Geospatial Information System (GIS) has been integrated with the weighted index overlay (WIO) methods for the identification and delineation of soil erosion susceptibility zones in the Kodai Hills. The high soil erosion probability zone was observed in areas with high terrain alteration, high relief and slopes with the intensity and duration of heavy precipitation during the monsoons.
Geospatial Information System (GIS) has been integrated with the weighted index overlay (WIO) methods for the identification and delineation of soil erosion susceptibility zones in the Kodai Hills. The high soil erosion probability zone was observed in areas with high terrain alteration, high relief and slopes with the intensity and duration of heavy precipitation during the monsoons.
Geospatial Information System (GIS) has been integrated with the weighted index overlay (WIO) methods for the identification and delineation of soil erosion susceptibility zones in the Kodai Hills. The high soil erosion probability zone was observed in areas with high terrain alteration, high relief and slopes with the intensity and duration of heavy precipitation during the monsoons.
Assessment of soil erosion probability in Kodaikanal, India using GIS and remote sensing Bagyaraj M. 1 , Ramkumar T. 1 , Venkatramanan S. 2* ,
Chung S. Y. 2
and Gurugnanam B. 3 1. Depart ment of Eart h Sciences, Annamalai Universit y, Annamalai Nagar-608 002, Tamil Nadu, INDIA 2. Depart ment of Eart h Environment al Sciences, Pukyong Nat ional Universit y, 599-1, Daeyeon-dong Nam, Busan - 608737, REPUBLIC of KOREA 3. Depart ment of Geology, Gandhi garm Rural Inst it ut e Deemed University, Gandhigarm-624 302, Tamil Nadu, INDIA *venkat ramanansenapat hi@gmail.com
Abstract Soil erosion hazard zone was demarcated with the help of Remote sensing data. Geographical Information System (GIS) has been integrated with the weighted index overlay (WIO) methods for the identification and delineation of soil erosion susceptibility zones in the Kodai Hills. The parameters considered for identifying the soil erosion probability zone of factors that are considered to influence the soil erosion are: soil series, land use/land cover, normalized difference vegetation index (NDVI), landform, drainage density, drainage frequency, lineament density, slope and relative relief generated using the resource sat (IRS P6 LISS IV MX) data and survey of India (SOI) toposheets of scale 1:50,000 and integrated them with weight index overlay (WIO) model based on GIS to identify the soil erosion probability zone of the study area.
In the present analysis the area is divided into three zones representing low (38.12%), moderate (50%) and high (11.89%) erosion proneness. The high soil erosion probability zone was observed in areas with high terrain alteration, high relief and slopes with the intensity and duration of heavy precipitation during the monsoons.
Keywords: Soil erosion, WIO, Geographical Information System, Remote Sensing, Kodaikanal.
Introduction Soil erosion is a serious problem throughout the world due to its adverse economic and eco-environmental impacts such as losses in land resources and decreases in land productivity, from erosion and shallow land sliding on basin slopes to sediment transport and deposition in the channel network. The management of mountain basins requires reliable methods for the analysis of sediment dynamics. 1,6,8,10,11,21,23,33,36 A detailed account of the procedure is involved in assessing soil erosion in various diverse environments along the management practices to reduce the problem 17 . The major factors that are responsible for soil erosion include climate, topography, soil characteristics, vegetation cover and land-use practices.
Soil erosion and sediment transport studies were made in the three experimental watersheds in the hilly mountains part of west Serbia (Yogoslana). The sweater sheds are characterized by similar natural features, the only difference being the degree of forest cover. The result of study shows that the degree of forest cover affects significantly the type and intensity of erosion processes as well as sediment transport. It was found that the lowest intensity of erosion and lowest sediment transport recorded in the watershed had more forest covers 18 . The over view of soil erosion control measures adopted in Hungary followed by a case study of the soil erosion process of a small watershed in the northern sub-continent of Lake Balaton has found only two per cent of the calculated soil loss actually observed in the catchment 16,35 . The amount of sediment eroded away from sites of road construction in hilly areas is ten times greater than the amount derived from agricultural land and about 200 times greater than from grassy area and about 2000 times greater from the forested area. This has distributed the stability of the hill side, inflicting serious damage to the hydrologic system. 7
The Arim Gadh catchment near Missouri has found that sediment yield from slopes vegetated with scrubs was 5 - 7 times greater than those with reserved forest 8 . Once the perennial vegetation off ores are removed and brought under cultivation, the slope plays an important role in erosion. The soil wash in cultivated areas is estimated to be 6 10 cm per year and in tea gardens it was found to be 30- 100 cm per annum 5 . Deforestation is the dominant factor causing rapid soil erosion 28 . The other factors, which have accelerated soil erosion in the area, include high altitudinal range, geology and heavy rainfall. Assessment of erosion status of a watershed is an essential pre-requisite of integrated watershed management. Due to the complexity of the variables involved in erosion, it becomes difficult to measure or predict the erosion in a precise manner.
