Freezing is an important factor in soil degradation. In order to predict soil erosion in mountainous areas, the freezing process and the frequency and extent of freezing must be understood. This study aims to identify a relationship between air temperature and a depth of freezing, taking into account soil water content. The soil is considered to be frozen when all the free water within the pores freezes which, according to our numerical model, mainly occurs at temperatures slightly below 0°C. On the other hand, the temperature is not homogeneous in the soil as it is driven by the process of heat diffusion from the soil-atmosphere interface to depth, controlled by soil thermal conductivity and heat capacity, which depend on ice and water contents. As a consequence, the relationship between air temperature and the thickness of the frozen layer is not direct, and the relevance of using air temperature as a measure of frozen depth is to be evaluated.

A small N-S-oriented claystone ridge in an Eastern Pyrenees badland is being monitored. A series of thermometers, water content sensors, and specific heat sensors are collecting data in 5-minute intervals on both sides of the ridge. The data show an attenuation of temperature oscillation with an increasing depth and a time delay of the surface temperature propagation. The differences in the soil temperatures on the north and south side are moreover showing the importance of solar radiation in the process. These observations are further integrated into a procedure allowing for the analysis of possible ad-hoc relationships between current and past air temperature and depth of the frozen layer.

How to cite: Bočková, K., Moya, J., Macías, A., and Vaunat, J.: Field monitoring of cyclic freezing process: effect of air temperature on frozen layer thickness, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-14413, https://doi.org/10.5194/egusphere-egu23-14413, 2023.

The splash phenomenon is one of the stages of water erosion, i.e. a process that causes soil degradation. The crux of splash is the detachment and displacement of soil particles due to the impact of raindrops. Another important aspect of the phenomenon is that a deformation called a crater is formed at the point of the impact. Research on craters helps to expand the knowledge of the splash phenomenon and may contribute to understanding its relationship with successive stages of water erosion. The quantities used to describe the deformation include static dimensions (e.g. depth or diameter of the crater) and dynamic quantities to describe the course of the deformation (e.g. the time for the crater to reach a stable shape). The small size of the craters and the high dynamics of their formation induce the use of accurate and advanced measurement methods.

The aim of the work is to present selected non-contact methods for measuring craters formed on soil after a drop impact.

The study was focused on photography, high-speed imaging, 3D surface modelling, and computed tomography (CT). The discussion of the methods includes a brief description of the measurements, the range of quantities that can be determined with the methods, and indication of their strengths and weaknesses. Knowledge of these issues allows making an informed choice of measurement methods when planning new experiments or projects, which may significantly affect the results that can be obtained.

References

Mazur, R.; Ryżak, M.; Sochan, A.; Beczek, M.; Polakowski, C.; Bieganowski, A. Soil Deformation after Water Drop Impact—A Review of the Measurement Methods. Sensors 2023, 23, 121. https://doi.org/10.3390/s23010121

This work was partially supported by the National Science Centre, Poland within the framework of project no. 2020/37/N/ST10/03363.

How to cite: Mazur, R., Beczek, M., Ryżak, M., Sochan, A., Polakowski, C., Gibała, K., and Bieganowski, A.: Non-contact methods for measuring craters formed on soil after a drop impact, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-5433, https://doi.org/10.5194/egusphere-egu23-5433, 2023.

Soil erosion is a result of detachment and transport of particles or small aggregates from the soil surface. Previous research has predominantly focused on studying the effects of either rainfall or wind on soil erosion processes as separate erosive agents. To date, there have been only few studies into the simultaneous effect of both agents operating at the same time on soil erosion and hydrological processes. In this research, the effect of wind velocities on the erosivity of rainfall was studied, comparing windless rain (WLR) and wind-driven rain (WDR) events when applied to a sandy loam soil. A moderate slope gradient of 11%, a simulated rainfall intensity of 90 mm hr^-1, a rainfall duration of 30 minutes and wind velocities up to 9 m s^-1 were used. The runoff, infiltrate and soil loss (including splash erosion measured at different heights) were compared between the different events and treatments. The soil surface roughness was measured before and after the rainfall event, using a hand-held laser scanner, to evaluate the effect of WLR and WDR on the surface morphology. The outcome of this study shows that, for the smooth surface under WDR, infiltrate volumes were less than under WLR and decreased with increasing wind velocity, while the runoff volumes increased under WDR compared with WLR. The rate of rainsplash erosion increased under the WDR event compared to the WLR event at all heights. The amount of splash-eroded particles decreased with height above the soil surface. We conclude that wind has an effect on the erosivity of rainfall; therefore, it should be considered in erosion studies.

Acknowledgments: This work was funded by OCP group, Morocco

How to cite: Bahddou, S., Otten, W., Whalley, R., Shin, H.-C., El Gharous, M., and Rickson, J.: Effect of wind-driven rain on runoff, infiltration and soil erosion, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-8245, https://doi.org/10.5194/egusphere-egu23-8245, 2023.

Land degradation is a global topic in climate change debates resulted from different types of human activities as well as from physical processes. Resilient, healthy soils are important to help reduce the ecological and economic impact of environmental change and extreme conditions. The development of adequate and broadly applicable indicators and thresholds is challenged by the great diversity of European soils and climate, as well as different political, economic, and social conditions which lead to different priority settings for targets and indicators. This work built upon current environmental awareness (e.g CAP, SDGs, etc.) to design a methodological framework for environmental performance metrics related to land degradation. The framework was oriented towards a data-driven environmental metric approach leveraging Copernicus Sentinel-2, existing open-access databases such as LUCAS (Land Use/Cover Area frame statistical Survey) and GEOSS (e.g., Soil Grids) vast dataset archives to provide metrics for environmental actors. Based both on the international literature and European commission documentations this work is focused on the combination of vital importance indicators of soil degradation and soil health. A novel deep learning architecture was implemented to support the final knowledge extraction with a pixel-based spatial resolution of 10m for the determination of Soil Organic Carbon (SOC) and Clay content. The above indicators are used as enhanced geospatial inputs to a soil erosion modelling approach providing improved predictions. A proper approach was followed for the SOC:clay ratio generation and with the soil erosion product combination to provide an ambitious land degradation index. An agricultural area in Northern Greece was used as a demonstration test site area for the proposed methodology.

How to cite: Samarinas, N., Tziolas, N., and Zalidis, G.: Assess land degradation status based on Earth Observation driven proxy indicator, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-14553, https://doi.org/10.5194/egusphere-egu23-14553, 2023.

