Scale effect in USLE and WEPP application for soil erosion computation from three Sicilian basins

Soil erosion increasingly poses a great threat to human food security and environmental quality. It is necessary to implement the assessment of soil erosion so as to provide precautionary measures and relevant suggestions for soil conservation. In this paper, the soil erosion model, revised universal soil loss equation (RUSLE), was used to quantify the soil loss in a large mountainous area of Lower Jinsha River Basin. The analysis of the datasets by means of geographic information systems (GIS) together with RUSLE led to the estimation of soil erosion. Results show that the average annual soil erosion was estimated at 52.1 t ha−1 year−1 and the total annual soil loss was 4.5 × 108 t. The highest erosion was found along the main course of Jinsha River. Soil erosion was serious in the elevation zone between 1,675 and 2,153 m and slope zone with slopes between 15° and 35°. As for land use types, cropland and grassland contributed 84.1 % to total soil loss due to the large areas and higher erosion rates. The results can be used to advice the local government in prioritizing the areas of immediate conservation practices.

Abstract. Soil erosion in Alpine areas is mainly related to extreme topographic and weather conditions. Although different methods of assessing soil erosion exist, the knowledge of erosive forces of the snow cover needs more investigation in order to allow soil erosion modeling in areas where the snow lays on the ground for several months. This study aims to assess whether the RUSLE (Revised Universal Soil Loss Equation) empirical prediction model, which gives an estimation of water erosion in t ha yr−1 obtained from a combination of five factors (rainfall erosivity, soil erodibility, topography, soil cover, protection practices) can be applied to mountain areas by introducing a winter factor (W), which should account for the soil erosion occurring in winter time by the snow cover. The W factor is calculated from the ratio of Ceasium-137 (137Cs) to RUSLE erosion rates. Ceasium-137 is another possible way of assessing soil erosion rates in the field. In contrast to RUSLE, it not only provides water-induced erosion but integrates all erosion agents involved. Thus, we hypothesize that in mountain areas the difference between the two approaches is related to the soil erosion by snow. In this study we compared 137Cs-based measurement of soil redistribution and soil loss estimated with RUSLE in a mountain slope affected by avalanches, in order to assess the relative importance of winter erosion processes such as snow gliding and full-depth avalanches. Three subareas were considered: DS, avalanche defense structures, RA, release area, and TA, track area, characterized by different prevalent winter processes. The RUSLE estimates and the 137Cs redistribution gave significantly different results. The resulting ranges of W evidenced relevant differences in the role of winter erosion in the considered subareas, and the application of an avalanche simulation model corroborated these findings. Thus, the higher rates obtained with the 137Cs method confirmed the relevant role of winter soil erosion. Despite the limited sample size (11 points), the inclusion of a W factor in RUSLE seems promising for the improvement of soil erosion estimates in Alpine environments affected by snow movements.

Accurate estimation of soil erosion at regional scale is essential to preserve fertile soils. Assessing soil erosion is crucial to plan conservation works in a river basin. The Shimsha basin is predominantly an agricultural basin, which faces soil erosion issues. This research evaluates the soil loss in the Shimsha basin, Karnataka, India, using the Revised Universal Soil Loss Equation (RUSLE) together with geographic information system (GIS) methods and remotely sensed (RS) data. RS data are used to obtain the important parameters of the RUSLE model, whereas GIS is used for mapping areas that are susceptible to erosion. Input datasets of rainfall, soil maps, land use land cover (LULC), and digital elevation model (DEM) are employed to calculate the corresponding RUSLE model factors. In the GIS setting, these factors—topography, rainfall erosivity, erodibility of soil, cover management, and conservation practices—are computed and merged to evaluate spatial patterns of soil loss throughout the watersheds. The estimated soil loss in the Shimsha basin varies from 0 to 4669.4 t ha −1 year −1 , and the annual mean total soil loss has been approximated to 2.81 million tonnes per year and an average soil loss of 3.24 t ha −1 year −1 . The findings indicate that soil erosion rates are highest in intensively cultivated areas and mountainous regions that have sparse vegetation and steep slopes. RUSLE and GIS integration approach offers a dependable estimate of soil loss. The results help in identification of priority areas, enabling effective planning along with execution of sustainable soil management activities.

