mann-Whitney (MW) tests and t-tests were used for this purpose, and the selected significance level was 0.05. Moreover, this conclusion applies for artificial and real terrain and for a wide range of tested RAMMS model rheological parameters. Further testing of the RAMMS-DF model in real situations is proposed in order to better understand its applicability and limitations under real conditions for debris flow hazard assessment or the planning of mitigation measures.

No abstract