Healthy soil can be defined as a dynamic ecosystem that performs a variety of essential functions such as controlling plant disease, nutrient cycling, improving soil function with positive effects for filtering and storing water, and nutrient capacity, and contributing to improving crop production. Healthy soil also contributes to mitigating climate change by maintaining or increasing its carbon content. Therefore, information on spatial variability of key soil properties is essential for prioritizing and tracking land management interventions, from the small-scale farm level to the global landscape level. In addition, healthy soil is critical for achieving several SDGs such as #2, 3, 6, 13, 15 and 17. However, the determination of soil properties through wet chemistry measurements is often expensive and time-consuming process, and consequently, soil analyses are restricted to a limited number of soil samples. A method that predicts the soil properties fast, inexpensively, and accurately is soil spectroscopy, which can provide immense opportunities for monitoring important soil health indexes. In this study, the evaluation of three algorithms for predicting three key soil properties, soil organic carbon (SOC), pH and Magnesium (Mg) using mid-infrared spectral data were studied using a dataset of more than 3400 samples. The soil samples were collected across Sub-Saharan Africa (SSA) region using the well-established Land Degradation Surveillance Framework (LDSF) method developed from World Agroforestry (ICRAF). The developed trained calibration models were based on the widely used in the soil spectroscopy research Partial Least Squares (PLS) and Random Forest (RF), and the not applied so far Bayesian Regularization for Feed-Forward Neural Networks (BRNN). The dataset was split into calibration (70%) and validation (30%) sets. Furthermore, the threshold of 5% was applied and thus, only the data with value that lie between 5% and 95% of each soil property were included. In this way, the extreme values that will bias the model and the predictions were excluded. Results has shown that the calibration model developed based on the BRNN algorithm yielded the more robust predictions among the three studied soil properties (R² of val 0.90, 0.92, 0.87 for pH, Mg, SOC, respectively). The predictions utilizing soil spectroscopy for determining soil properties in this study are showing its extremely potential to be beneficial in the support of soil health.

How to cite: Pittaki, Z., Vagen, T.-G., Karari, V., Weullow, E., Ateku, D. A., and Winowiecki, L. A.: Towards a more sustainable global soil health monitoring through soil spectroscopy, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-15335, https://doi.org/10.5194/egusphere-egu23-15335, 2023.

Soil erosion and associated land degradation, accelerated by anthropogenic activities, are serious worldwide environmental problems that can reduce crop productivity and cause on- and off-site damages resulting from sediment transport and deposition. A common feature of the agriculture-dominated landscape of the Prairie Pothole Region in western Canada is depressional wetlands (typically 10 to 20, small seasonal to permanent waterbodies per square kilometer).  These wetlandscapes are fragile agro-ecosystems that have been disturbed and impacted through human-induced soil erosion, receiving influxes of sediment and associated constituents (e.g., phosphorus, organic carbon and pesticides) from the surrounding local and regional sources (e.g., cultivated fields). The general objective of this study was to estimate the severity of soil loss and sedimentation rates, and sediment flux from surrounding cultivated fields into wetlands. The study area included two sub-watersheds in the Prairie Provinces of Manitoba and Alberta, within the Canadian portion of the Prairie Pothole Region. Soil and sediment cores were collected along multiple transects from the uppermost portion of each catchment to the central area of the wetland.  Transects were distributed within catchments to capture the variability resulting from the topographic complexity. The results of this study demonstrated high rates of soil loss in the upper slope and middle slope positions, and high rates of deposition in lower slope and foot slope positions of wetlandscapes. Furthermore, the areal average of transect data provided sediment delivery ratio estimates of 57% for Manitoba and 35% for Alberta, which indicates that a relatively high amount of the mobilized sediment was exported beyond the cultivated fields towards the wetlands, mostly deposited in riparian areas.

How to cite: Zarrinabadi, E. and Lobb, D. A.: Assessing soil erosion and sediment flux using 137Cs within topographically complex landscapes of Prairie Pothole Region of Canada, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-9841, https://doi.org/10.5194/egusphere-egu23-9841, 2023.

Global efforts to restore the world’s degraded croplands require knowledge on the degree and extent of accelerated soil organic carbon (SOC) loss induced by soil erosion. However, methods for assessing where and to what extent erosion takes place is still inadequate for precise detection of erosion hotspots at high spatial resolution.

In this study, we attempted to develop a spectra-based soil erosion mapping approach to pinpoint eroded hotspots in a typical catchment located in the black soil region of Northeast China as characterized by undulating landscapes. We built a ground-truth dataset consisting of three classes of soils representing Severe, Moderate, and Low erosion intensity because of their inter-class contrasts in estimated erosion rates from ¹³⁷Cs tracing. The spectral separability of different erosion classes was first tested by a combined principal component and linear discriminant analysis (PCA-LDA) against laboratory hyperspectral data and then validated against Sentinel-2 derived broadband spectra.

We will present results on the performance of the PCA-LDA model to classify soil erosion intensity classes based on laboratory and satellite-based soil spectra. We further identified distinctive spectral features representative of shifting soil albedo and biochemical composition due to erosion-induced SOC mobilization. A classification scheme comprising the spectral features was applied to the Sentinel-2 bare soil composite for pixel-wise soil erosion mapping, from which 15.9% of the catchment area was detected as erosion hotspots while the Moderate class occupied 65.4%.Our study highlights the potential of the spectral-based remote sensing approach for  better targeted cropland management to combat soil degradation.

How to cite: Qi, L. and Shi, P.: Detection of soil erosion hotspots with accelerated carbon losses in the black soil region of Northeast China as driven by Sentinel-2 multispectral remote sensing, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-4089, https://doi.org/10.5194/egusphere-egu23-4089, 2023.

Vegetative barriers have proven their effectiveness in controlling erosion and promoting other ecosystem services in agricultural areas. This has led to their conservation and promotion as landscape elements by the Common Agricultural Policy and other policy initiatives. However, predicting their efficiency in reducing hydrologic connectivity presents a large uncertainty (Gumiere et al., 2011).