This research explores the use of the revised universal soil loss equation (RUSLE) process and remote sensing (RS) and geographic information system (GIS) techniques to estimate soil erosion in Kundapura taluk, Udupi, Karnataka. The average annual predicted soil loss ranges between 292.16 and 41044.92 t yr-1. The loss of soil <292.16 t yr-1 near coastal areas has been recorded, and 1741.86 t yr-1 was found intensely plantation (mainly areca nut plantation and paddy field) area and degraded forest area. Soil erosion rate was predicted moderately high (4802.34 t yr-1) for agriculture and slope areas, which needs proper soil conservation measures to reduce erosion. The high, very high, and extreme rate (10601.16-41044.92 t yr-1) of soil erosion was found in the dense forest because of moderate slope value and the high slope length and steepness factor. The computed sediment yield (2010-2018) values range from 137.29- 504.65 t yr-1. In the coming years, huge man-made interventions in the Western Ghats region can significantly increase soil erosion. Also, most soil erosion changes happen at the micro-level and need much-localized study for specific regions. The study area belongs to the coastal region of Kundapura, which faces more soil erosion problems. This research may be helpful to understand and quantify soil erodibility rate. This study effectively proves the important role of RS and GIS in understanding geological processes and that the RULSE method can effectively be used for erosion mapping.

Erosion by water on agricultural areas has been studied by USDA (SCS, ARS, FS, NRCS) since the 1930‘s, initially consisting of natural runoff plots at numerous locations in the country that provided data that culminated in the development of the Universal Soil Loss Equation (USLE) in 1965 in Agriculture Handbook 282 (Wischmeier and Smith, 1965). Rainfall simulation equipment developed by ARS scientists located at Purdue University during the early 1960‘s was used to supplement the natural rainfall plot data and provided the ability to rapidly study soil erosion processes in more controlled laboratory and field experiments. The USLE update in Agriculture Handbook 537 (Wischmeier and Smith, 1978) contained a soil erodibility nomograph, developed from the results of extensive field rainfall simulation studies across the Midwest U.S. Transitions to process-based soil erosion modeling approaches in the 1980‘s-1990‘s relied upon laboratory and field rainfall simulation experiments to study processes and develop parameterization approaches for sensitive model input parameters. The Water Erosion Prediction Project (WEPP) model solves differential equations down a slope profile, that compute changes in sediment load and detachment/deposition at discrete points (Flanagan and Nearing, 1995; Flanagan et al., 2001). Unfortunately, most data available for validation of process-based hillslope interrill/rill erosion is from natural runoff plots (USLE database) or other natural or simulated rainfall studies where soil (sediment) loss was only measured at the plot end. Newer experimental measurement techniques, equipment, and approaches are needed to allow examination of changes in sediment detachment/deposition/loads with very fine spatial resolution down a hillslope and down a rill channel. The current modeling techniques also assume either uniformly spaced rill channels down a slope profile (default is 1-m spacing), or alternately simulations can be conducted where all surface flow is assumed to be distributed at a uniform shallow depth across the entire flow width. However, in real world situations the soil conditions, landscape complexities and increases in slope lengths can result in situations where no evident rilling occurs, rill spacings change with increasing slope lengths, or surface flows converge or diverge which may result in channelized rill flows changing to broader sheet flows. These complex surface flow situations need to be accounted for in any new water erosion prediction technologies, as they can greatly affect soil detachment and deposition processes. Additionally, subsurface hydrology effects also need to be better accounted for in any new technologies. Current erosion prediction models poorly account for water moving within the soil, and how saturation conditions can weaken the soil surface and make it more susceptible to detachment. A comprehensive understanding of how subsurface water moves and can converge in fields (and may emerge to the surface) is needed in new modeling approaches that can also be linked to initiation and development of ephemeral gullies. Additionally, better understanding and modeling of the effects of subsurface drainage and controlled drainage systems on hydrology and soil erosion is required. This also highlights the needs to move to erosion prediction systems that truly represent the landscape and fields as 2-D/3-D surfaces. Current prediction technologies are largely limited to estimating soil erosion processes at points down a line to represent a rectangular area. Long-term temporal effects of various practices on soil health and soil hydrology and erosion also should be accounted for in new erosion models. Currently, baseline soil erodibility is assumed to be a constant value for a particular soil and cropping system as a function of soil texture, organic matter, and other properties. However, fields that are in long-term pasture, or long-term no-till row crop management systems along with the use of cover crops may substantially change soil organic matter, soil structure, infiltration characteristics, and erodibilities that are not currently reflected in USDA models. This presentation will explore these and other areas of research needed.