The latest advances in remote sensing and geographical information technologies have provided very useful methods of surveying and identifying various aspects of watershed terrain behavior and also the integrated modelling approach utilizing the parameters controlling soil erosion is the effective mean of practical assessment of soil erosion hazard. Several studies carried out in different parts of the world have demonstrated capability of GIS technique for quantitatively assessing soil erosion hazard based on various approaches and equations. 9,12,19,20,22,25,26,32,37,38
Disaster Advances Vol. 7 (2) February 2014 37
To estimate the average annual soil loss from an area, RUSLE is often used. To adopt the RUSLE, large sets of data starting from rainfall, soil, slope, crop and land management are needed in detail. In developing countries all the necessary data are often not available or require ample time, money and effort to prepare such data sets. In the present study, an attempt has been made to assess the spatial distribution of potential soil erosion zones and rate of soil erosion at a of scale 1:50,000 covering Kodaikanal area by an efficient, fast and simple methodology using the remote sensing and GIS data integration and analysis, despite the lack of direct observation data.
Study Area General Setup: Study area (Fig.1) of Kodaikanal hills is located in the Dindigul district of Tamil Nadu with an area of 1039.46 km. 2 It is geographically located between 77 14 26 and 77 45 28 E longitudes and 10 6 25 and 10 26 54 N latitudes. In the survey of India toposheet, it forms part of 58 F/7, 8, 11 and 12 on 1:50,000 scale. It is part of the Western Ghats which is designated as one of the eighteen biodiversity hotspots of the world. On a tourism perspective, the region is widely hailed as the Princess of Hill Stations owing to the popularity it gained during British Raj. The climatic condition of the study area is characterized by humid conditions. Relative humidity is high during retreating NE monsoon season (October to mid December). The mean temperature of Kodaikanal taluk is 15.93 0 C with a mean summer temperature of 17.29 0 C (June, July and August) and mean winter (December, January and February) temperature of 14.10 0 C. The average annual rainfall is 1436.87 mm. The maximum elevation is 2517mts located in the south west portion.
Geology: The Kodaikanal hills form one of the chief charnockite massifs which are bluish grey and coarse grained more or less of uniform type of charnockite formation of older granulites and traversed at places by pink granites. The geological sequences of the rock are observed as mentioned below:
Recent and sub recent-Black humus soil, laterite and lithomorgic clay. Archean - Pegmatite, quartz veins and pink granite. Feldspathic gneiss (leptynitictype), Charnockite Pyroxene granulite, Calc-granulite.
Charnockite rocks in the Kodaikanal hills have been intensely altered resulting in the formation of laterite and lithomorgic clays. This is due to the intensive tropical conditions of the area and extensive laterisation of the parent rock. During heavy precipitation in rainy seasons, the capping laterite and clay materials have been eroded to down side thereby attributing to the denudation of the plateau. In the Kodaikanal hill ranges, a brown colored humus material is found overlaying the lateritic material.
Methodology: In the present study resource sat (IRS P6 LISS IV MX) sensor data of 7 March 2006 were collected and registered to Survey of India (SOI) topographical sheets at 1:50,000 scale in the ERDAS image analysis system version 9.3. In the survey of India toposheet, it forms part of 58 F/7, 8; 11 and 12 on 1:50,000 scales used for the preparation of spatial databases and they are land use/land cover, normalized difference vegetation index (NDVI), landform, drainage density, drainage frequency, lineament frequency and the topographic attributes of the region such as slope and relative relief. Each category is subdivided into different classes by its value or feature for the identification of soil erosion probability zones methodology flow chart (Fig. 2).