This communication presents an analysis of trapping efficiency of sediment, runoff, and nutrients (P and N) by vegetative barriers aimed to provide a statistical approach to efficiency, based on a review of available studies, considering two climates: humid and arid/semi-arid (Muñoz et al, 2022). For this, different independent variables were grouped and identified to explore its influence: i) those defining the vegetative barrier dimension (width, slope of the plot, and area ratio buffer/plot) and ii) those related to experimental conditions (type of vegetation, soil protection of the non-buffered area, type of climate, type of experimental measurement and origin of the rainfall). A more detailed analysis was performed with the studies which reproduce similar situations to the ones occurring naturally (natural rainfall and paired plots).

In general, average trapping efficiencies for runoff and sediment are 40.1 and 62.6% and ranging between -81.0 to 99.9% and -109 to 100%, respectively. For nutrients, values of trapping efficiencies had an average of 44.9 and 38.4% for phosphorus and nitrogen, respectively. The lack of data and the large variability among and within the measurements from the studies considered in our review only allowed to detect slight trends and statistically significant differences in some cases for the different variables analysed.

In order to provide guidelines to farmers and technicians, a probabilistic model was developed for sediment trapping efficiency regarding the width of the vegetative barrier and the climatic region. The model showed that in 92% of the cases, a vegetative barrier will reduce erosion in humid climates while this trapping efficiency will be 100% in semi-arid/arid conditions. Grouping the vegetative barriers’ width in different intervals, it was observed that the maximum trapping efficiency is ~80 % for a width of 2.75 to 3-m in arid/semi-arid climate and 5 to 6-m in humid regions.

Acknowledgement: This work supported by the project PID2019-105793RB-I00 financed by the Spanish Ministry of Science and Innovation, and project TUdi, GA 101000224, of the European Union’s Horizon 2020 research and innovation programme

References

Gumiere, S. J., Le Bissonnais, Y., Raclot, D., & Cheviron, B. (2011). Vegetated filter effects on sedimentological connectivity of agricultural catchments in erosion modelling: A review. Earth Surface Processes and Landforms, 36(1), 3–19. https://doi.org/10.1002/esp.2042.

Muñoz, J. A., Guzmán G., Soriano M.A., & Gómez J. A. (2022). Trapping efficiency of vegetative barriers in agricultural landscapes. An operational model from a review of available information. Manuscript submitted for publication.

How to cite: Muñoz, J. A., Guzmán, G., Soriano, M. A., and Gómez, J. A.: Trapping efficiency of vegetative barriers in agricultural landscapes. An operational model from a review of available information, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-477, https://doi.org/10.5194/egusphere-egu23-477, 2023.

Tunisia is an agriculture country. Such as many other Mediterranean countries precipitations  remains a decisive factor not only for the different agricultural uses of lands (rainfed or irrigated system) but also for the soils erosion. The latter  is accentuated by agricultural practices (tillage, pesticide inputs, low crop residue restitution…) which are often productive but do not protect natural resources. All these factors have led to the development of conservation agriculture based on no-tillage as a mean to combat soil erosion. In fact, in Tunisia, no-tillage areas increased from 52 ha in 1999 to 17000 ha in 2020. Based a set of 20 plots covering 6 soils types and located in the semi-arid area of the country, a periodic monotoing of a set of soil parameters were done during three years, which include soil sensitivity to erosion according Le Bissonnais method, soil organic matter content and soil microbial respiration. For each plot, a half  of the surface was no-tilled and the other half was conducted according the conventional method based on soil returning. The results show a rapid effect of no-tillage on soil erosion, since an improvement of 18%, 42% and 39% of soil resistance to erosion, respectively after the first, second and third year of switching to no-tillage system. The soil microbial activity response was also significative, whit a progressive increase of soil respiration in the no-tilled treatments compared to tilled treatments. The microbial respiration was higher in non-hydromorphic soils and particularity in the red Mediterranean soils, calcic-magnesic soils and isohumic soils were moisture conditions was the most favorable for a development of soil microorganisms. For soil organic matter content, the evolution trend was not detectable in relation to the slow evolution of this soil parameter. However, the evolution of particulate organic matter content (a part visible of soil organic matter, with a size larger than 2mm) shown an increase in no-tilled treatments comparative to the tilled treatments. The increase of the particulate organic matter content was more important in vertisols, podzol, calcic-magnesic soils and isohumic soils in relation with their higher wheat production compared to other studier soil types.

How to cite: Annabi, M., Bahri, H., Cheikh M'hamed, H., Barbouchi, M., and Toukebri, W.: Conservation Agriculture to Rebuilt Soil Fertility in Northern Tunisia, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-16736, https://doi.org/10.5194/egusphere-egu23-16736, 2023.

Gully erosion is regarded as one of the worst land degradation processes in the world. It is a complex geomorphic process resulting in on-site land degradation due to the removal of soil, and off-site land degradation, due to mobilised soil, pollutants, and pesticides adversely affecting soil and water resources. Climate change predictions show that the frequency of high-magnitude rainfall events will increase, thereby exacerbating the degradation impacts of gullying. By assessing long-term datasets (>30 years), the relationship between gully evolution and climate variability can be observed, providing potential insight into gully evolution under climate change. We aim to isolate climate as a driving factor by investigating three sites under similar land-use, environments transformed into conservation areas but exhibiting contrasting climates. The sites are in South Africa (SA), located in the arid Karoo, a Lowveld area with Savanna land cover and a humid Grassland region. A triangulation of methods was implemented, including i) digitising gully change from remotely sensed imagery spanning up to 78 years, ii) conducting field measurements, and iii) interviewing land managers. In the arid Karoo, gullies were discontinuous, forming successive chains waning into deposition zones, only to be reactivated again with an abrupt headcut downstream of the depositional area. Continuous gullies were found in the Grassland and Savanna regions. Remotely sensed image analysis shows a mean annual linear expansion along the main drainage line was the largest in the Savanna, 2.8 m, compared to 0.5 m and 0.4 m recorded in the Grassland and Karoo. However, the mean planimetric area changes of gullied extent were 6 and 8.4 times larger in the Grasslands compared to the Karoo and Savanna. Fieldwork confirmed active gully processes at all three sites, identifying active sub-surface processes, only found in the Grassland, as the leading cause for the morphological difference between the extensive Grassland gullies and the narrower linear gully features in the Savanna and Karoo. However, certain gully headcuts in the Karoo present as bulbous transforming into narrower downstream channels, most likely a result of being artificially enlarged from failed gabion installations at the headcut. During interviews with land managers, concerns about contemporary gully erosion were noted at all three sites, with the implementation of mitigation measures ongoing. There is consensus among the three methodologies, identifying that gully erosion remains stubbornly active, even after transforming to a conservation-orientated land-use. In the Karoo and Savanna, human influence and rainfall variability were attributed as causal factors of gullying, inherited from ill-placed farm roads and overgrazing, respectively. In the Grasslands, soil type is deemed the dominant driver of gullying, although animal tracks (specifically those from antelope) forming pathways for water after rainfall was contentious. Our analysis shows that gully erosion severity follows the climate gradient of SA. However, it remains difficult to completely isolate climate as a driver due to different inherent soil properties, geology, and slope. We expect an increase in gully erosion for all three sites under climate change predictions, although more pronounced in the Grassland region.