Erosion by water on agricultural areas has been studied by USDA (SCS, ARS, FS, NRCS) since the 1930‘s, initially consisting of natural runoff plots at numerous locations in the country that provided data that culminated in the development of the Universal Soil Loss Equation (USLE) in 1965 in Agriculture Handbook 282 (Wischmeier and Smith, 1965). Rainfall simulation equipment developed by ARS scientists located at Purdue University during the early 1960‘s was used to supplement the natural rainfall plot data and provided the ability to rapidly study soil erosion processes in more controlled laboratory and field experiments. The USLE update in Agriculture Handbook 537 (Wischmeier and Smith, 1978) contained a soil erodibility nomograph, developed from the results of extensive field rainfall simulation studies across the Midwest U.S. Transitions to process-based soil erosion modeling approaches in the 1980‘s-1990‘s relied upon laboratory and field rainfall simulation experiments to study processes and develop parameterization approaches for sensitive model input parameters. The Water Erosion Prediction Project (WEPP) model solves differential equations down a slope profile, that compute changes in sediment load and detachment/deposition at discrete points (Flanagan and Nearing, 1995; Flanagan et al., 2001). Unfortunately, most data available for validation of process-based hillslope interrill/rill erosion is from natural runoff plots (USLE database) or other natural or simulated rainfall studies where soil (sediment) loss was only measured at the plot end. Newer experimental measurement techniques, equipment, and approaches are needed to allow examination of changes in sediment detachment/deposition/loads with very fine spatial resolution down a hillslope and down a rill channel. The current modeling techniques also assume either uniformly spaced rill channels down a slope profile (default is 1-m spacing), or alternately simulations can be conducted where all surface flow is assumed to be distributed at a uniform shallow depth across the entire flow width. However, in real world situations the soil conditions, landscape complexities and increases in slope lengths can result in situations where no evident rilling occurs, rill spacings change with increasing slope lengths, or surface flows converge or diverge which may result in channelized rill flows changing to broader sheet flows. These complex surface flow situations need to be accounted for in any new water erosion prediction technologies, as they can greatly affect soil detachment and deposition processes. Additionally, subsurface hydrology effects also need to be better accounted for in any new technologies. Current erosion prediction models poorly account for water moving within the soil, and how saturation conditions can weaken the soil surface and make it more susceptible to detachment. A comprehensive understanding of how subsurface water moves and can converge in fields (and may emerge to the surface) is needed in new modeling approaches that can also be linked to initiation and development of ephemeral gullies. Additionally, better understanding and modeling of the effects of subsurface drainage and controlled drainage systems on hydrology and soil erosion is required. This also highlights the needs to move to erosion prediction systems that truly represent the landscape and fields as 2-D/3-D surfaces. Current prediction technologies are largely limited to estimating soil erosion processes at points down a line to represent a rectangular area. Long-term temporal effects of various practices on soil health and soil hydrology and erosion also should be accounted for in new erosion models. Currently, baseline soil erodibility is assumed to be a constant value for a particular soil and cropping system as a function of soil texture, organic matter, and other properties. However, fields that are in long-term pasture, or long-term no-till row crop management systems along with the use of cover crops may substantially change soil organic matter, soil structure, infiltration characteristics, and erodibilities that are not currently reflected in USDA models. This presentation will explore these and other areas of research needed.