Drainage density and drainage frequency were classified into different orders. In GIS, drainage channel segments were ordered numerically as order number 1 from a streams headwaters to a point downstream. The stream segment that results from the joining of two first order streams was assigned order 2. Two second order streams formed a third order stream and so on. The sub basin area, perimeter, cumulative length of streams and basin length were measured in GIS and are expressed as A, P, L and Lb respectively. Parameters such as ruggedness number (Rn), drainage density, bifurcation ratio, stream frequency (Fu), texture ratio, form factor (Rf), circulatory ratio (Rc), elongation ratio (Re) and constant of channel maintenance (C) were evaluated with established mathematical equations. 34
Rainfall runoff erosive factor: Rainfall data were obtained from the Bayan Lepas weather station belonging to Malaysian Meteorological Services Department. The monthly average rainfall data of 10 years (20022011) was used to calculate the R factor as per formula 2 : 12 Rainfall factor = 4.17 * (Pi 2 /P) 152
(i=1)
where Pi is the monthly average rainfall (mm) for the month i, P is the annual average rainfall (mm) and represents the annual average R. When P and Pi values are substituted from rainfall data in the above equation, the resultant value of R can be obtained. R factor was obtained from Kodai weather station and calculated using above equation. For Kodaikanal it is 74.60 mm/ha. This constant factor was used for the entire study area.
The rainfall data for the period of 2002 to 2011 were collected in the Statistical Department wing (PWD), Govt. of Tamil Nadu. The rainfall data ware assessed for all the seasons. These results were taken into GIS platform to prepare the spatial distribution maps. Winter, Summer, Southwest and Northeast monsoon seasons spatial distribution maps result reveal that 903.58 km 2 , 840.99 km 2 , 894.99 km 2 and 889.17 km 2 area fall in high rainfall received shadow zone respectively. Last 10 years (1998 to Disaster Advances Vol. 7 (2) February 2014 38
2007) annual average rainfall spatial distribution map result reveals that 883.05 km 2 areas fell under high rainfall shadow zone with more than 766 mm rainfall. It shows the annual average rainfall distribution is very high in the study area. The high rainfall domain occupies 84.9 % of the study area (Fig.3).
Soil characteristics of different landforms: The study area is having six types of soil series (Soil Survey and land use Organization) Cavaiparai-Samayanure, Shavadiparai- Vylogam, Irugur-miscellaneous, Irugur-Palaviduthi, Irugur- Vyogam and Pilamedu-Vylogam. The Chavadiparai- Samayanure soil series moderate steep slopes are the main causative factors for these soils. The Chavadiparai- Vylogam soil series covered in a small area (0.30%) includes steep slope, drainage density and stream frequency as the influencing factors for the genesis of this soil. Irugur- miscellaneous soil series having moderate slope, high drainage density, are the main causative factors for these soils. These soils occupied 5% of the study area.
Irugur-Palaviduthi covers a majority of the study area of 65.82% (Fig. 4). The high drainage density and steep slopes are the main causative factors for these soils. These soils are in association with structural hill areas. The Irugur- Vylogam soils having steep to very steep slopes, very high drainage density and stream frequency are the main contributing factors for the development of Irugur- Vylogam soils. The Pilamedu-Vylogam soil series covers 3.41% of the total geographical area. It may be seen that the soil series Cavaiparai-Samayanure, Irugur- miscellaneous and Pilamedu-Vylogam soil series is more prone to erosion than any other unit.
Land use/land cover: The land use/land cover pattern of the area was very important because Western Ghats are designated as one of the eighteen biodiversity hotspots of the world. As a biodiversity hotspot it has the variety of endemic flora and fauna. Its richness of epiphytes, orchids and thirteen varieties of shola forests are unique. Besides, this area is prone to the erosion, a phenomenon of debris flow associated with torrential rain falls during the monsoons. In the present study, the standard methods of visual interpretation of remote-sensing data were adopted to demarcate the various zones of natural and manmade patterns. The various land use/land cover classes delineated to include Forest, Agriculture, Waste Land and Harvested Land and built-up land water body (Fig. 5).