How to cite: Olivier, G., Van De Wiel, M. J., and De Clercq, W. P.: Historical evolution of gullies: Impact of climate, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-10002, https://doi.org/10.5194/egusphere-egu23-10002, 2023.

Mediterranean ecosystems are already under pressure from the combined impacts of direct human activities and anthropogenic climate change. In this highly human-managed region, land degradation and accelerated dryland expansion threaten the biological systems and the natural resources that sustain agriculture and forests [1]. The question of "what is the driving force of the changes," and specifically whether anthropogenic climate change or direct human management is to blame, is crucial. The attribution of these changes plays a key role in better managing the arid and semiarid Mediterranean agro-ecosystems, especially considering a different future climate. Here we examine the net effect of anthropogenic climate change-induced changes in the state of two significant land degradation factors, aridity and rainfall erosivity, across the Mediterranean.

We rely on a state-of-the-art counterfactual stationary climate dataset, a hypothetical climate in a world without climate change [2], to estimate impact indicators since the onset of industrial revolution (1900-2019). We compare results against aridity and rainfall erosivity calculated from observational (factual) climate data to examine historical imposed long-term changes attributed to human induced climate change. The difference between past impacts and counterfactual impacts is a proxy for a broad-scale concept facilitating a top down approach to define degradation and physical vulnerabilities of Mediterranean agro-ecosystems, in the frame of REACT4MED project. We further analyze scenarios of future climate change to unravel future trends. The information is further downscaled at pilot area level to support the co-development and analysis of local-scale indicators. Results will serve as a basis for the detection of potential future degradation trends used in the identification of potentially suitable areas in the Mediterranean for up or out-scaling of restoration measures beyond REACT4MED pilot areas.

[1] Daliakopoulos, N., et al. "Yield response of Mediterranean rangelands under a changing climate." Land Degradation & Development 28.7 (2017): 1962-1972.

[2] Mengel, M., et al. “ATTRICI v1.1 – counterfactual climate for impact attribution.” Geosci. Model Dev., 14, 5269–5284.

Acknowledgements: This work has received support from REACT4MED (GA 2122) PRIMA funded project, supported by Horizon 2020.

How to cite: Koutroulis, A., Grillakis, M., and Tsilimigkras, A.: Historical evolution and future storylines of degradation drivers in Mediterranean arid and semiarid agro-ecosystems, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-8611, https://doi.org/10.5194/egusphere-egu23-8611, 2023.

Gully is an issue of the climate crisis in most coastal areas of the world. More than 70% of the world's sandy coastlines would be eroded in recent decades. Anthropogenic climate change is one of key factors controlling soil erosion occurrence. Soil erosion control is extremely important for soil conservation work and territorial planning. The municipality of Nacala, province of Nampula, Mozambique, has gully susceptibility as a known problem. The Japan International Cooperation Agency (JICA), concerned about the intense urban erosion in the Nacala region, which causes problems not only for the population living in critical areas but also for the proper functioning of the Nacala Port, has been making efforts to identify the causes and mechanisms of the phenomenon and suggest corrective and preventive measures covering the area of the Mocone basin, Nacala. Thus, we obtained information on the causes and mechanisms of evolution of these phenomena to allow the joint search for the features that determine the natural susceptibility to erosion of the municipal territory. We also identified that the soil type in the Mocone basin is sandy-clayey, with an infiltration capacity of 50mm/hour, erodible, with a drainage network mainly characterized by rills and gullies, with periodic surface runoff, only during heavy rains. Our studies indicated a catchment with a significant area upstream of the erosion process (2.34 km²). The maximum project flows obtained in the our preliminary hydro-meteorological studies, considering intense rains with a return period of 100 years, and with duration equal to the time of concentration of the catchment basin (90 min), also proved to be quite significant, with values above of 25 m³/s. Accelerated erosion in this location seems to occur, preferentially, in sectors where those natural factors have been exacerbated by anthropic factors due to inadequate occupation of the physical environment. This urbanization resulted in waterproofing, an increase in surface runoff volumes and a reduction in the concentration time of this basin. In addition, the rainwater drainage of some streets was released directly into the natural terrain, without the proper hydraulic works to reduce energy. The rapid development of the erosion branch towards the Nacala Port proves this close relationship between inadequate urban drainage and the evolution of the analyzed erosion process. The results obtained also prove, once again, the importance of analysis and mitigation approaches for large scale erosion processes that contemplate the catchment basin in its entirety, where erosion control practices or changes in soil cover are necessary.

How to cite: Ferreira Lima, I., Kobayashi, C., da Silva, B., So, Y., and Hitomi, Y.: Influence of anthropogenic climate change on soil erosion occurrence: Gully Nacala, Mozambique, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-16894, https://doi.org/10.5194/egusphere-egu23-16894, 2023.

As part of the earth's dynamic systems, badland landscapes are chiefly generated by the interaction between bedrock weathering, climate seasonality, and the controversial contribution of the hillslope and fluvial erosion processes. Although contemporary definitions clearly point out that badland initiation is a function of overland-flow-dominated gully channels, gravitational processes-dominated mass movements, and subsurface processes-driven piping, the following related questions remain to be tackled regarding their topographic position in the landscape: (1) Are the badland landforms signature of hillslope or fluvial erosion domain? (2) How do climate and regional settings (i.e., macro landforms, homogenous or heterogenous lithologies, etc.) influence the topographic position of badlands? To address these questions, we have selected climatically distinct badlands from different continents where; digital elevation models are available: Utah and South Dakota (USA), Upper Llobregat and Murcia (Spain), Northern Apennines and Basilicata (Italy), Mediterranean and Plateau – Margin Transition Badlands (Turkey), and Southern Taiwan. The badland boundaries were visually inspected and manually digitized based on diagnostic morphologic indicators. We have utilized the slope and drainage area relationship by using a 5m digital terrain model to identify topographic thresholds at TopoToolbox, a MATLAB-based software for topographic analysis. Preliminary results show that most badland areas occupy a maximum of 10⁴ to 10⁵ m² contribution area in the landscapes. Although contribution areas relatively represent uniform thresholds in the sites, the local gradient (S), which is proportional to the contribution area, tends to be higher in the sub-humid mountainous badlands (Upper Llobregat, Northern Apennines) and wet tropical SW Taiwan than in semi-arid badlands (Basilicata, Murcia). We conclude that the topographic signature of badlands in the context of the sub-catchment scale may depict an appropriate instance of a transitional domain from a diffusive erosional process to a fluvial erosion process. Our findings may serve as a foundation for a better understanding of the classification and automatic detection of badland landscapes, also known as erosional hot spots.