Estimation of soil erosion using USLE and GIS in Awassa Catchment, Rift valley, Central Ethiopia

Modelling runoff and soil erosion in logged forests: Scope and application of some existing models

The quantitative estimate of soil erosion, in space and time, is valuable information to initiate land degradation measures at a watershed level. In this study, two models, Morgan Morgan Finney (MMF) and universal soil loss equation (USLE), were used in GIS environment to assess the soil erosion, as a function of land use/land cover, soil and topography in a mountainous watershed in the Kashmir Himalayan region, India. The two modelled soil erosion estimates were validated using the available land degradation maps of the area in order to determine their efficacy for soil erosion modelling. The results from the two models showed some similarity between the two soil erosion estimates. However, keeping in view the soil deposition being taken into consideration by MMF (47.33% of watershed area), the disagreement with the USLE soil estimates is understandable. USLE estimated 72.52% of watershed area under 0–1 kg m−2 year−1 while as the MMF model estimated only 41.27% of the watershed area in this category. In both the model results, almost equal area of the watershed has been classified with erosion > 10 kg m−2 year−1 category. Based on the model validation with the available land degradation data, the USLE estimates of soil erosion were found more reliable because of the good correlation with the land degradation maps. The erosion estimates worked out in this study, particularly the categories under very high, high, severe and very severe eroded areas, shall go a long way in framing up the strategies for mitigation and control of soil erosion in the mountainous Himalayan watershed.

<abstract> <bold><sc>Abstract. </sc></bold>Performing a comprehensive sensitivity/uncertainty analysis is a valuable step in understanding and using a predictive hydrologic/water quality (H/WQ) model. This article applies one-factor-at-a-time (OAT) sensitivity analysis (SA) and first-order error analysis (FOEA)/Monte Carlo simulation with Latin hypercube sampling (LHS) uncertainty analysis techniques for evaluation of a complex, process-based water erosion prediction tool, the USDA Water Erosion Prediction Project (WEPP) model (version 2010.1). Assessment of the WEPP hillslope profile model on a Midwestern U.S. Miami silt loam soil for three cropping/management scenarios and three erosion process cases (as defined by topography) is described. WEPP model runoff, soil loss, and corn (Zea mays L.) yield output responses in the form of expected values and error variances were determined to illustrate model prediction uncertainty. The OAT SA showed that WEPP runoff and soil loss output responses were most sensitive to changes in the baseline effective hydraulic conductivity (K_b) and sand content. WEPP model corn yield output response was most sensitive to crop input parameters affecting the simulation of biomass development. The FOEA showed that the largest contributions to runoff, soil loss, and corn yield total error variance came from K_b and sand/clay content, K_b and baseline soil erodibility factors, and the biomass energy ratio of a crop and harvest index, respectively. The FOEA total variances presented in this study for runoff and soil loss were considerably larger than the corresponding Monte Carlo LHS simulation total variances. The Monte Carlo LHS total variance results were reasonable, making Monte Carlo LHS appear to be a better alternative for quantifying WEPP output response error variance. The Monte Carlo LHS soil loss output responses were also compared to Universal Soil Loss Equation (USLE) soil loss predictions. The USLE soil loss estimates were within the Monte Carlo LHS 90% prediction intervals for six of the nine cropping/management and erosion process cases. Results of this study illustrate the usefulness of combining SA and Monte Carlo LHS for providing detailed uncertainty analysis information for complex, physically based models such as WEPP.

Soil erosion is considered a very critical environmental issue that has repercussions for almost every aspect of the world. In developing countries, such as India, soil erosion continues to be a major limitation. A prediction and assessment of erosion prone areas is of utmost importance for soil fertility and water management. Recent technological advancements have provided useful models through which remotely-sensed data for a large scale area can be analyzed and interpreted. This study aims to adopt an erosion model that is unique to the physiography, biological and climatic conditions of the Indian Himalayan Region. The Revised Universal Soil Loss Equation (RUSLE) model estimates the average annual soil loss A in tonnes ha-1 year-1. Recognizing the conditions of the region, the RUSLE developed by Renard et al., (1997) was applied in conjunction with Geographic Information System (GIS) for estimating soil loss. All parameters of the model were thoroughly studied, starting from reviews and research papers on soil erosion assessment at national and catchment levels. The study follows the RUSLE soil model in estimating the rate of soil erosion at state and district level. The model was developed around coarse resolution data requirements, with practicality in providing annual soil loss rate for a large study area. It provides a means to describe specific districts that are vulnerable to soil erosion, rendering immediate action for soil conservation practices. To determine the spatial distribution of average annual soil erosion within the study area, cell-based parameters of the RUSLE were multiplied in the specified 500m x 500m spatial resolution using the raster calculator function in ArcGIS 10.0 software. The spatial pattern of soil erosion indicates that maximum erosion takes place in the north, north-western and eastern regions of the study area while the areas with low erosion rates are located in the eastern-most part of the study area. Keywords: Soil erosion, Remote sensing, Geographic information system, Soil erodibility

Soil degradation and erosion in semi-arid regions can significantly impact agricultural development, environmental sustainability, and hydrological balance. Understanding the impacts of land use changes and soil and water conservation (SWC) technique implementation on soil erosion and sediment yield is critical to planning effective watershed management. This study aims to evaluate the impacts of environmental changes in the Merguellil watershed (Central Tunisia) over the last forty years. To achieve this, remote sensing techniques and a geographic information system (GIS) will be employed to classify Landsat images from 1980 to 2020. Additionally, the Revised Universal Soil Loss Equation model will be utilized to estimate soil erosion rates, while the sediment delivery distributed model will be employed for sediment yield modeling. Spatiotemporal changes in land use and land cover and in areas treated with SWC techniques were analyzed as the main factors influencing changes in erosion and sediment yield. The combined impact of land use change and SWC techniques resulted in a decrease in the annual soil erosion rate from 18 to 16 t/ha/year between 1980 and 2020 and in sediment yield from 9.65 to 8.95 t/ha/year for the same period. According to the model’s predictions, both soil erosion and sediment yield will experience a slight increase with further degradation of natural vegetation and a reduction in the efficiency of SWC works. This emphasizes the importance of continued efforts in adopting and sustaining SWC techniques, as well as preserving natural vegetation cover, to proactively combat soil degradation and its adverse effects on the environment and communities. Continuous dedication to these measures is crucial to preserving our ecosystem, promoting sustainable practices, and protecting the well-being of both the environment and society.

Rainfall erosivity, or the capacity of rainfall to cause erosion, is a topic which has garnered substantial confusion over the years. This confusion stems from a number of discrepancies between isoerodent maps of the United States (and the world) and erosivity (or erosion index) benchmark values in the literature. Resolving these discrepancies is important since several of the world‘s soil erosion models, applications, and assessments are based on Universal Soil Loss Equation (USLE) technologies such as USLE, RUSLE, or RUSLE2. A brief history of these discrepancies is provided below. McGregor et al. (1995) observed a 30% difference (greater) in erosivities obtained from 29 breakpoint gauges in the Goodwin Creek Watershed, Mississippi as compared to isoerodent maps for the same location as used in the RUSLE model. Similar findings had been reported prior to the development of RUSLE (McGregor et al., 1980) regarding erosion index (EI) values used in the USLE. There are a number of reasons for these differences. Climate change and climate variability may have contributed, but there is also the potential impact of differing data types, which is discussed more below. An updated isoerodent map was prepared for the arrival of RUSLE2, which used more recent climate data (1960-1999). However, this update switched from using breakpoint precipitation data to a less precise and more ‘lossy‘ fixed-interval precipitation data for erosivity calculations (McGehee et al., 2021). This more commonly available data type does not preserve precipitation characteristics as well as the breakpoint data type, and thus, introduces bias/error in erosivity calculations when compared to the approach of Wischmeier & Smith (1958) and Wischmeier (1959), who discovered the original relationship of erosivity to soil loss. The RUSLE2 erosivity map values were greater than those from RUSLE and USLE by about 10%, but these values still fall short of benchmark erosivity values from McGregor et al. (1995) and Flanagan et al. (2020) by 19% and 32%, respectively. It is unclear how much of these differences are a result of the differing time periods, measurement technologies, or erosivity calculation, gap-filling, and interpolation methods. McGehee et al. (2021) demonstrated the impact of different data types (e.g., breakpoint vs. fixed-intervals for common gauge precisions) on erosivity calculations, and McGehee et al. (2022) updated the national isoerodent map of the United States. Erosivity estimates based on fixed-interval precipitation data continued to fall below breakpoint-derived benchmarks, but these differences were substantially less than existing erosivity maps. These results point to potential insufficiencies in currently recommended intensity dampening correction and storm omission practices when using fixed-interval precipitation data to estimate erosivity values and map them over large areas. Work aimed at resolving these issues is ongoing. This presentation will provide a brief history of erosivity estimation and mapping in the United States, the most recent progress towards a more reliable dynamic national erosivity map for the United States and the world, and a glimpse of the future of these efforts.

Scaling effects on sediment yield in the upper Yangtze River are investigated through constructing relationships between the specific sediment yield(SSY) and drainage area and factors influencing soil erosion based on data of drainage basins upstream of 268 hydrological stations.The factors influencing soil erosion used in the study are those in the Revised Universal Soil Loss Equation(RUSLE) and are calculated from DEM,rainfall,landuse,soil type data banks for the upper Yangtze River.Results show that there is a negative power relationship between SSY and the drainage area of all the gauged drainage basins in the upper Yangtze River and this relationship can be ascribed to the changes in rainfall erosivity factor R and soil erodibility factor K in RUSLE with the drainage area.Between the drainage area of 104-1.58×105km2 exists an abrupt decline of SSY with the increase of drainage area and it is a main component of the overall decreasing trend of SSY with the drainage area.In the regression relationship of SSY with the drainage area and all RUSLE factors,SSY is positively and exponentially related with the drainage area,revealing that deposition of eroded sediment occurs in a short distance in the upper Yangtze River.The relationships of SSY with the drainage area and RUSLE factors of the drainage basins of the main first-order tributaries of the upper Yangtze River are also constructed and analyzed.The results show different relationships for different tributaries because of different reasons.The scaling effects on sediment yield and causes for changes in SSY with the drainage area in the Jingsha basin are similar to those in the whole upper Yangtze River.The causes for changes in SSY with the drainage area in the Min-Tuo(Minjiang-Tuojiang) basin are also associated to the changes in factors R and K with the drainage area,but SSY is negatively related with the drainage area in the multiple regression relationship between SSY and the drainage area and factors in RUSLE due to sediment deposition in the lower reaches of the tributary.In the Jialing basin,the main factors for the occurrence of the positive relationship between SSY and the drainage area are the slope length and steepness factor LS and support practice factor P,and the positive relationship between SSY and the drainage area also keeps in the multiple regression between SSY and factors in RUSLE and the drainage area.There is not a significant regression relationship between SSY and the drainage area in the Wu-Chishui-Heng(Wujiang-Chishui-Hengjiang) basins due to intensive human activities.

Soil erosion and sediment yield are strongly affected by land use change. Spatially distributed erosion models are of great interest to predict soil erosion loss and sediment yield. Hence, the objective of this study was to determine sediment yield using Revised Universal Soil Loss Equation (RUSLE) model in Geographical Information System (GIS) environment at Cameron Highlands, Pahang, Malaysia. Sediment yield at the study area was determined using RUSLE model in GIS environment The RUSLE factors were computed by utilizing information on rainfall erosivity (R) using interpolation of rainfall data, soil erodibility (K) using soil map and field measurement, vegetation cover (C) using satellite images, length and steepness (LS) using contour map and conservation practices using satellite images based on land use/land cover. Field observations were also done to verify the predicted sediment yield. The results indicated that the rate of sediment yield in the study area ranged from very low to extremely high. The higher SY value can be found at middle and lower catchments of Cameron Highland. Meanwhile, the lower SY value can be found at the north part of the study area. Sediment yield value turned out to be higher close to the river due to the topographic characteristic, vegetation type and density, climate and land use within the drainage basin.