Bircher, Pascal (2025). Development of a novel soil erosion risk map for arable land in Switzerland. (Thesis). Universität Bern, Bern

Preview

Text 25bircher_p.pdf - Thesis Available under License Creative Commons: Attribution-Noncommercial-No Derivative Works (CC-BY-NC-ND 4.0). Download (26MB) | Preview

Abstract

Soil erosion is a global problem and has a negative impact on food security. Long-term mapping helps to determine the extent of soil loss and to take appropriate mitigation measures. Estimating soil loss is a critical scientific challenge due to limitations in data availability to quantify actual soil loss from erosion. To address this limitation, models have been employed to quantify potential soil erosion risk. The most widespread models are the Universal Soil Loss Equation / Revised Universal Soil Loss Equation -USLE/RUSLE. The USLE/RUSLE models have been expressed in literature as an equation: A = R * K * L*S * C * P, with A representing the estimated average long term soil loss (actual erosion risk) in tons per hectare per year, R the rainfall-runoff erosivity factor, K the soil erodibility factor, L the slope length factor, S the slope steepness factor, C the cover- and management factor and with P the support practice factor. However, not all factors in the USLE/RUSLE models have been tested internationally or adapted for Switzerland. The LS- and C-factor are the most sensitive factors in the (R)USLE-model and require testing of the LS-factor with different flow algorithms and at different spatial resolutions. In particular, the topographical LS-factor in the (R)USLE calculation including different multiple flow algorithms and different resolutions are internationally less known and have also not been tested for Switzerland. Further, problems persist with calibrating and validating erosion models. Often the standard (R)USLE approach without any adaptation to regional conditions is applied in literature. In addition, for cover- and management and support practice factors (CP) only standard values are used in literature and software tools (like a CP-Tool) for calculation of actual soil erosion risk at field levels are rare. Software tools for calculating the CP-factors at field size level are not available, thus the calculation of actual soil erosion risk was not possible for Switzerland. Further, the 2010 developed potential erosion risk map of Switzerland (ERM2 2010) with a resolution of 2 m was available only for agriculturally used area (covering crop- and permanent grassland) of Switzerland, as data on cropland areas was not available to create an erosion risk map only for croplands. The ERM2 2010 was also not calibrated with measured soil loss data and was not corrected for factors specific to Swiss arable landscapes. Consequently, only the standard version of the (R)USLE was applied for Switzerland thus making the ERM2 2010 inadequate as a decision basis for land management and governance in Switzerland. This thesis thus aims to improve the knowledge basis for managing soil erosion in Switzerland by creating a corrected nationwide potential erosion risk map as a tool for decision support and developing software tools for scenario calculations based on adapted CP-factor. To achieve these objectives and to fill the above-mentioned gaps, four sub-objectives were defined: (1) to compare suitable multiple flow algorithms for LS-factor calculation to improve soil loss prediction, and based on this, (2) to compare modelled and measured soil loss, (3) to develop software tools for calculating actual erosion risk, and (4) to subject arable land areas to a correction factor thereby replace the ERM2 2010 of Switzerland with an improved erosion risk map - ERM2 2019 for the arable land. These four sub-objectives resulted into four publications. The first three publications were peer-reviewed and the fourth was a final report to the Swiss Federal Office for Agriculture. In Paper 1, I showed the importance of flow barriers such as roads, railway lines and forests for disturbing or interrupting the flow paths by comparing different multiple flow algorithms on different resolutions (2m vs 25 m). The identified water flow paths show patterns of possible soil loss areas and hints for mitigation measures against soil loss. The LS-factor calculated based on compared multiple flow algorithms was then incorporated into modelling soil loss (Paper 2), used as input for software tools to calculate actual erosion risk (Paper 3) and for developing a new erosion risk map for Switzerland (Paper 4). In Paper 2, I compared modelled and long-term measured soil loss on five representative areas. These five areas located between the northern Prealps and the Jurassic Alps depict the agricultural areas in the Swiss Plateau. The mismatch of the modelled soil loss compared to the measured soil erosion was about the factor 8. This result showed the need to calibrate the modelled soil loss. For the comparison of different CP-Tools in paper 3 a correction factor was applied on the modelled soil loss in paper 4. Thus, in Paper 3, I introduced two software tools to calculate the C-factor, which captures the influence of cropping and management practices and the P-factor on erosion. I then connected this CP-factor with the novel potential erosion risk map of Switzerland 2019 with a Geographical Information System-tool (GIS-tool) programmed in QGIS. The developed software tool for calculating the CP factor uses input data like crop sequences, tillage practices, inter-cropping period, and direction of tillage. Secondly, this tool was coupled with a GIS to obtain the actual erosion risk. The advantage of combining the CP-Tool with the GIS-tool (Paper 3) is that management scenarios can be calculated on selected fields and can provide an impression of different soil erosion mitigation measures. Both software tools (CP-Tool and GIS-tool) allow calculating the actual erosion risk on field block level and comparing different scenarios. This enables the use of the tools for decision support in soil conservation for research and practice such as by the Federal Office of Environment and the Federal Office of Agriculture for regulatory purposes, extension advice and for training purposes. In Paper 4, the potential erosion risk map (ERM2 2019) covers the relatively constant factors of the RUSLE comprising the topographical factor LS, the soil erodibility factor (K) and the rain erosivity factor (R). The sensitive cover- and management and support practice factors (CP) are not included in the ERM2 2019, because the CP-factor highly varies on field block level and within years. A modification on the LS-factor and a conservative reduction factor of 5 was applied on the modelled soil loss for the ERM2 2019. Through these adaptions the ERM2 2019 provides a more realistic assessment of soil erosion risk on arable land than the ERM2 2010, where the standard RUSLE-approach was applied. The results from this research and the development of the software have filled the research gaps relating to application of the RUSLE on arable land. In addition, the study fulfils the objectives to create an upgraded erosion risk map for Switzerland. Some limitations are that since the start of the study, the basic input data for calculating the RUSLE, such as soil maps for K-factor creation and improvements for distinguishing arable and permanent grassland (cadastral survey) have been significantly improved in 2021. This led to an update of the arable land layer in 2022 for the erosion risk map in Switzerland (Federal metadata catalogue) in a followed up project. In future studies, better soil maps may also be available to improve the soil erodibility factor map for Switzerland.

Item Type:	Thesis
Dissertation Type:	Cumulative
Date of Defense:	25 March 2025
Subjects:	500 Science > 550 Earth sciences & geology
Institute / Center:	08 Faculty of Science > Institute of Geography 10 Strategic Research Centers > Centre for Development and Environment (CDE)
Depositing User:	Hammer Igor
Date Deposited:	08 Apr 2025 16:30
Last Modified:	21 Apr 2025 21:46
URI:	https://boristheses.unibe.ch/id/eprint/5984

Actions (login required)

View Item