This study has been produced benefiting from the 2232 International Fellowship for Outstanding Researchers Program of the Scientific and Technological Research Council of Turkey (TUBITAK) through grant 118C329 and TUBITAK 2214-A International Research Fellowship Programme.

How to cite: Avcioglu, A., Schwanghart, W., Görüm, T., Yetemen, Ö., Moreno-de las Heras, M., and Yang, C. J.: Topographic signature of badlands landscapes, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-646, https://doi.org/10.5194/egusphere-egu23-646, 2023.

The models currently used to predict post-fire soil erosion risks are limited by high data demands and long computation times. An alternative is to map the potential hydrological and sediment connectivity using indices to express the general properties of the landscape under evaluation and map the possible connectivity between the different parts of a catchment.

In this study, we aim to answer the question: Do these alternative approaches identify post-fire sediment mobilization hotspots?  To achieve this, we assess the spatial variability distribution of the location of soil erosion hotspots using the Index of Connectivity (IC), Revised Universal Soil Loss Equation (RUSLE model) and the Sediment Export (SE) and compare it to the simulation results of a more complex Landscape Evolution Model (LAPSUS model). Additionally, we evaluate statistical measures of association between the four tools. Furthermore, IC, RUSLE model and SE are used due to their simplistic representation of erosion and ease of application, and the LAPSUS model is used as the best representation of erosion and sediment transport in the study area.

Our results show that the three tools (IC, RUSLE model and SE) tested in this study are suitable for identifying sediment mobilization hotspots, i.e., areas where the erosion rates are above the 90^th percentile, in recently burnt areas, and differences between their performance are minor. These findings can be considered for post-fire and water contamination risk management, especially for fast prioritization of areas needing emergency post-fire intervention.

How to cite: Parente, J., Nunes, J., Baartman, J., and Föllmi, D.: Testing simple approaches to map sediment mobilization hotspots after wildfires, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-261, https://doi.org/10.5194/egusphere-egu23-261, 2023.

A better understanding of soil erosion is not possible without including subsurface erosion. Soil piping may significantly contribute to the overall erosion problem in a given area and may therefore change the conditions and methods for controlling soil degradation. Therefore, there is an urgent need to identify regionally and globally sites where soil piping occurs which then may require a change of the strategies to control soil erosion. In this project, we are constructing the very first data-driven piping erosion susceptibility map of Europe. The crucial point is to identify piping-affected areas by mapping the soil piping-related features, i.e. pipe roof collapses (PCs) and pipe outlets in the European Union and the UK. Mapping is based on an in-depth literature review in combination with detailed mapping using Google Earth imagery, and LiDAR data (if available). The database currently consists of 6841 piping-related features (6171 PCs, and 670 outlets), among which the location of 88% features is certain (within a resolution of 25 m). Almost 28% (1889 features) were located based on detailed fieldwork, 25% (1726) were extracted from published papers, and 47% based on a detailed analysis of Google Earth imagery and LiDAR data (19% and 28%, respectively). This database is currently used to construct the very first data-driven piping erosion susceptibility map of Europe.

This research is part of the project “Building excellence in research of human-environmental systems with geospatial and Earth observation technologies” that received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 952327.

How to cite: Bernatek-Jakiel, A., Vanmaercke, M., Poesen, J., Biernacka, A., Borrelli, P., Derii, A., Hałys, J., Holden, J., Jakab, G., Panagos, P., Piątek, D., Regensburg, T. H., Rodzik, J., Nadal-Romero, E., Stolarczyk, M., Wacławczyk, P., and Zgłobicki, W.: A European database of soil piping-related features: a major step towards producing a piping erosion susceptibility map of Europe, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-1467, https://doi.org/10.5194/egusphere-egu23-1467, 2023.

Soil erosion (SE) and saturated hydraulic conductivity (Ksat) are the essential indicators for land degradation and soil nutrient loss. Both processes are linked with each other as one can enhance the other resulting in positive feedback loops. The increase in soil erosion might block the macropores and reduce Ksat. In contrast, the reduction in Ksat triggers the runoff and increases soil erosion. However, modelling and predicting soil erosion, hydraulic properties are usually not adequately considered. We tried to link soil erosion directly with Ksat on a global scale. For this, we used global soil erosion and saturated hydraulic conductivity maps to yield combined soil risk classes. SE and Ksat maps were obtained from Borrelli et al. (2017) and Gupta et al. (2021), respectively. Each map was classified into six classes based on the previous studies. Moreover, both maps' classes were combined by creating six merged classes to develop the modified soil risk map. The modified soil risk map highlights regions with low Ksat and high SE. Furthermore, the final map was validated using continental and/or national sediment yield (SY) datasets. SY classes were compared with final risk classes to validate the accuracy of the map. The modified soil risk map showed higher accuracy compared to the SE map when compared with SY classes. This study demonstrates the first attempt to link soil erosion to soil hydraulic properties on large scales.

How to cite: Gupta, S., Borrelli, P., Panagos, P., and Alewell, C.: Modified global soil risk map using soil erosion and saturated hydraulic conductivity maps, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-1892, https://doi.org/10.5194/egusphere-egu23-1892, 2023.