Fig. 1: Study area map Disaster Advances Vol. 7 (2) February 2014 39
Fig. 2: Flow chart of the present study
Fig. 3: Annual Average Rainfall SDSS Map
Disaster Advances Vol. 7 (2) February 2014 40
Fig. 4: Classes of soil series map
Fig. 5: Classes of land use/land cover map
Disaster Advances Vol. 7 (2) February 2014 41
Landforms: The geographical features were interpreted from the satellite imagery and the features and GIS database was generated for dissected plateau, structural hill, structural valley, valley fill and pediment. Dissected plateaus occupied almost 24.84% of total geographic areas. Structural hills have been identified in almost all the study area and are delineated based on the image characteristics viz. valley fills are delineated with the help of satellite image. It is having more vegetation. The valley fills the cover in small areas. The composite slopes are also identified. From the ranking awarded above it may be seen that the geomorphological unit Structural Hill and Pediment is more prone to slides than any other unit (Fig.6).
Drainage density and drainage frequency: Drainage density is considered to be one of the important parameters for assessing the soil erosion zone in an area. Since drainage density is a useful index for understanding the nature of the surface material and their permeability and infiltration characteristics, drainage density has been used in conjunction with other parameters such as slope geomorphology, rock types, lineament density etc. Hill's slope evolution in an area is controlled by the sediment transport processes which change in response to the evolving topography and by their interaction with stream processes at the slope base. 29 In mountainous regions, drainage density provides an indirect measure of groundwater conditions which have an important role to play in landslide and other erosion activities. 24,30,31
Drainage density is a measure of stream spacing and a higher drainage density represents a relatively higher number of streams per unit area and thus a rapid storm response. It also represents conditions favorable for higher erosion from the catchment.
The simplest way to calculate drainage density boon a regional scale is to divide the study area into grid squares of one sq.km each and to measure the total stream length in each grid square and to group the derived data in drainage density categories. The methodology stated above has been used to obtain a drainage density map of the study area. The values obtained range from 150 to 3,959 m/km 2 finally classified into three classes of high (>3,000 m/km 2 ), moderate (2000 to 3000 m/km 2 ) and low (<2000 m/km 2 ) density (Fig.7). The stream frequency is controlled by various factors like climate, the amount of precipitation, vegetation cover, permeability of mantle and bedrock. It is also influenced by the relief of the terrain. In the present study, the area has been divided into one sq.km grid and the number of streams in one sq. km counted based on this. The area has been divided into three groups in the stream frequency map (Fig.8):
1) Low (< 7 / sq.km) 2) Medium (7 10 / sq.km) 3) High (> 10 / sq.km).
Lineament Density and Lineament Frequency: The study of lineament density may help to identify the weathered zones in an area which is very essential in the studies relating to, soil erosion, landslides etc. The lineament density map for the present study area was prepared using lineament map prepared by interpretation of satellite data. The lineament map was prepared on 1:50,000 scales overlaid by a 2cm grid sheet. The total length of the lineament in each grid was noted down and the process was repeated for all the grids covering the study area. By carefully examining the values obtained, the data were grouped into four classes as low (lineament density less than 1000 m/sq. km, moderate (1000-1500m/sq. km.), high (1500-2000 m/sq. km.) and very high (> 2000 m/sq. km, Fig. 9). General trend shown by the lineaments present in the study area are NNESSW and NESW. The lineament frequency map was generated using the Spatial Analyst extension of ArcGIS. The raster layer obtained was reclassified into two classes on using the reclassification tool.
Slope and relative relief: The slope map of the study area was prepared by adopting the widely used Wentworths average slope method. The slopes affect the velocity of both surface and subsurface flow and hence soil water content, soil formation, erosion potential and a large number of important geomorphic processes. Digital elevation model (DEM) is derived using contour information from the topographical map for estimation of slope in degrees. The identified slope category varies from 0 o to >30 o degree in the study area and are classified into five classes like, 0 o 10 o (gentle), 10 o 20 o (moderate), 20
o 25 o (high), 25 o 30 o (very high) and >30 o (steep) (Fig.10).
Elevation is useful to classify the local relief and locate points of maximum and minimum heights within terrains. Relative relief portrays the difference in elevation at a given point. The factor of safety decreases with the increase in height. Thus, for two slopes having identical geo- mechanical and geometrical parameters except for height, the higher slope will be more susceptible to erosion and landslide. Run-off is higher and infiltration is lower in areas of steeper topography.
