Assessment of Soil Erosion by RUSLE Model using Remote Sensing and GIS of Indla-Ghatkhed, District-Amravati, Maharashtra

The catchment boundary of Indla Ghatkhed watershed covers an area about 14..62 sq km. The erosion is a natural geomorphic process occurring continually over the earth’s surface and it largely depends on topography, vegetation, soil and climatic variables and, therefore, exhibits pronounced spatial variability due to catchments heterogeneity and climatic variation. This problem can be circumvented by discrediting the catchments into approximately homogeneous sub-areas using Geographic Information System (GIS). Soil erosion assessment modeling was carried out based on the Revised Universal Soil Loss Equation (RUSLE). A set of factors are involved in RUSLE equation are A = Average annual soil loss (mt/ha/year), R = Rainfall erosivity factor (mt/ha/year), k = Soil erodibility factor, LS = Slope length factor, C = Crop cover management factor, P = Supporting conservation practice factor. These factors extracted from different surface features by analysis and brought in to raster format. The output depicts the amount of sediment rate from a particular grid in spatial domain and the pixel value of the outlet grid indicates the sediment yield at the outlet of the watershed.

Computing slope length (USLE): return to original definitions.

<p>Soil erosion is an important process of consideration in different erosion risk models and in planning soil conservation. Common erosion models, such as the USLE and its derivatives are widely used. In this context, the slope length is the variable with the most difficulties due to the different scales and procedures available that lead to very different results. Furthermore, many of the calculation procedures are based on a hydrological network definition that poses many problems in areas with a complex topography.</p><p>We propose an algorithm implemented in GIS, returning to the original field perspective form defined by the USLE and RUSLE, which is detached from the hydrological network definition. The calculation procedure is based on 5 m DEM and defines overland water flow at the field scale.</p><p>This method has been applied in three areas with different climate and geomorphology. The results are similar to those derived from aerial photograph observation.</p><p><strong>References</strong></p><p>Honghu Liu, Jens Kiesel, Georg Hörmann, Nicola Fohrer. (2011). Effects of DEM horizontal resolution and methods on calculating the slope length factor in gently rolling landscapes. Catena, 87, 368–375</p><p>Renard, K.G., Foster, G.R., Weesies, G.A., Mc. Cool, D.K y Yoder, D.C. (1997). Predicting Soil Erosion by Water: A Guide To Conservation Planning With The Revised Universal Soil Los Equation. Agricultural Handbook 703. USA: US Department of Agriculture.</p><p><strong>Acknowledgements</strong></p><p>Authors are grateful to Authors are grateful to Agroseguro funding this research.</p>

Possibilities of including surface runoff barriers in the slope-length factor calculation in the GIS environment and its integration in the user-friendly LS-RUSLE tool

The effect of the morphology is key aspect of erosion modelling. In Universal Soil Loss Equation (USLE) type methods, this effect is expressed by the topographic factor (LS). The LS calculation in GIS is performed by a unit contributing area (UCA) method and can mainly be influenced by the pixel resolution, by the flow direction algorithm and by the inclusion of a hydrologically closed unit (HCU) principle, the cutoff slope angle (CSA) principle and the ephemeral gullies extraction (EG) principle. This research presents a new LS-RUSLE tool created with the inclusion of these principles in the automatic user-friendly GIS tool. The HCU principle using a specific surface runoff interruption algorithm, based on pixels with NoData values at the interruption points (pixels), appears to be key. With this procedure, the occurrence of overestimation results by flow conversion was rapidly reduced. Additionally, the reduction of extreme L and LS values calculated in the GIS environment was reached by the application of the CSA and EG principles. The results of the LS-RUSLE model show the prospective use of this tool in practice.

MODELS TO PREDICT SLOPE LENGTH FROM OTHER WATERSHED ATTRIBUTES

Soil erosion by water is an extensive and increasing problem worldwide. Albeit, this problem has been recognized as a significant hazard in Iraq, yet the number of studies on this topic is very limited. Most of the models used for estimating soil erosion contain parameters for slope length factor (LS). A major constraint is the difficulty in extracting the LS factor. Accordingly, the current study was initiated with the main objective of deriving models to predict the slope length from relatively easy to measure basin characteristics with a reasonable accuracy. To achieve the above objective, standard methodologies were employed to describe 30 main basins with the upper part of Iraq in terms linear, areal and relief morphometric parameters. The majority of the delineated watersheds were characterized by having high slope lengths indicating lower drainage density and higher erosion rate. Linear and non-linear least squares techniques were applied to predict the slope length from other basin characteristics. Different indicators were used to test the performance of the proposed models and the approach was validated using K-fold procedure at independent basins. The results indicated that the 4-parameter regression model outperformed the remaining models of watershed slope length. The regressors of this model are bifurcation ratio, perimeter, and basin length and slope gradient.

Topography significantly influencing low flows in snow-dominated watersheds

Watershed topography plays an important role in determining the spatial heterogeneity of ecological, geomorphological, and hydrological processes. Few studies have quantified the role of topography in various flow variables. In this study, 28 watersheds with snow-dominated hydrological regimes were selected with daily flow records from 1989 to 1996. These watersheds are located in the Southern Interior of British Columbia, Canada, and range in size from 2.6 to 1780 km2. For each watershed, 22 topographic indices (TIs) were derived, including those commonly used in hydrology and other environmental fields. Flow variables include annual mean flow (Qmean), Q10 %, Q25 %, Q50 %, Q75 %, Q90 %, and annual minimum flow (Qmin), where Qx % is defined as the daily flow that occurred each year at a given percentage (x). Factor analysis (FA) was first adopted to exclude some redundant or repetitive TIs. Then, multiple linear regression models were employed to quantify the relative contributions of TIs to each flow variable in each year. Our results show that topography plays a more important role in low flows (flow magnitudes ≤ Q75 %) than high flows. However, the effects of TIs on different flow magnitudes are not consistent. Our analysis also determined five significant TIs: perimeter, slope length factor, surface area, openness, and terrain characterization index. These can be used to compare watersheds when low flow assessments are conducted, specifically in snow-dominated regions with the watershed size less than several thousand square kilometres.

Topography significantly influencing low flows in snow-dominated watersheds

Watershed topography plays an important role in determining the spatial heterogeneity of ecological, geomorphological, and hydrological processes. Few studies have quantified the role of topography on various flow variables. In this study, 28 watersheds with snow-dominated hydrological regimes were selected with daily flow records from 1989 to 1996. The watersheds are located in the Southern Interior of British Columbia, Canada and range in size from 2.6 to 1,780 km2. For each watershed, 22 topographic indices (TIs) were derived, including those commonly used in hydrology and other environmental fields. Flow variables include annual mean flow (Qmean), Q10%, Q25%, Q50%, Q75%, Q90%, and annual minimum flow (Qmin), where Qx% is defined as flows that at the percentage (x) occurred in any given year. Factor analysis (FA) was first adopted to exclude some redundant or repetitive TIs. Then, stepwise regression models were employed to quantify the relative contributions of TIs to each flow variable in each year. Our results show that topography plays a more important role in low flows than high flows. However, the effects of TIs on flow variables are not consistent. Our analysis also determines five significant TIs including perimeter, surface area, openness, terrain characterization index, and slope length factor, which can be used to compare watersheds when low flow assessments are conducted, especially in snow-dominated regions.

Determination of landslide susceptibility in linear infrastructure. Case: aqueduct network in Palacé, Popayan (Colombia)

This research aimed to predict the occurrence of mass movements in the aqueduct network of Palacé, in the municipality of Popayan (Colombia). We evaluated the quality of SRTM and ASTER digital terrain models by comparing them with contour lines using a map scale of 1: 25000. The landscape parameters derived from the SRTM-DEM were analyzed with a multivariate procedure using algorithms implemented in free software, along with thematic information of the study area (coverage, distance to faults, rivers and precipitation). We selected non-redundant variables with the non-parametric ACP technique, and obtained a susceptibility prediction model using logistic regression, with two types of variables: dependent (landslides inventory from field observation) and independent (slope, slope length factor, topographic wetness index, flow path length, soil units and rate of convergence) resulting in a susceptibility map, reclassified into categories according to the values of probability. The prediction model could not be quantitatively assessed because of the absence of studies with a semi-detailed scale, but the estimation of the mean square error of elevation, from which the terrain parameters were derived, the level of detail and the performance of the classifier with ROC curve, yielded a zoning consistent with the findings of the field visits.