Soil erosion represents a serious challenge for agricultural production and for the environment. Soil erosion impacts, such as reduction of fertile soil, alteration of the carbon cycle and pollution and eutrophication of water bodies, represent a significant management concern for the European Union. Modelling can help define efficient and targeted mitigation strategies by identifying the long-term controlling factors and the areas where, and periods during which, soil is at high risk of erosion. However, to define such strategies, there remains a lack of modelling approaches a) able to provide with longer term baseline information which to measure the success or otherwise of mitigation strategies at the catchment scale and b) accessible and robust enough to be used, understood and trusted by users with more or less expertise, including researchers, land managers and policy makers. In response, this project will improve the robustness and accessibility of quantitative methods for supporting agricultural land management. The objectives of this project are: (i) to develop an accessible soil erosion model, iMPACt-erosion, based on interactive Jupyter Notebooks, to support agricultural land management at the catchment scale, (ii) to apply a robust model evaluation based on the use of both long-term soil loss observations and global sensitivity analysis to achieve greater confidence in model predictions and (iii) to identify the soil erosion controlling processes and vulnerable areas and periods to define targeted and effective mitigation strategies until 2100. We present the model and the first model evaluation results, which show that simulations are consistent with observed soil loss rates (estimated from both fallout radionuclides and tree mound measurements) in the last 60 years in olive orchard catchments in South Spain.

How to cite: Peñuela, A. and Vanwalleghem, T.: iMPACt-erosion: A robust and accessible model for a long-term effective management of agricultural soil erosion, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-13611, https://doi.org/10.5194/egusphere-egu23-13611, 2023.

The (R)USLE and its many derivates are the most used models to estimate soil loss by water at regional scale. Within the (R)USLE the C-factor is used to assess the impact of crop and soil management on erosion. The exact, temporal and spatial explicit determination of the C-factor is crucial to create reliable maps on soil loss estimates, monitoring the impact of agricultural practices on soil erosion and evaluating the efficiency of policy instruments.

The derivation of accurate high-resolution C-factor data for larger regions remains a challenge, as a variety of spatio-temporal data (crops/crop rotation, tillage management, interannual variation of rainfall erosivity) is needed. Based on recent improvements in calculation methods and the availability of earth observation data, the C-factor can be estimated using agricultural statistics and parcel-based information on crop cultivation. Consequently, we developed two types of spatially explicit and regional C-factor datasets for Germany, mapping the impact of agricultural practice on soil erosion in a more reliable and spatially detailed way: i) harmonized mean C-factors for municipalities for 1999, 2003, 2007, 2010, 2016 and 2020, and ii) parcel-based C-factors for the period 2017-2020 based on crop type maps from satellite data.

In this contribution, we will present the implemented methods and will discuss the impact of the new detailed C-factor maps on soil loss estimates in a spatial resolution of 10 x 10 m for German cropland. The results show an increase in the C-factors for municipalities during the last two decades by 17 %. The parcel-based data reveals the C-factor ranges within the municipalities: While the mean C-factor for all parcels across Germany is 0.107 (conventional cropping; 2017-2020), the standard deviation within the municipalities ranges from 0 to 0.141 at a mean standard deviation of 0.045. This results in differences in erosion rates of up to 7 t / (ha • a), highlighting the importance of spatially explicit C-factor estimates.

How to cite: Steinhoff-Knopp, B. and Saggau, P.: The impact of new detailed (R)USLE-C-factor maps for Germany on soil loss estimates, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-9334, https://doi.org/10.5194/egusphere-egu23-9334, 2023.

In Switzerland, soil erosion by water is a challenge due to the complex topography. It is stipulated by law that a soil removal rate of 2 t/ha/yr (for soil depth < 70 cm) or 4 t/ha/yr (for soil depth > 70 cm) on arable land must not be exceeded as a long-term average. Furthermore, an impairment of soil fertility and of water bodies and near-natural habitats due to erosion should be avoided by all means.

In this study, we calculated the actual erosion risk on arable land of Switzerland for 2021 using newly available data sources on crops and tillage practices. The erosion risk was calculated with the Revised Universal Soil Loss Equation:

A = R∗K∗L∗S∗C∗P

With R = rainfall and surface runoff factor, K = soil erodibility factor, L = slope length factor,  S = slope gradient factor,  C = soil cover and cultivation factor, and P = erosion control factor.

The product of the factors R∗K∗L∗S represents the potential erosion risk. For the calculation, we used the available potential erosion risk map by Bircher et al. (2019).

To derive realistic C factors adapted to Swiss climate conditions, the following procedure was applied: average crop-specific C  factors for two geographical regions (valley, mountain) and three distinct tillage types (plough mulch, direct seeding) were deduced from data of the Swiss agri-environmental monitoring programme (SAEDN; Gilgen et al. 2023), which provides detailed management data on the field level of approximately 300 farms since 2009. The categorised C  factors were allocated to individual parcels using geo-referenced crop area polygons of the Swiss cantons as well as direct payment data from the Federal Office for Agriculture on crop-specific tillage types.

Since no data on the P factor was available, it was estimated using expert knowledge. Calculation was carried out on a 2x2m grid and summarised at municipality level.

Our calculations represent the best yet for Switzerland's actual erosion risk. However, a comparison with the erosion that has actually occurred, as carried out by Bircher et al. (2022), is still pending. The described method can be used in the future for monitoring that focuses on the environmental impact of farmers' management changes. A prerequisite for this is that sufficiently detailed data on farm management continues to be available.

References:

Bircher, P., Liniger, H.P., and Prasuhn, V. (2019): Aktualisierung und Optimierung der Erosionsrisikokarte (ERK2): Die neue ERK2 (2019) für das Ackerland der Schweiz: Schlussbericht. Bundesamt für Landwirtschaft, Bern 

Bircher, P., Liniger, H.P., and Prasuhn, V. (2022): Comparison of long-term field-measured and RUSLE-based modelled soil loss in Switzerland. Geoderma Regional, 31, e00595, https://doi.org/10.1016/j.geodrs.2022.e00595

Gilgen, A., Blaser, S., Schneuwly, J., Liebisch, F., and Merbold, L. (2023): The Swiss agri-environmental data network (SAEDN): Description and critical review of the dataset. Agricultural Systems, 205, 103576, https://doi.org/10.1016/j.agsy.2022.103576

How to cite: Gilgen, A., Blaser, S., Schneuwly, J., Hutchings, C., and Prasuhn, V.: Modelling Switzerland’s actual erosion risk on arable land, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-6421, https://doi.org/10.5194/egusphere-egu23-6421, 2023.

Soil erosion has always been a threat to the environment and agricultural practices throughout the world. For a country like India, where agriculture contributes primarily to its economy, it becomes a major problem. Identifying these vulnerable regions and planning for mitigation is crucial for sustainable soil resource management. For this, mapping or modeling soil erosion at a national scale is required to understand the variability of soil losses throughout the country. Performing field-based experiments to estimate soil losses over a large country is always expensive and tedious. Revised Soil Loss Equation (RUSLE), an empirical model, has been more prominent worldwide due to its simplicity and less forcing data requirements. In this study, average annual Potential Soil Erosion (PSE) was estimated over India using IRED (Indian Rainfall Erosivity Dataset), ISED (Indian Soil Erodibility Dataset), SRTM (Shuttle Radar Topographic Mission), DEM (Digital Elevation Model), and LULC (Land-use/Land-cover) obtained from NRSC (National Remote Sensing Institute) India. PSE was further analyzed using LULC categories and soil types to visualize the impact of soil erosion in each class. As erosion significantly affected agricultural activities, financial losses over the nation were also estimated, considering the severity of soil erosion. Further, Sediment Delivery Ratio (SDR) and Specific Sediment Yield (SSY) were also mapped over the national boundary to visualize the actual soil displacement at a grid scale of 250 m. Using the SSY map, the incoming sediment load to reservoirs/dams/lakes was also estimated considering its watershed areas. This study will be helpful for the experts in the field of sustainable soil resources management for planning mitigation measurements against soil losses in India.

Keywords: Soil Erosion, Sediment Yield, Sediment Delivery Ratio, India, RUSLE, LULC

How to cite: Raj, R., Saharia, M., and Chakma, S.: National Scale Soil Erosion and Sediment Yield Assessment over India, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-883, https://doi.org/10.5194/egusphere-egu23-883, 2023.

The crop cover-management (C-) factor in arable landscapes describes the soil erosion susceptibility associated with seasonally cultivated crops. Previous informatic and computational limitations have led many modelling studies to prescribe C-factor values and assume spatial and temporal stationarity. However, the multiple influencing factors ranging from parcel-scale crop cultivation and management to regional-scale rainfall regimes motivate new methods to capture this variation when identifying at-risk areas. Modern data systems (in this case: field parcel vector data, Sentinel-2 time series, rainfall erosivity time series networks, EU-scale land survey (LUCAS) and European regional statistical data) provide spatially and/or temporally dense information sources on which scalable model parametrisation frameworks can be built and updated. Here, we define a multi-component method to derive the C-factor by associating time series of canopy and residue surface cover from Sentinel-2 and climate-specific rainfall erosivity with Integrated Administration and Control System (IACS) field parcel data from European Union Member States. A standardised approach is emphasised to increase the future interoperability and inter-comparability of soil erosion modelling studies deploying the C-factor. Additionally, field parcel simulation units with associated crop declarations provide a new reference scale to link predictions of soil erosion risk with specific management decisions and declarations by farmers. After implementing the method on a homogenised subsample of 8600 field parcels covering available IACS data regions, several key findings are outlined: 1) time series information provides new opportunities to predict the time-criticality of erosion in specific crop cultivations, 2) the varying (a-)synchronicity between seasonal crop canopy cover and heavy rainstorms means that spatial variability is inherent within the C-factor across Europe, and 3) the addition of agricultural management practices (e.g. tillage practice descriptions) to open-access IACS repositories can facilitate more comprehensive evaluations of the C-factor and soil erosion risk in the future.

How to cite: Matthews, F., Panagos, P., Borrelli, P., and Verstraeten, G.: A crop phenology-based approach to quantify the C-factor at the field-parcel scale in Europe, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-8889, https://doi.org/10.5194/egusphere-egu23-8889, 2023.

Soil erosion by water is considered as one of the most serious processes degrading the soil functionality of arable soils around the globe. To assess the risk of water erosion at larger scales, the universal soil loss equation (USLE) and its derivates are commonly used for the implementation of policy instruments.

An important factor needed within (R)USLE model exercises is the K-factor, which reflects the natural erodibility of the topsoil. The spatial explicit determination of the K-factor is crucial to create reliable soil loss estimate maps. However, the accurate estimation of the K-factor at the regional level is challenging due to different existing calculation methods and as a variety of spatial data (e.g. soil texture fractions, rock fragments, soil organic matter) is needed. The latter severely limits the choice of available data, which differ in spatial resolution and information content of the required parameters. This leads to a high potential of uncertainty in the regional estimation of K-factors, but also of the soil erosion estimates at regional level.

Therefore, the aim of the study presented is to determine the spatial and quantitative accuracy of different soil data sets and calculation methods for estimating the K-factor for Germany. Furthermore, the influence of possible uncertainties on the estimation of soil erosion risk by combining all factors of the (R)USLE should be evaluated. Based on the aim of this study we modelled K-factors for three available harmonized German-wide datasets: I) LUCAS soil dataset of Europe), II) BÜK1000 and III) BÜK200 (soil overview maps for Germany) using the existing equations of A) Wischmeier and Smith (1978) and its extension of B) Auerswald et al. (2014) representing the original K-factor nomograph. For the validation of the three datasets, K-factors based on 2234 arable soil profiles across Germany from the Agricultural Soil Inventory (BZE_LW) with most detailed soil information were calculated.

The results reveal significant differences between calculation methods and data sets for the German-wide assessment of K-Factors. The LUCAS soil dataset overestimates average K-Factors of the BZE-LW by ~ +21 % (RMSE = 0.015, R² = 0.30), while BÜK200 underestimates K-Factors by ~ -30 % (RMSE = 0.016; R² = 0.34). In contrast the calculation method has a low impact on the average K-Factor estimation (~ 5 %) while the RMSE is comparable high with 0.012 (R² = 0.57).

The results imply that the choice of the calculation method and dataset at regional level is important and that detailed information on soil texture (e.g. very fine sand fraction) are crucial and strongly needed in order to improve the estimation of reliable K-Factors and soil erosion risk on regional-scales. 

How to cite: Saggau, P. and Steinhoff-Knopp, B.: Uncertainties in the regional estimation of Soil Erodibility. A German wide evaluation of the K-Factor comparing current datasets and calculation methods., EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-9045, https://doi.org/10.5194/egusphere-egu23-9045, 2023.

The Universal Soil Loss Equation (USLE) was initially developed to support the implementation of conservation measures to minimize soil loss by water erosion, i.e. sheet and rill erosion, on a local scale and in the context of agricultural land use. The approach was refined over decades and became the Revised Universal Soil Loss Equation (RUSLE).  At the same time the scope of applications has grown considerably. Due to its rather simple structure and relatively low demand for input data, it has been used for the assessment of soil loss from water erosion for ever growing spatial entities, i.e. regional scale catchments or whole countries. Recently this has been applied on a global scale in order to identify global hotspots of soil erosion. This coherent approach for a global comparison is most welcome against the background of the large number of country-specific assessments which are rather difficult to compare.

However, there are two issues of concern. First, one needs to remember that RUSLE-derived soil loss assessments do not account for gully erosion, which might not be linearly scaled with sheet and rill erosion. Furthermore, information on the uncertainties of RUSLE based erosion assessments are not frequently reported. This especially concerns the impact of specific combinations of varying unique input data on the results.

In this contribution we compare in high spatial resolution the results of two RUSLE-based soil loss by water erosion assessments conducted for six 100 by 100 km large study sites in South Africa. The first assessment was conducted about two decades ago and was based on the then available data covering the whole of South Africa. The second assessment is a revision, which includes the latest input data for rainfall erosivity, soil erodibility, the topographic factor as well as land cover and management. When compared, the results of the current soil loss estimates are an order of magnitude lower than the previous estimates. Does this difference represent a temporal trend or just the inherent uncertainties reflecting different input data and slightly different data processing? This is the question discussed in this contribution.

How to cite: Baade, J., Zoller, K., and Le Roux, J.: Soil loss by water erosion assessment uncertainties – experiences from South Africa, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-3578, https://doi.org/10.5194/egusphere-egu23-3578, 2023.

In 1993, the Government of Navarre (northern Spain) began the installation and operation of a network of experimental watersheds in order to assess, among other aspects, soil erosion in representative agricultural areas of the territory. Initially, sediment sampling at the outlet of each of the five basins was performed on a daily basis, despite which it was possible to get a highly fruitful and novel knowledge on sediment export (Merchán et al., 2019). However, in the last 16 years, with the aim of studying sediment export in more detail, the sampling frequency was increased so that the behavior of the sedimentogram at the event level could be known. In these cases, when the amount of sediment was large enough, the sediment texture was also determined. In addition, from the beginning of the observations, a turbidimeter was used to record turbidity data every 10 minutes. The aim of this work is to deepen the knowledge of sediment export dynamics in representative agricultural watersheds of Navarre by analyzing the database described above and focusing in specific events. To do this, first, the entire database was represented in graphs that include variables such as sediment texture, samples taken per event, daily mean precipitation, turbidity, flow rate, etc. Next, events with a minimum of six samples were selected and the linear relationship between turbidity and sediment concentration was analyzed using simple linear regressions, as this is the method used in similar works. Subsequently, these same event data were added to set up monthly samples where again linear regressions were performed. Apart from the simple linear analysis, where the linear relationship with turbidity was analyzed as the only predictor variable, different artificial intelligence methods have been explored, such as the generalized linear model (GLM), support vector machine (SVM), multivariate adaptive regression splines (MARS) and random forest (RF), adding additional variables such as accumulated precipitation, and season or water level in the analysis. The results from all these statistical studies have been disappointing, since no pattern or generalization has been found to predict sediment concentration from the variables considered. These results suggest that the sediment export behavior of small agricultural watersheds is particularly complex and controlled by spatially and temporally varying variables. It is evident that at least some of these variables have not been taken into account in the study. The high variability found in sediment textures supports the hypothesis that the erosive behavior of watersheds is of great complexity. We believe that the consideration of variables such as vegetation on slopes and channels and its evolution can be helpful in the analysis.

Merchán, D., Luquin, E., Hernández-García, I., Campo-Bescós, M. A., Giménez, R., Casalí, J., & Del Valle de Lersundi, J. (2019). Dissolved solids and suspended sediment dynamics from five small agricultural watersheds in Navarre, Spain: A 10-year study. CATENA, 173, 114–130. https://doi.org/10.1016/J.CATENA.2018.10.013

How to cite: Barberena, I., Luquin, E., Campo-Bescós, M. Á., Eslava, J., Gimenez, R., and Casalí, J.: Challenges and progress in the detailed estimation of sediment export in agricultural watersheds in Navarra (Spain) after two decades of experience, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-13925, https://doi.org/10.5194/egusphere-egu23-13925, 2023.

Badlands are highly erosive landforms of dissected morphology, which can be found on soft rocks and unconsolidated sediments, with little or no vegetation, that are useless for agriculture. The erosion rates of these areas are high, causing important environmental and economic problems. For that reason, detecting the main conditioning factors that control badland occurrence and identifying susceptible areas is higly important to prevent soil erosion phenomena and their negative consequences.
This work attempts to assess badland susceptibility and their governing factors at a multi-scale level, using a random forest (RF) modelling approach. Previous RF-based research have demonstrated that RF modelling is a powerful tool for making predictions in the same spatial context and scale where the model has been trained. However, upscaling RF-modelling results to obtain accurate predictions in other, more extensive spatial contexts than that used for model training, remains an important challenge.
For that, the Upper Llobregat River Basin (ULRB, 504.8 km2) and Catalonia region (CAT, 32000 km2) have been selected as study areas. We have evaluated the viability of training a RF model for the analysis of badland suceptibility in the small spatial context of the ULRB, and further testing it to the more extensive spatial context of CAT. Revealing the most important factors that control badland distribution in the territory has been another goal in the present study. Eleven governing factors and two badland inventories developed for these study areas have been used for model training and testing. Model performance has been analyzed through validation tests and three different evaluation metrics: AUC, Kappa coefficient and accuracy. The outcomes of this work manifest that the two variables that have the most important relevance for badland occurrence are lithology and NDVI. In addition, our results showed that upscaling RF model results defined in the ULRB to the more extensive spatial context of CAT in order to predict badland occurrence, it’s possible but not ideal. Last, badland susceptibility maps of ULRB and Catalonia have been obtained with a very high accuracy (96% and 97% respectively), confirming the feasibility and uselfuness of RF approach for badland susceptibility assessment.

How to cite: Torra, O., Hürlimann, M., Puig-Polo, C., and Moreno-de las Heras, M.: Predicting badland occurrence in Catalonia by applying random forest techniques and a multi-scale approach, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-149, https://doi.org/10.5194/egusphere-egu23-149, 2023.