In addition, saturation of a slope reduces the shear resistance of the regolith and increases the shear forces through drag. 24,27,30 Survey of India provided base layers in a digital format on 1:50000 scale. A DEM was prepared using line contours of 20 meter interval. The elevation of the area depicted in clearly shows that the Kodai hill rises abruptly from the adjacent plains from about 300 m to 1000 m. in some parts of the hills such as abruptness continues even up to 2500 m (Fig.11) The relative relief map thus generated shows a value ranging between 0 and 2500 m/km 2 and reclassified into class 1 (<300 m/km 2 ); class 2 (300500 m/km2); class 3 (>500 m/km 2 ).
Disaster Advances Vol. 7 (2) February 2014 42
Normalized difference vegetation index (NDVI): NDVI calculated from the visible and near-infrared light reflected by vegetation and the healthy vegetation absorbs most of the visible light that hits it and reflects a large portion of the near-infrared light. Unhealthy or sparse vegetation reflects more visible light and less near-infrared light. Calculations of NDVI for a given pixel always result in number that ranges from minus one (1) to plus one (+1); however, no green leaves give a value close to zero. A zero means no vegetation and close to +1 (0.80.9) indicates the highest possible density of green leaves. The IRS P6 LISS 3 multispectral image was used with the vegetation index function available in the ERDAS Imagine 9.2 software to derive the NDVI image. To avoid the negative values and for easy handling of digital data, NDVI values obtained for IRSP6 L3 data (23.5 m spatial resolution) were rescaled as shown in eq. 1:
NDVI = [(Band 4-Band 3/Band 4 + Band 3) + 1] (1)
Thus produced rescaled NDVI map shows a range of values between 0.63 and 1.75 in which the low values are characteristics of cleared areas and zones with sparse vegetations and the higher values indicate densely covered areas.
Results and Discussion Delineation of soil erosion probability zonesweighted index overlay method (WOI): The ranking of themes done in an unbiased manner is based on the frequency of occurrence of soil erosion under a particular theme. Each thematic map such as land use/land cover, land form, drainage density, drainage frequency, lineament frequency, slope and relative relief provides certain clues for the preparation of soil erosion probability zones. Thus the major processes involved are theme weightage fixing and their further analysis in GIS platform. The weightages of individual themes and feature score were fixed and added to the layers by considering their role in the soil erosion. GIS modelling technique of index overlay method was used to produce soil erosion map. In this modelling method the map classes occurring in each input map are assigned by different scores in addition to the maps themselves receiving weights. The average score is then defined by:
S = Sij Wi / Wi 3
where S is the weight score of an area object (polygon, pixel), Wi is the weight for the its input map and Sij is the rating score of the class map.
The process involves raster overlay analysis and is known as weighted index overlay (WIO) of several methods available for determining interclass/inter-map dependency, a probability weighted approach has been adopted that allows a linear combination of probability weights of each thematic map (Wt). Different categories of derived thematic maps have been assigned scores (Wf), depending upon their role in making the terrain susceptible to soil erosion. 37 The maximum value is given to the feature with highest susceptibility and the minimum being to the lowest susceptible feature (Table 1).
The procedure of weighted linear combination dominates in raster-based GIS software systems. Spatial analyst extension of ArcGIS 9.3 was used for converting the features to raster and also for final analysis. Then using raster calculator, all the themes are added and the soil erosion prone area map is prepared. In this method, the total weights of the final integrated map were derived as sum or product of the weights assigned to the different layers according to their susceptibility.
Assessment of annual soil erosion rateE30 model using NDVI and slope: The proposed method provides a greater flexibility to estimating the soil erosion rate for any location within the study area because by this method the soil erosion rate for each of the pixels could be estimated individually. 13-15 The soil erosion model given in eq. 2 was used to estimate the annual rate of soil erosion in the area under investigation. This model is mainly governed by slope gradient and vegetation index and the annual soil erosion rate (E) is defined as:
E = E 30 (S/S 30 ) 0.9 (2)
where S = gradient of the point under consideration, S 30 = tan (30) and E 30 =rate of soil erosion at 30 slope and defined as given in eq. 3:
The maximum and minimum (average) rate of soil erosion at 30 slope in the study areas collected from the field stations were 22.25 mm/year and 0.562 mm/year in the study area as shown in eq. 3. By calculating the E 30 value for each pixel using eq. 3, soil erosion from each pixel with a different slope was calculated using eq. 2. A raster map of slope gradient was prepared with the pixel size of 20 m, using a DEM to provide the slope information for eq. 2. The final map thus produced has given a continuous raster with values varying from pixel to pixel indicating the soil erosion rate in the study area. Thus by using the GIS-based proposed methods, soil erosion probability zones and soil erosion rate map were prepared.
Finally, the erosion probability zone assessment of the Kodaikanal area was carried out using seven different parameters which are generally accepted as soil erosion contributing factors and widely used in soil erosion probability zones assessment studies in the literature. These are soil series, land use/ land cover, landform, drainage density, drainage frequency, lineament density, slope and relative relief slope. The integrated final map has generated a range of values from 1662 which is reclassified into Disaster Advances Vol. 7 (2) February 2014 43
three zones (Fig.12) based on the quintile classification method available in the spatial analyst option.
Fig. 6: Classes of landform map
Fig. 7: Classes of drainage density map
Disaster Advances Vol. 7 (2) February 2014 44
Fig. 8: Classes of drainage frequency map
Fig. 9: Classes of lineament density map
Disaster Advances Vol. 7 (2) February 2014 45
Fig. 10: Classes of slope map
Fig. 11: Classes of relative relief map
Disaster Advances Vol. 7 (2) February 2014 46
Table 1 Theme and feature class weights of Kodaikanal
Theme Theme weight (Wt) Feature class Feature class weight (Wf) Rainfall 20 High Rainfall Medium Rainfall 10 7 Soil series
18 300 m 0 < 300 5 300 - 500 m 8 > 500 m 10 Disaster Advances Vol. 7 (2) February 2014 47
Fig. 12: Soil erosion zone map (In figure Km2 should be Km 2 )
In quintile classification, the range of possible values is divided into unequal-sized intervals so that the number of values is the same in each class. Classes at the extremes and middle have the same number of values. Because the intervals are generally wider at the extremes, this option is useful to highlight changes in the middle values of the distribution.
The soil erosion probability of the area is classified as high, moderate, and poor. The high erosion probability zones occupy 38.12% of the total area; moderate and low soil erosion prone zones occupy 50% and 11.89% of the study area respectively. After generating the soil erosion probability zones, it is very important to identify the type of individual feature classes which play a vital role in the making the area vulnerable to soil erosion. In the present analysis it was found that land use/land cover types such as cleared areas, crop land and rubber plantations, particularly in replanting time present in the slide slope plateau, highly elevated areas with high slope and high drainage density make the terrain more prone to soil erosion. The rate severity and nature of the erosion will be more unpredictable at the time of monsoon seasons.
A quantitative assessment of average annual soil loss on grid basis was made using a new methodology known as WIO and NDVI slope of the area. Lack and non- availability of data needed to process the RUSLE method necessitated the application of the proposed methodology in the study area to assess the spatial distribution of rate of soil erosion in the studied sub watershed. The use of remote sensing data and digital elevation model in GIS and ERDAS enabled the determination of the spatial distribution of the parameters needed for the analysis.
The overall estimated soil erosion rate in the study area was varying from 0.04 mm year 1 to 61.80 mm year 1 with an average of 30.92 mm year 1 . The spatial patterns of soil erosion rate were overlaid with soil erosion probability map of the area to cross validate the accuracy of both the maps and it was observed that areas with high soil erosion rate and high erosion probability zones were showing similar spatial domains and patterns. The result of cross validation of both maps indicates the accurate choice of parameters and methodology for the present study.
Conclusion The study area is characterized by undulating topography dominated by high-altitude hills with an upland plateau flanked by steep slopes. Major portion of the area is covered with cultivated land followed by forest land. The implementation of WIO and E 30 method enables to classify the area into different zones based on probability of soil erosion and the rate of soil erosion in each pixel is Disaster Advances Vol. 7 (2) February 2014 48
ultimately helpful to derive suitable protection measures. The implementation of WIO and NDVI
method enables to classify the area into different zones on the basis of probability of soil erosion and the rate of soil erosion in each pixel is ultimately helpful to derive suitable protection measures.
The maximum rate of soil erosion is estimated to be 61.80 mm year 1 and this corresponds to areas with high soil erosion probability (38.12% of the total area). The generated soil erosion probability image, predicted amount of soil erosion rate and its spatial distribution can provide a basis for comprehensive and sustainable land management for the study area. Spatial relationships between erosion rate and elevation indicate that severe and extreme soil erosion is characteristic of plateau margins and highly undulating hills with varying terrain alteration where no conservation practices exist. Therefore, these areas, especially the extreme and severe erosion zones, need immediate attention from soil conservation point of view.
Acknowledgement We express our sincere thanks to University Grant Commission (UGC), Government of India for extending the financial support to carry out this work.
References 1. Ande O.T., Alaga Y. and Oluwatosin G.A., Soil erosion prediction using MMF model on highly dissected hilly terrain of Ekiti environs in southwestern Nigeria, Inter Jour of Physical Sci., 4(2), 5357 (2009)
2. Arnoldus H. M., An approximation of the rainfall factor in the USLE, In Boodt M. de and Gabriels D., eds., Assessment of erosion, Chichester, Wiley, 127132 (1981)
3. Bonham-Carter G. F., Geographic Information Systems for geoscientists, Modelling with GIS, Elsevier Science Ltd., 398 (1994)
4. Carson B., Erosion and Sedimentation Process in the Nepalese Himalayas, ICIMOD Occasional Paper no.1, Kathmandu, 1-14 (1985)
5. Chinnamani S., Soil and Water Conservation in the Hills of Western Ghats, In Ramesh A., ed., Resource Geography, Heritage Publishers, New Delhi, 235- 21I (1984)
6. Dabral P. P., Baithuri N. and Pandey A., Soil erosion assessment in a hilly catchment of North Eastern India using USLE, GIS and remote sensing, Water Resour Manage, 22, 17831798 (2008)
7. Dendy E.E., Sedimentation in Nations Reservoirs, Journal of soil and water conservation, 23, 135-137 (1968)
8. Hlaing K.T., Haruyama S. and Aye M.M., Using GIS-based distributed soil loss modeling and morphometric analysis to prioritize watershed for soil conservation in Bago river basin of Lower Myanmar, Front Earth Sci. China, 2(4), 465478 (2008)
9. Ismail J. and Ravichandran S., RUSLE 2 model application for soil erosion assessment using remote sensing and GIS, Water Resour Manage, 22, 83102 (2008)
10. Jain A.K. and Kothyari U.C., Sediment yield estimation using GIS, Hydrol Sci J.,45(5), 771786 (1997)
11. Jain S.K., Singh P., Saraf A.K. and Seth S.M., Estimation of sediment yield for a rain, snow and glacier fed river in the western Himalayan region, Water Resour Manage, 7, 377393 (2003)
12. Jianrong F., Bingwei T. and Dong Y., Cause analysis of gully erosion in Yuanmou Basin of Jinshajiang valley, Wuhan Univ. J Nat Sci., 13(3), 343349 (2008)
13. Hazarika M.K. and Honda K., Estimation of soil erosion using remote sensing and GIS, its calculation and economic implications on agricultural production, In Stott D.E., Mohtar R.H. and Steinhardt G.C., eds., Sustaining the global farm, USDAARS national soil erosion research laboratory, Purdue, 10901093 (2001)
14. Honda K., Samarakoon L., Ishibashi A., Mabuchi Y. and Miyajima S., Remote sensing and GIS technologies for denudation estimation in Siwalik watershed of Nepal, In Proceedings of the 17th Asian Conference on Remote sensing, Colombo, Sri Lanka, 48 th November, B21B26 (1996)
15. Honda K., Samarakoon L. and Ishibashi A., Erosion control engineering and geoinformatics, river planform changes and sediment yield estimation in a watershed of Siwalik, Nepal, In Singh R.B. et al, eds., Space informatics for sustainable development, Oxford & IBH Publishing Co. Pvt. Ltd., New Delhi, 6370 (1998)
16. Kertesz A. and Loezy D., Soil Erosion Control in Hungary, In Josef Krecek, Rajwar G.S. and Martin J. Haigh, eds., Hydrological Problems and Environmental Management in Highlands and Headwaters, Oxford and IBH Publishing Co. Pvt. Ltd., New Delhi, 63-70 (1996)
17. Kirkby M.J. and Morgan R.P.C., Soil erosion, John Wiley and Sons Ltd., New York, 178 (1980)
18. Kostadinov S., Soil Erosion and Sediment Transport Depending on Land use in the Watershed, In Josef Krecek, Rajwar G.S. and Martin J. Haigh, eds., Hydrological Problem and Environmental Management in Highlands and Headwaters , Oxford and IBH Publishing Co. Pvt. Ltd., New Delhi, 43-51 (1996)
19. Kouli M., Soupios P. and Vallianatos F., Soil erosion prediction using the revised universal soil loss equation (RUSLE) in a GIS framework, Chania, Northwestern Crete, Greece, Environ Geol., 57, 483497 (2009)
20. Krishna Bahadur K.C., Mapping soil erosion susceptibility using remote sensing and GIS, a case of the Upper Nam Wa Watershed, Nan Province, Thailand, Environ Geol., 57, 695705 (2009)
Disaster Advances Vol. 7 (2) February 2014 49
21. Marchi L. and Fontana G.D., GIS morphometric indicators for the analysis of sediment dynamics in mountain basins, Environ Geol, 48, 218228 (2005)
22. Millward A.A. and Mersey J.E., Adapting the RUSLE to model soil erosion potential in a mountainous tropical watershed, Catena, 38, 109129 (1999)
23. Mitasova H., Hofierka J., Zlocha M. and Iverson R.L., Modeling topographic potential for erosion and deposition using GIS, Inter Jour of Geograph Inform Scie., 10(5), 629641 (1996)
24. Nagarajan R., Roy A., Vinod Kumar R., Mukherjee A. and Khire M.V., Landslide hazard susceptibility mapping based on terrain and climatic factors for tropical monsoon regions, Bull Eng Geol Environ, 58, 275287 (2000)
25. Naik M.G., Rao E.P. and Eldho T.I., Finite element method and GIS based distributed model for soil erosion and sediment yield in a watershed, Water Resour Manage, 23, 553579 (2009)
26. Onyando J.O., Kisoyan P. and Chemelil M.C., Estimation of potential soil erosion for river perkerra catchment in Kenya, Water Resour Manage, 19, 133143 (2005)
27. Pandey A., Chowdary V.M. and Mal B.C., Identification of critical erosion prone areas in the small agricultural watershed using USLE, GIS and remote sensing, Water Resour Manage, 21, 729 746 (2007)
28. Pant B.R., Joshi R.C and Jalal D.S., Erosion in Hill Environment. Himalayas, Man and Nature, 12, 18-19 (1988)
29. Rajkumar P., Sanjeevi S., Jayaseelan S., Isakkipandian G., Edwin M., Balaji P. and Ehanthalingam G., Landslide susceptibility mapping in a hilly terrain using remote sensing and GIS, Jour Indian Soc of Rem Sen, 35(1), 3142 (2007)
30. Saha A.K., Gupta R.P., Sarkar I., Arora M.K. and Csaplovics E., An approach for GIS based statistical landslide susceptibility zonation with a case study in the Himalayas, Landslides, 2, 61 69 (2005)
31. Sarkar S. and Kanungo D.P., An integrated approach for landslide susceptibility mapping using remote sensing and GIS, Photogramm Eng Remote Sensing, 70(5), 617625 (2004)
32. Sharma J.C., Prasad J., Saha S.K. and Pande L.M., Watershed prioritization based on sediment yield index in eastern part of Don valley using RS and GIS, Indian J Soil Conserv., 29(1), 7 13 (2001)
33. Singh O., Sarangi A. and Sharma C.M., Hypsometric integral estimation methods and its relevance on erosion status of north- western lesser Himalayan watersheds, Water Resour Manage., 22, 15451560 (2008)
34. Strahler A.N., Quantitative geomorphology of basins and channel networks, In Chow V.T., ed., Handbook of Applied Hydrology, McGraw Hill Book Company, New York (1964)
35. