<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "https://jats.nlm.nih.gov/publishing/1.3/JATS-journalpublishing1-3.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" article-type="research-article" dtd-version="1.3" xml:lang="en">
<front>
<journal-meta>
<journal-id journal-id-type="publisher-id">gcj1</journal-id>
<journal-title-group><journal-title>Remote Sensing of Land</journal-title></journal-title-group>
<issn publication-format="electronic">2582-3019</issn>
<publisher><publisher-name>Gatha Cognition</publisher-name></publisher>
</journal-meta>
<article-meta>
<article-id pub-id-type="doi">10.21523/gcj1.20040101</article-id>
<article-id pub-id-type="publisher-id">gca101</article-id>
<article-categories><subj-group subj-group-type="heading"><subject>Original Research Paper</subject></subj-group></article-categories>
<title-group><article-title>Modeling Soil Erosion Vulnerability Using RUSLE: A Case Study Upstream of Nangal Reservoir, India</article-title></title-group>
<contrib-group>
<contrib contrib-type="author"><contrib-id contrib-id-type="orcid" authenticated="false">https://orcid.org/0000-0003-1233-6487</contrib-id><name><surname>Nehra</surname><given-names>Sunil</given-names></name><xref ref-type="aff" rid="aff1"/></contrib>
<contrib contrib-type="author"><contrib-id contrib-id-type="orcid" authenticated="false">https://orcid.org/0000-0002-3070-1233</contrib-id><name><surname>Sur</surname><given-names>Koyel</given-names></name><xref ref-type="aff" rid="aff1"/></contrib>
<contrib contrib-type="author"><contrib-id contrib-id-type="orcid" authenticated="false">https://orcid.org/0000-0001-5639-2910</contrib-id><name><surname>Verma</surname><given-names>Vipan</given-names></name><xref ref-type="aff" rid="aff1"/></contrib>
</contrib-group>
<aff id="aff1"><institution>Punjab Remote Sensing Centre (PRSC), PAU Campus, Near Kitchlu Nagar block ‘F-end, Ludhiana, Punjab (India)-141004.</institution></aff>
<pub-date publication-format="electronic" date-type="pub"><day>30</day><month>10</month><year>2020</year></pub-date>
<volume>4</volume>
<issue>1-2</issue>
<fpage>1</fpage><lpage>11</lpage>
<history><date date-type="received"><day>09</day><month>09</month><year>2020</year></date><date date-type="rev-recd"><day>26</day><month>10</month><year>2020</year></date><date date-type="accepted"><day>27</day><month>10</month><year>2020</year></date></history>
<permissions><copyright-year>2020</copyright-year><license license-type="open-access" xlink:href="https://creativecommons.org/licenses/by-nc-nd/4.0/"><ali:license_ref>https://creativecommons.org/licenses/by-nc-nd/4.0/</ali:license_ref><license-p>This is an open access article distributed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0).</license-p></license></permissions>
<self-uri xlink:href="https://gathacognition.com/article/gca101/modeling-soil-erosion-vulnerability-using-rusle-a-case-study-upstream-of-nangal-reservoir-india"/>
<self-uri content-type="pdf" xlink:href="https://gathacognition.com/uploads/admin/articlefiles/GCJ1M4581.pdf"/>
<abstract>
<p>Simulation of soil erosion is an important phenomenon in watershed management studies because of the dynamic nature of soil. The present study demonstrates the capabilities of remote sensing coupled with GIS to extract soil erosion vulnerabilities using RUSLE model. Therefore this study understands spatial distribution of soil erosion at a macro-watershed scale. The input parameters for RUSLE model include the rainfall and erosivity (R) derived using the TRMM satellite precipitation product for a long term implementing regression model. The soil erodibility (K) factor is estimated using available soil map, and the topographic factor (LS) from 12m ALOS PALSAR DEM. The management practice information was extracted from the land use / cover map based on 30m multi spectral LANDSAT-8 of three season datasets. The RUSLE model predicted very low, low, moderate, high and very high erosion in 84.8%, 7.9%, 4.4%, 2.3% and 0.7% of reviewed area, respectively. The results were validated with ground infromation and NDVI, being the main identifiers of soil erosion. As presence of high vegetation indicates higher strength to hold the soil particles together and prevent soil erosion, correlation coefficient was generated to find accuracy of the results with NDVI of the area. The strong positive correlation (0.5) provided scientific support that prediction of soil erosion by RUSLE can be effectively used for watershed management and combating action plan for soil erosion.</p>
</abstract>
<kwd-group kwd-group-type="author"><kwd>Soil</kwd><kwd>GIS</kwd><kwd>Remote Sensing</kwd><kwd>RUSLE</kwd><kwd>Watershed Management</kwd></kwd-group>
</article-meta>
</front>
<body>
<sec id="s1"><title>Introduction</title>
<p>Climate change is expected to impact on soil erosion to greater extent (<xref ref-type="bibr" rid="R4076">Gupta and Kumar, 2017</xref>) because spatial and temporal patterns of long term soil erosion and deposition dynamics are results of complex interactions. Several natural processes of the Earth system dynamics such as rainfall, flowing water, underground water, vegetation growth, soil detachment, transport etc. has a great role in soil erosion. Erosion processes driven by hydrological factors are extremely complicated therefore over the years researchers are trying to find solutions to predict soil erosion quantitatively over large landscapes (<xref ref-type="bibr" rid="R4073">Finlayson and Montgomery, 2003</xref>). Soil erosion has currently evolved as a global environmental issue which is responsible for nearly 85% of land degradation around the globe, reducing over time the global crop productivity by 17% (<xref ref-type="bibr" rid="R4083">Nyesheja <italic>et al</italic>., 2019</xref>). Soil erosion therefore adversely affects the rate of crop yield and water quality (<xref ref-type="bibr" rid="R4081">Lal, 1998</xref>). Due to this worldwide rise in concern, estimating soil erosion by different models at catchment levels has become very popular because it helps to provide both quantitative and qualitative estimation (<xref ref-type="bibr" rid="R4084">Phinzi <italic>et al</italic>., 2019</xref>). Models developed till date to calculate soil erosion can be broadly divided into (a) empirical, (b) conceptual and (c) those based on physical processes (<xref ref-type="bibr" rid="R4069">Aksoy and Kavvas, 2005</xref>; <xref ref-type="bibr" rid="R4079">Kinnell, 2010</xref>). Universal Soil Loss Equation (USLE) is an empirical model introduced in the mid-1960s, this model gained wide popularity over time (<xref ref-type="bibr" rid="R4095">Wischmeier and Smith, 1965</xref>). Over the past few decades, USLE has undergone several significant modifications and up gradation resulting to improved versions of empirical model such as the Modified USLE (MUSLE) developed by <xref ref-type="bibr" rid="R4094">Williams and Berndt, 1977</xref>, then came Soil Loss Estimation Model for Southern Africa (SLEMSA) developed by <xref ref-type="bibr" rid="R4072">Elwell (1977</xref>), Revised USLE (RUSLE) proposed by <xref ref-type="bibr" rid="R4086">Renard <italic>et al</italic>., (1991)</xref> and <xref ref-type="bibr" rid="R4087">Renard <italic>et al</italic>. (1997)</xref>, European Soil Erosion Model (EuroSEM) (<xref ref-type="bibr" rid="R4082">Morgan <italic>et al</italic>., 1998</xref>), USLE-M (<xref ref-type="bibr" rid="R4080">Kinnell and Risse, 1998</xref>), and RUSLE (<xref ref-type="bibr" rid="R4074">Foster <italic>et al</italic>., 2003</xref>). Amongst these models, RUSLE has been proven to be the most frequently used empirical model because it is also computer based model which provides a clear perspective for understanding the interaction of erosion and its causative factors (<xref ref-type="bibr" rid="R4070">Alexakis <italic>et al</italic>., 2013</xref>). The biggest advantage of RUSLE is that it can be stimulated in the Geographic Information System (GIS) environment where satellite based datasets are used to derive several factors for the model as input parameters. Annual soil loss is estimated from the combination of the six factors contained in RUSLE: rainfall erosivity (R), soil erodibility (K), slope length (L), slope steepness (S), soil use and management (C) and support practices (P).</p>
<p>Remote sensing provides several important inferences from the earth observation data and allow us to gain insights into interactions between physical processes and environmental conditions that control erosion and landform evolution. Satellite derived images are preferred than traditional methods which are related to field measurements because although they gave detailed and accurate measurement at plot scale but employing this methods at catchment levels are really difficult and require considerable amount of time, money, and effort. RUSLE provides an ideal framework for assessing soil erosion and its factors (<xref ref-type="bibr" rid="R4076">Ganasri and Ramesh <italic>et al.,</italic> 2016</xref>). The RUSLE model can identify soil loss possibility in GIS environment by pixel cell-by-cell method which helps to recognize individual spatial classes of the soil erosion in a big area effectively (<xref ref-type="bibr" rid="R4089">Shinde <italic>et al.</italic>, 2010</xref>). RUSLE model can be calculated at different geospatial scales in the digital platform. The research objectives of this study are: (1) to develop a methodology that combines remote sensing data and GIS with RUSLE to estimate spatial distribution of soil erosion at a macro-watershed scale with validation of the results using rigorous field work over the study area.</p>
</sec>
<sec id="s2"><title>Materials And Methods</title>
<sec id="s2-1"><title>2.1 Study Area</title>
<p>The present study was carried out over a macro- watershed upstream of Nangal reservoir with a total area of 8390 ha, selected mainly because of the spatial diversity of plant cover and the variety of soil classes and slope (<xref ref-type="fig" rid="F712">Figure 1</xref>). The climatic characteristics indicate generally dry hot summer (except in the south-west monsoon season) and a bracing cold winter. The temperature ranges from minimum of 4°C in winter to 45°C in summer. The study area lies in the foothill regions of the Siwaliks where potential soil erosion rate is maximum. The point of interest in the study was to see whether RUSLE integrated in GIS environment was being able to be captured the soil erosion at this spatial scale and variation.</p>
<fig id="F712"><label>Figure 1</label><caption><title>Study area: Nangal watershed of Punjab, India</title></caption><graphic xlink:href="https://gathacognition.com/uploads/admin/articlefigure/bb872d257a0f66e774fe405bd712e494.jpg" mimetype="image" mime-subtype="jpeg"/></fig>
</sec>
<sec id="s2-2"><title>2.2 Data Used</title>
<p>Three sets of Landsat-8 images (<italic>Kharif, Zaid and Rabi)</italic> were used in this work for generating thematic layers of land use / land cover (<xref ref-type="table" rid="T399">Table 1</xref>). ALOS PALSAR digital elevation model was used to derive elevation data. Rainfall data used for deriving R factor used in RUSLE was generated using Tropical Rainfall Mapping Mission (TRMM) datasets of last 18 years. Available soil map on 1:50000-scale was used for soil erodibility (K) extraction. Field survey was also conducted in three different seasons for validation of the final output by collecting GCPs.</p>
<table-wrap id="T399"><label>Table 1</label><caption><title>Data</title></caption><table frame="hsides" rules="groups">
<tbody>
<tr><td>Data</td><td>Source</td><td>Resolution</td></tr>
<tr><td>Digital Elevation Model</td><td>https://urs.earthdata.nasa.gov/</td><td>12.5 m</td></tr>
<tr><td>LANDSAT 8-OLI</td><td>www.earthexplorer.com</td><td>30 m</td></tr>
<tr><td>Soil data</td><td>Soil and Land Survey of India</td><td>1:50000</td></tr>
<tr><td>Rainfall data</td><td>TRMM (Monthly Data) https://pmm.nasa.gov/data-access/downloads/trmm</td><td>0.25º×0.25º</td></tr>
</tbody>
</table>
</table-wrap>
</sec>
<sec id="s2-3"><title>2.3 Analysis and Visualization</title>
<p>Implementation of the RUSLE model in GIS involves creating a work flow in computer based environment that includes input data processing, model computation, and analysis of results (<xref ref-type="fig" rid="F713">Figure 2</xref>). The GIS provided powerful tools for analysis, visualization of modeling and specialized tools to perform model calibration and validation.</p>
<fig id="F713"><label>Figure 2</label><caption><title>Methodology</title></caption><graphic xlink:href="https://gathacognition.com/uploads/admin/articlefigure/fc413e35f5149fd71f362ee49c8a9f6a.jpg" mimetype="image" mime-subtype="jpeg"/></fig>
</sec>
<sec id="s2-4"><title>2.4 RUSLE Model</title>
<p>RUSLE is often used to monitor soil erosion in agriculture watersheds ranging from macro to mini watersheds across the world (<xref ref-type="bibr" rid="R4091">Udayakumara <italic>et al</italic>., 2010</xref>). This model was particularly selected for this study because of its demonstrated effectiveness. The RUSLE (<xref ref-type="bibr" rid="R4087">Renard <italic>et al</italic>., 1997</xref>) model can be expressed as equation (1) and the methodology followed to map soil erosion.</p>
<p><italic>\(A=R×K×LS×C×P\)</italic></p>
<p><italic>A</italic> = Computed soil loss per unit area per year (t/ha per year); <italic>R</italic> = Rainfall erosivity factor (MJ mm ha-1h-1year-1); <italic>K</italic> = Soil erodibility factor (t ha MJ-1mm-1); <italic>LS</italic> = Slope length and steepness factor (dimension less); <italic>C </italic>= Cover management factor (dimension less); <italic>P</italic> = Support practice factor (dimension less).</p>
</sec>
<sec id="s2-5"><title><italic>2.4.1 </italic><italic>Rainfall Erosive Factor (R) </italic></title>
<p>The rainfall erosivity factor is measured by rainfall intensity thus R basically explains the effect of raindrop on the ground. The modeling of R factor in GIS environment requires continuous satellite retrieved precipitation data. Therefore, monthly rainfall data product of TRMM satellite was downloaded over the study area for 18 years (1998-2015) and used to calculate the R-factor. Random 50 points were plotted over the study area. The locations and altitude of each point was calculated. Multiple regression models were used to derive a coefficient against each point’s altitude, longitude, latitude and rainfall. Finally rainfall intensity was calculated using the equation given by Van der .</p>
<p>\(R=a.p_j\)</p>
<p>\(p_j\) is rainfall intensity in mm, a = coefficient derived from the multiple regression model.</p>
<fig id="F715"><label>Figure 3</label><caption><title>Rainfall pattern (1998-2015)</title></caption><graphic xlink:href="https://gathacognition.com/uploads/admin/articlefigure/6568555f83ff90603614011f9d57e9a0.jpg" mimetype="image" mime-subtype="jpeg"/></fig>
<fig id="F714"><label>Figure 4</label><caption><title>Thematic maps: (a) Rainfall erosivity factor, (b) Soil erodibility factor, (c) Slope length and steepness factor and (d) Land cover management factor</title></caption><graphic xlink:href="https://gathacognition.com/uploads/admin/articlefigure/41ac89797eb22a5a40f6635b9bae1002.jpg" mimetype="image" mime-subtype="jpeg"/></fig>
</sec>
<sec id="s2-6"><title><italic>2.4.2 </italic><italic>Soil Erodibility Factor (K) </italic></title>
<p>Soil erodibility is a function of soil texture, structure (e.g., macro porosity, aggregate properties), organic matter content, hydraulic properties and wetness of soil due to rain water and runoff (<xref ref-type="bibr" rid="R4071">Blanco-Canqui and Lal, 2008</xref>). In this study available soil map was used to derive the soil erodibility (K) classes using standard classification scheme (<xref ref-type="table" rid="T400">Table 2</xref>). Since this watershed has thick forested layer whereas detailed field survey were not possible.</p>
<table-wrap id="T400"><label>Table 2</label><caption><title>Classification scheme for K-factor</title></caption><table frame="hsides" rules="groups">
<tbody>
<tr><td>Texture class</td><td>Average</td><td>Less than 2%</td><td>More than 2%</td></tr>
<tr><td>Clay</td><td>0.22</td><td>0.24</td><td>0.21</td></tr>
<tr><td>Clay loam</td><td>0.3</td><td>0.33</td><td>0.28</td></tr>
<tr><td>Coarse sandy loam</td><td>0.07</td><td></td><td>0.07</td></tr>
<tr><td>Fine sand</td><td>0.08</td><td>0.09</td><td>0.06</td></tr>
<tr><td>Fine sandy loam</td><td>0.18</td><td>0.22</td><td>0.17</td></tr>
<tr><td>Heavy clay</td><td>0.17</td><td>0.19</td><td>0.15</td></tr>
<tr><td>Loam</td><td>0.3</td><td>0.34</td><td>0.26</td></tr>
<tr><td>Loamy fine sand</td><td>0.11</td><td>0.15</td><td>0.09</td></tr>
<tr><td>Loamy Sand</td><td>0.04</td><td>0.05</td><td>0.04</td></tr>
<tr><td>Loamy very fine sand</td><td>0.39</td><td>0.44</td><td>0.25</td></tr>
<tr><td>Sand</td><td>0.02</td><td>0.03</td><td>0.01</td></tr>
<tr><td>Sandy clay loam</td><td>0.2</td><td></td><td>0.2</td></tr>
<tr><td>Sandy loam</td><td>0.13</td><td>0.14</td><td>0.12</td></tr>
<tr><td>Silt loam</td><td>0.38</td><td>0.41</td><td>0.37</td></tr>
<tr><td>Silt clay</td><td>0.26</td><td>0.27</td><td>0.26</td></tr>
<tr><td>Silt clay loam</td><td>0.32</td><td>0.35</td><td>0.3</td></tr>
<tr><td>Very fine sand</td><td>0.43</td><td>0.46</td><td>0.37</td></tr>
<tr><td>Very fine sandy loam</td><td>0.35</td><td>0.41</td><td>0.33</td></tr>
</tbody>
</table>
</table-wrap>
</sec>
<sec id="s2-7"><title><italic>2.4.3 </italic><italic>Slope Length and Steepness Factor </italic></title>
<p>The topographical factors slope length (L) and the slope steepness (S) represents a ratio of soil loss below specified condition to any site (), it is a unit less factor.The highest and the lowest slope represent the maximum potentiality of soil erosion as LS factor generate high overland flow velocity and similarly higher runoff (<xref ref-type="bibr" rid="R4096">Zhao <italic>et al</italic>., 2020</xref>). To obtain slope length (L) and slope steepness (S), ALOS PALSAR DEM data was used. This resolution was chosen by accurately defining flow direction in the watershed, using the “flow direction” tool of the Arc GIS software. Slope length corresponded to 10.0 m, when water flow was perpendicular to the pixel line and 14.14 m when it was diagonally oriented. The slope length factor was computed using the equation below:</p>
<p>\(L= (µ/22.1) m\)</p>
<p>Where, <italic>L</italic> = Slope length factor, µ = horizontal projected slope length (m)</p>
<p>m = slope length exponent, (µ=flow accumulation X cell size)</p>
<p>In this equation ‘m’ slope length varies based on slope steepness, m equals 0.5 if the slope is 4.5% or more; 0.4 if the slope is 3-4.5%; 0.3 if the slope is 1-3% and 0.2 of uniform gradient of less than 1% (<xref ref-type="bibr" rid="R4095">Wischmeier and Smith, 1978</xref>).</p>
</sec>
<sec id="s2-8"><title><italic>2.4.4 </italic><italic>Cover Management Factor (C) </italic></title>
<p>The C-factor is important because it helps to indicate how soil loss potential will be distributed in time under different land use/land covers and during crop rotations, construction or other management activities (<xref ref-type="bibr" rid="R4090">Thomas <italic>et al</italic>., 2018</xref>). To identify the C-factors, GIS techniques were used to extract thematic layer of land use / land cover of the study region from the satellite imagery. Then the literature based C-factor values were assigned for different land use / land cover in the study area according to <xref ref-type="table" rid="T401">Table 3</xref>.</p>
<table-wrap id="T401"><label>Table 3</label><caption><title>Classification scheme for C-factor (Kent, 1972)</title></caption><table frame="hsides" rules="groups">
<tbody>
<tr><td>Land use/land cover</td><td>C-factor</td></tr>
<tr><td>Degraded forest</td><td>0.0006</td></tr>
<tr><td>Dense forest</td><td>0.003</td></tr>
<tr><td>Double cropped</td><td>0.20</td></tr>
<tr><td>Fallow</td><td>0.80</td></tr>
<tr><td>Single cropped</td><td>0.62</td></tr>
<tr><td>Steeply sloping</td><td>0.80</td></tr>
<tr><td>Undulating land with or without grass</td><td>0.14</td></tr>
<tr><td>Water body</td><td>1.00</td></tr>
</tbody>
</table>
</table-wrap>
</sec>
<sec id="s2-9"><title><italic>2.4.5 </italic><italic>Support Practice Factor (P) </italic></title>
<p>Support practice factor represents the positive impacts of support practices on the soil it is often described as the ratio of soil loss by a support practice to that of straight-row farming up and down the slope (<xref ref-type="fig" rid="F714">Figure 4e</xref>). The P-factor indicates conservation practices adopted in the watershed including terracing, bunding, strip cropping, check dam etc.The value of P- factor ranges from 0-1, in which 0 values represent high-quality preservation practice and the value resembling one indicates poor protection practices (<xref ref-type="bibr" rid="R4082">Morgan <italic>et al</italic>., 1998</xref>). In catchment areas often several mitigation practices like terracing and contour tillage among other practices cannot be reflected in land use maps (<xref ref-type="bibr" rid="R4075">Fu <italic>et al</italic>., 2005</xref>). Therefore, P-factor information was calculated through empirical method proposed by (<xref ref-type="bibr" rid="R4093">Wener, 1972</xref>):</p>
<p>\(P = 0.2+0.03×S \) ,</p>
<p>Where, <italic>S</italic> = Slope steepness, <italic>P</italic> = P-factor</p>
</sec>
<sec id="s2-10"><title><italic>2.4.6 </italic><italic>Annual Soil Loss (A) </italic></title>
<p>Soil loss determined using the RUSLE model using the Arc GIS software, was classified into five classes of soil loss: 0-2; 2-4; 4-6; 6-8 and &gt;8 t/ha/yr.</p>
<fig id="F716"><label>Figure 4</label><caption><title>Thematic maps: (e) Support practice factor and (f) Soil loss estimations</title></caption><graphic xlink:href="https://gathacognition.com/uploads/admin/articlefigure/a99422e9fcaeaa15cf7b6ca6936a5303.jpg" mimetype="image" mime-subtype="jpeg"/></fig>
</sec>
</sec>
<sec id="s3"><title>Results And Discussions</title>
<sec id="s3-1"><title>3.1 Rainfall Erosivity Factor (R)</title>
<p>Since many studies often indicate that rainfall is the main indicator of soil erosion in any region therefore the pattern of rainfall in this region needs to be studied carefully. The monthly rainfall datasets were summed up to get the annual rainfall of each year over the study area. More than 400 mm of rainfall is witnessed in all the years and an average of 800 mm has been observed in this region over 18 years (<xref ref-type="fig" rid="F715">Figure 3</xref>). In the present analysis average annual rainfall was used for R-factor calculation (<xref ref-type="fig" rid="F714">Figure 4a</xref>).</p>
<p>The estimated R-factor ranges from 165 to 183 mm.ha<sup>-1 </sup>hr<sup>-1</sup>/year in the study area. The R-factor calculated has been divided into five major zones as seen in <xref ref-type="table" rid="T402">Table 4</xref> and <xref ref-type="fig" rid="F714">Figure 4(a)</xref>. 165-169 mm.ha-<sup>1</sup>hr<sup>-1</sup>/year covers 5307 ha, 169-172 mm.ha<sup>-1</sup>hr<sup>-1</sup>/year covers 667 ha, 172-176 mm.ha<sup>-1</sup> hr<sup>-1</sup>/year covers 563ha, 176-180 mm.ha<sup>-1</sup> hr<sup>-1</sup>/year covers 526ha and 180-183 mm.ha<sup>-1</sup> hr<sup>-1</sup>/year covers 1327 ha. This region experiences orographic type of rainfall, the foot hill region in the northern part of the study area experiences heavy rainfall and gradually the erosivity rate decreases as the land slopes down towards the plain.</p>
<table-wrap id="T402"><label>Table 4</label><caption><title>Rainfall erosivity factor</title></caption><table frame="hsides" rules="groups">
<tbody>
<tr><td>R-factor</td><td>Area (ha)</td></tr>
<tr><td>165-169</td><td>5307</td></tr>
<tr><td>169-172</td><td>667</td></tr>
<tr><td>172-176</td><td>563</td></tr>
<tr><td>176-180</td><td>526</td></tr>
<tr><td>180-183</td><td>1327</td></tr>
<tr><td>Total</td><td>8390</td></tr>
</tbody>
</table>
</table-wrap>
</sec>
<sec id="s3-2"><title>3.2 Soil Erodibility Factor (K)</title>
<p>The study area is classified into four major soil erodibility classes based on soil texture. The study area comprises of four major soil textural classes i.e. loamy sand, sandy loam, silt and loam (<xref ref-type="fig" rid="F714">Figure 4b</xref>). Maximum area is covered by loamy sand with an average soil erodibility factor of 0.3. The sandy loam soils possess an average erodibility factor of 0.13. Silty loam soils found in small patches provides an average of 0.38 soil erodibility factor. Very small patches of loam soils are found in the southern part of the study area. The silt loam soils K value as 0.38 occupies 328 ha of land, whereas sandy loam soils and occupies 1504 ha of land. <xref ref-type="table" rid="T403">Table 5</xref> shows the distribution of the soil erodibility factor.</p>
<table-wrap id="T403"><label>Table 5</label><caption><title>Soil erodibility factor</title></caption><table frame="hsides" rules="groups">
<tbody>
<tr><td>Texture</td><td>Average</td><td>Less than 2%</td><td>More than 2%</td><td>Area (ha)</td></tr>
<tr><td>Loam</td><td>0.3</td><td>0.34</td><td>0.26</td><td>184</td></tr>
<tr><td>Loamy sand</td><td>0.04</td><td>0.05</td><td>0.04</td><td>6374</td></tr>
<tr><td>Sandy loam</td><td>0.13</td><td>0.14</td><td>0.12</td><td>1504</td></tr>
<tr><td>Silt loam</td><td>0.38</td><td>0.41</td><td>0.37</td><td>328</td></tr>
<tr><td>Total</td><td></td><td></td><td></td><td>8390</td></tr>
</tbody>
</table>
</table-wrap>
</sec>
<sec id="s3-3"><title>3.3 Slope Length and Steepness Factor (LS)</title>
<p>Topographic factor represents the basic influence of slope length and slope steepness on erosion process. LS- factor was calculated in ARCGIS by considering the flow accumulation and slope in percentage as the input factors. From the analysis, it is observed that the value of topographic factor increases as the flow accumulation and slope increases. The general slope of the region ranges between 0 and 1.8 from east to west. Maximum slope of the area is between 0 and 0.3 which occupies 6895 ha of land in the study area. Only 53 ha of land is under very high steepness factor influence which is in the hilly forested region bounded on the right hand part of the study area. <xref ref-type="table" rid="T404">Table 6</xref> shows the distribution of slope length and steepness factor and <xref ref-type="fig" rid="F714">Figure 4c</xref> demonstrates slope steepness factor.</p>
<table-wrap id="T404"><label>Table 6</label><caption><title>Slope length and steepness factor</title></caption><table frame="hsides" rules="groups">
<tbody>
<tr><td>LS-factor</td><td>Area (ha)</td></tr>
<tr><td>0 -0.3</td><td>6895</td></tr>
<tr><td>0.3 -0.6</td><td>785</td></tr>
<tr><td>0.6-1.2</td><td>442</td></tr>
<tr><td>1.2 - 1.8</td><td>214</td></tr>
<tr><td>&gt;1.8</td><td>53</td></tr>
<tr><td>Total</td><td>8390</td></tr>
</tbody>
</table>
</table-wrap>
</sec>
<sec id="s3-4"><title>3.4 Cover Management Factor (C)</title>
<p>Land use and land cover helps to understand land utilization pattern of any region and in turn land utilization helps to understand the erosion rate of that particular land cover. Thus, cover management is one of the crucial factors. The study region is mainly covered by dense forest, which is around 5448 ha in the eastern part and 0.004 is the C-factor in this region because the rainforest in this region contains huge tall trees with high density canopy cover which protects the ground below from random erosion factors (<xref ref-type="fig" rid="F714">Figure 4d</xref>). Around 1301 ha land is covered by crop land which is vulnerable because this region goes under three cropping season thus soil particles are open to more erosion this region has a cover factor of around 0.28. Urbanization has contributed towards erosion to a great extent and it is most vulnerable land use because it spoils the top soil totally and covers an artificial concrete floor. Thus, 787 ha of land in the study area gave C-factor of 1. A very small part of the land is bare which are almost a waste lands it covers an area of 183 ha and has a C-factor of around 0.18 this land is affected only by natural geomorphic agents and not by any other external agents. Land without scrub has been given 0.7 C-factor because scrub does not have much of canopy cover to protect the soil. Water body in this region covers around 554 ha which is 0 C-factor because it is the least receptive land cover towards erosion. <xref ref-type="table" rid="T405">Table 7</xref> shows crop management factor distribution.</p>
<table-wrap id="T405"><label>Table 7</label><caption><title>Crop management factor</title></caption><table frame="hsides" rules="groups">
<tbody>
<tr><td>Land use classes</td><td>C-factor</td><td>Area (ha)</td></tr>
<tr><td>Water bodies</td><td>0</td><td>554</td></tr>
<tr><td>Dense forest</td><td>0.004</td><td>5448</td></tr>
<tr><td>Land without scrub</td><td>0.18</td><td>183</td></tr>
<tr><td>Crop land</td><td>0.28</td><td>1301</td></tr>
<tr><td>Land with scrub</td><td>0.7</td><td>117</td></tr>
<tr><td>Settlement</td><td>1</td><td>787</td></tr>
<tr><td>Total</td><td></td><td>8390</td></tr>
</tbody>
</table>
</table-wrap>
</sec>
<sec id="s3-5"><title>3.5 Support Practice Factor (P)</title>
<p>The analysis clearly shows that the area can be divided into two major classes according to support factor. The crop land in this region although is prone to more erosion due to exhaustive agriculture practices but these regions also go for green manuring at frequently after each crop rotation in three different seasons therefore, P- factor in this region is 0.28 (<xref ref-type="fig" rid="F714">Figure 4e</xref>). However, rest of the land use and land covers like water body, dense forest, land without scrub, land with scrub and settlement does not have any managerial practices over this region therefore, 1 is distributed over these areas (<xref ref-type="table" rid="T406">Table 8</xref>).</p>
<table-wrap id="T406"><label>Table 8</label><caption><title>Support factor</title></caption><table frame="hsides" rules="groups">
<tbody>
<tr><td>Land use/ land cover</td><td>P-factor</td><td>Area (ha)</td></tr>
<tr><td>Crop land</td><td>0.28</td><td>1301</td></tr>
<tr><td>Water bodies</td><td>1</td><td rowspan="5">7089</td></tr>
<tr><td>Dense forest</td><td>1</td></tr>
<tr><td>Land without scrub</td><td>1</td></tr>
<tr><td>Land with scrub</td><td>1</td></tr>
<tr><td>Settlement</td><td>1</td></tr>
<tr><td>Total</td><td></td><td>8390</td></tr>
</tbody>
</table>
</table-wrap>
</sec>
<sec id="s3-6"><title>3.6 Annual Soil Loss (A)</title>
<p>Integration of soil erosion parameters in RUSLE model in GIS environment provided annual soil loss at pixel by pixel basis to study the spatial distribution of soil erosion in the study area (<xref ref-type="table" rid="T407">Table 9</xref>, <xref ref-type="fig" rid="F714">Figure 4f</xref>). The area can be graded into five vulnerability zone like very low, low, moderate, high and very high.</p>
<p>It is seen that 7116 ha area is under erosion rate of 0-2 t<sup>-1</sup>/ha<sup>-1</sup>/yr, 663 ha area is under erosion rate of 2-4 t<sup>-1</sup>/ha<sup>-1</sup>/yr, 370 ha area is under erosion rate of 4-6 t<sup>-1</sup>/ha<sup>-1</sup>/yr, 370 ha area is under erosion rate of 6-8 t<sup>-1</sup>/ha<sup>-1</sup>/yr, 186 ha area is under erosion rate of 6-8 t<sup>-1</sup>/ha<sup>-1</sup>/yr and 55ha area is under erosion rate of 6-8 t<sup>-1</sup>/ha<sup>-1</sup>/yr. Immediate attention needs to be given to the 55 ha of land which is very highly prone to soil loss. Very high erosion rate is witnessed in the lower foothill parts of the hilly mountains because the river gushes down the slope and erodes the slopes by lateral and vertical movement of water downwards. Since Nangal Dam controls the water management factors in two major agricultural states of Punjab and Haryana, it can be affected adversely any time due to the high erodibilty in the upper reaches of the study area, thus management practices needs to be implemented properly to save the natural resource in the region.</p>
<table-wrap id="T407"><label>Table 9</label><caption><title>Soil losses</title></caption><table frame="hsides" rules="groups">
<tbody>
<tr><td>Soil loss (t/ha/yr)</td><td>Area (ha)</td><td>Vulnerability</td></tr>
<tr><td>0-2</td><td>7116</td><td>Very low</td></tr>
<tr><td>2-4</td><td>663</td><td>Low</td></tr>
<tr><td>4-6</td><td>370</td><td>Moderate</td></tr>
<tr><td>6-8</td><td>186</td><td>High</td></tr>
<tr><td>&gt;8</td><td>55</td><td>Very high</td></tr>
</tbody>
</table>
</table-wrap>
</sec>
<sec id="s3-7"><title>3.7 Validation of Results</title>
<p>Several workers reported integration of RUSLE in the GIS environment sans validation of soil loss estimation, which needs to be carried out to evaluate the authenticity of the results. The study area was very challenging as most of the places were inaccessible due to geographical hindrances and being upstream of the reservoir. Therefore, validation was done by collecting maximum points possible in the lower reaches on both sides of the river and few points in the lower foothills of the forested hilly areas. Random places were chosen and total of 35 points were visited, and ground information was gathered in correspondence to their latitude and longitude values using GPS. Further a Normalized Differential Vegetation Index (NDVI) map was generated in the GIS environment which included three seasons’ datasets to extract the vegetation cover percentage in the study area and a correlation of 0.5 was found between the NDVI and the soil loss erosion map t/ha/yr (<xref ref-type="fig" rid="F719">Figure 5</xref>). This positive strong correlation suggests quite good accuracy level of the soil loss estimation map generated in the GIS environment (<xref ref-type="fig" rid="F719">Figure 6</xref>).</p>
<fig id="F717"><label>Figure 5</label><caption><title>Correlation between vegetation cover and soil losses</title></caption><graphic xlink:href="https://gathacognition.com/uploads/admin/articlefigure/3d4b0ad68fabf7acf7d822fd1ee89680.jpg" mimetype="image" mime-subtype="jpeg"/></fig>
<fig id="F719"><label>Figure 6</label><caption><title>Distribution of ground truth points with photographs</title></caption><graphic xlink:href="https://gathacognition.com/uploads/admin/articlefigure/7817c7c13bd7deaaf482e414236bfef7.jpg" mimetype="image" mime-subtype="jpeg"/></fig>
</sec>
</sec>
<sec id="s4"><title>Conclusions</title>
<p>RUSLE can be easily integrated in the GIS platform and the inputs required for this model can be derived from multi-sources like satellite based data, available maps and in-situ observations. The topographic parameters can be readily extracted from DEMs facilitating GIS based erosion modeling. In spite of advances in sensor technologies, uncertainty and coarseness of critical parameters like spatial distribution of soils, and rainfall still remain the weak components of erosion modeling using RUSLE. The model results reasonably match with the observed field survey and majority of the area falls under moderate to low risk of soil erosion. Only, 55 ha of land is under threats of very high erosion rate per year. The results obtained from the study can assist in developing management scenarios and provide options to policy makers for managing soil erosion hazards in the most efficient manner for sustainable management plans.</p>
</sec>
</body>
<back>
<ack>
<title>Acknowledgements</title>
<p>Special thanks to Director, Punjab Remote Sensing Centre, Ludhiana (India) for the support and guidance. Authors are thankful to NASA and JAXA for sharing satellite imageries in the free domain.</p>
</ack>
<glossary>
<title>Abbreviations</title>
<p><bold>ALOS</bold>: Advanced Land Observation Satellite; <bold>DEM</bold>: Digital Elevation Model; <bold>GIS</bold>: Geographical Information System; <bold>NDVI</bold>: Normalised Differential Vegetation Index; <bold>PALSAR</bold>: Phased Array type L-band Synthetic Aperture Radar; <bold>RUSLE</bold>: Revised Universal Soil Loss Equation; <bold>TRMM</bold>: Tropical Rainfall Measuring Mission.</p>
</glossary>
<fn-group>
<fn fn-type="coi-statement">
<p>The authors declare that they have no conflicts of interest associated with this publication. This research article is in compliance with the ethical standards of this journal. This article does not contain any information that requires informed consent.</p>
</fn>
</fn-group>
<ref-list>
<title>References</title>
<ref id="R4069"><mixed-citation>Aksoy, H. and Kavvas, M. L., 2005. A review of hill slope and watershed scale erosion and sediment transport models. <italic>Catena</italic>, 64, 247‑271. <pub-id pub-id-type="doi">10.1016/j.catena.2005.08.008</pub-id></mixed-citation></ref>
<ref id="R4070"><mixed-citation>Alexakis, D. D., Hadjimitsis, D. G. and Agapiou, A., 2013. Integrated use of remote sensing, GIS and precipitation data for the assessment of soil erosion rate in the catchment area of “Yialias” in Cyprus. <italic>Atmospheric Research</italic>, 131, 108-124. <pub-id pub-id-type="doi">10.1016/j.atmosres.2013.02.013</pub-id></mixed-citation></ref>
<ref id="R4071"><mixed-citation>Blanco-Canqui, H. and Lal, R., 2008. Principles of soil conservation and management. Springer Science and Business Media.</mixed-citation></ref>
<ref id="R4072"><mixed-citation>Elwell, H.A., 1977. Soil loss estimation for southern Africa (p. 22) Harare: Zimbabwe Department of Conservation and Extension, Research Bulletin.</mixed-citation></ref>
<ref id="R4073"><mixed-citation>Finlayson, D. and Montgomery, D., 2003. Modeling large-scale fluvial erosion ingeographic information systems. <italic>Geomorphology</italic>, 53(1-2), 147-164. <pub-id pub-id-type="doi">10.1016/S0169-555X(02)00351-3</pub-id></mixed-citation></ref>
<ref id="R4074"><mixed-citation>Foster, G. R., Yoder, D. C., Weesies, G. A., McCool, D. K., McGregor, K. C. and Bingner, R. L., 2003. RUSLE 2.0 user’s guide. USDA-Agricultural Research Service, Washington, DC.</mixed-citation></ref>
<ref id="R4075"><mixed-citation>Fu, B. J., Zhao, W. W., Chen, L. D., Zhang, Q. J., Lü, Y. H., Gulinck, H. and Poesen, J., 2005. Assessment of soil erosion at large watershed scale using RUSLE and GIS: A case study in the Loess Plateau of China. <italic>Land degradation and development</italic>, 16(1), 73-85. <pub-id pub-id-type="doi">10.1002/ldr.646</pub-id></mixed-citation></ref>
<ref id="R4076"><mixed-citation>Ganasri, B. P. and Ramesh, H., 2016. Assessment of soil erosion by RUSLE model using remote sensing and GIS-A case study of Nethravathi Basin. <italic>Geoscience Frontiers</italic>, 7(6), 953-961. <pub-id pub-id-type="doi">10.1016/j.gsf.2015.10.007</pub-id></mixed-citation></ref>
<ref id="R4077"><mixed-citation>Gupta, S. and Kumar, S., 2017. Simulating climate change impact on soil erosion using RUSLE model- A case study in a watershed of mid-Himalayan landscape. <italic>Journal of Earth System Science</italic>, 126(3), 43. <pub-id pub-id-type="doi">10.1007/s12040-017-0823-1</pub-id></mixed-citation></ref>
<ref id="R4078"><mixed-citation>Kent, 1972. National engineering handbook, section 4: Hydrology. Washington<italic>, </italic>DC.</mixed-citation></ref>
<ref id="R4079"><mixed-citation>Kinnell, P. I. A., 2010. Event soil loss, runoff and the Universal Soil Loss Equation family of models: A review. <italic>Journal of Hydrology</italic>, 385, 384‑397. <pub-id pub-id-type="doi">10.1016/j.jhydrol.2010.01.024</pub-id></mixed-citation></ref>
<ref id="R4080"><mixed-citation>Kinnell, P. I. A., and Risse, L. M., 1998. USLE-M: Empirical modeling rainfall erosion through runoff and sediment concentration. <italic>Soil Science Society of America Journal</italic>, 62(6), 1667-1672. <pub-id pub-id-type="doi">10.2136/sssaj1998.03615995006200060026x</pub-id></mixed-citation></ref>
<ref id="R4081"><mixed-citation>Lal, R., 1998. Soil erosion impact on agronomic productivity and environment quality. <italic>Critical reviews in plant sciences</italic>, 17(4), 319-464. <pub-id pub-id-type="doi">10.1080/07352689891304249</pub-id></mixed-citation></ref>
<ref id="R4082"><mixed-citation>Morgan, R. P. C., Quinton, J. N., Smith, R. E., Govers, G., Poesen, J. W. A., Auerswald, K., Torri, C. D. and Styczen, M. E., 1998. The European Soil Erosion Model (EUROSEM): A dynamic approach for predicting sediment transport from fields and small catchments. <italic>Earth Surface Processes and Landforms: The Journal of the British Geomorphological Group</italic>, 23(6), 527-544. <pub-id pub-id-type="doi">10.1002/(SICI)1096-9837(199806)23:6&lt;527::AID-ESP868&gt;3.0.CO;2-5</pub-id></mixed-citation></ref>
<ref id="R4083"><mixed-citation>Nyesheja, E. M., Chen, X., El-Tantawi, A. M., Karamage, F., Mupenzi, C. and Nsengiyumva, J. B., 2019. Soil erosion assessment using RUSLE model in the Congo Nile Ridge region of Rwanda. <italic>Physical Geography</italic>, 40(4), 339-360. <pub-id pub-id-type="doi">10.1080/02723646.2018.1541706</pub-id></mixed-citation></ref>
<ref id="R4084"><mixed-citation>Phinzi, K. and Ngetar, N. S., 2019. The assessment of water-borne erosion at catchment level using GIS-based RUSLE and remote sensing: A review. <italic>International Soil and Water Conservation Research</italic>, 7(1), 27-46. <pub-id pub-id-type="doi">10.1016/j.iswcr.2018.12.002</pub-id></mixed-citation></ref>
<ref id="R4085"><mixed-citation>Renard, K. G. and Foster, G. R., 1983. Soil conservation: Principles of erosion by water. <italic>Dryland Agriculture</italic>, 23, 155-176.</mixed-citation></ref>
<ref id="R4086"><mixed-citation>Renard, K. G., Foster, G. R., Weesies, G. A. and Porter, J. P., 1991. RUSLE: Revised universal soil loss equation. <italic>Journal of Soil and Water Conservation</italic>, 46(1), 30-33.</mixed-citation></ref>
<ref id="R4087"><mixed-citation>Renard, K. G., Foster, G. R., Weesies, G. A., McCool, D. K. and Yoder. D. C., 1997. Predicting soil erosion by water: A guide to conservation planning with the revised universal soil loss equation (TUSLE). <italic>Agriculture handbook,</italic> 703.</mixed-citation></ref>
<ref id="R4088"><mixed-citation>Renard, K. G., Foster, G. R., Yoder, D. C. and McCool, D. K., 1994. RUSLE revisited: Status, questions, answers, and the future. <italic>Journal of Soil and Water Conservation</italic>, 49(3), 213-220.</mixed-citation></ref>
<ref id="R4089"><mixed-citation>Shinde, V., Tiwari, K. N. and Singh, M., 2010. Prioritization of micro watersheds on the basis of soil erosion hazard using remote sensing and geographic information system. <italic>International Journal of Water Resources and Environmental Engineering</italic>, 5(2), 130-136.</mixed-citation></ref>
<ref id="R4090"><mixed-citation>Thomas, J., Joseph, S. and Thrivikramji, K. P., 2018. Assessment of soil erosion in a tropical mountain river basin of the southern Western Ghats, India using RUSLE and GIS. <italic>Geoscience Frontiers</italic>, 9(3), 893-906. <pub-id pub-id-type="doi">10.1016/j.gsf.2017.05.011</pub-id></mixed-citation></ref>
<ref id="R4091"><mixed-citation>Udayakumara, E. P. N., Shrestha, R. P., Samarakoon, L. and Schmidt-Vogt, D., 2010. People’s perception and socioeconomic determinants of soil erosion: A case study of Samanalawewa watershed, Sri Lanka. <italic>International journal of sediment research</italic>, 25(4), 323-339. <pub-id pub-id-type="doi">10.1016/S1001-6279(11)60001-2</pub-id></mixed-citation></ref>
<ref id="R4092"><mixed-citation>Van Der Knijff J. M., Jones R. J. A. and Montanarella L., 2000. Soil Erosion Risk Assessment in Europe. Ispra, European Soil Bureau.</mixed-citation></ref>
<ref id="R4093"><mixed-citation>Wener, C. G. 1972. Soil Conservation in Kenya, Nairobi.<italic> Ministry of Agriculture, Soil Conservation Extension Unit</italic> (1981).</mixed-citation></ref>
<ref id="R4094"><mixed-citation>Williams, J. R. and Berndt, H. D., 1976. Sediment yield prediction based on watershed hydrology. <italic>American Society of Agricultural Engineering</italic>. <pub-id pub-id-type="doi">10.13031/2013.35710</pub-id></mixed-citation></ref>
<ref id="R4095"><mixed-citation>Wischmeier, W. H. and Smith, D. D., 1978. Predicting rainfall erosion losses: A guide to conservation planning (No. 537). Department of Agriculture, Science and Education Administration.</mixed-citation></ref>
<ref id="R4096"><mixed-citation>Zhao, G., Gao, P., Tian, P., Sun, W., Hu, J. and Mu, X., 2020. Assessing sediment connectivity and soil erosion by water in a representative catchment on the Loess Plateau, China. <italic>Catena</italic>, 185, 104284. <pub-id pub-id-type="doi">10.1016/j.catena.2019.104284</pub-id></mixed-citation></ref>
</ref-list>
</back>
</article>
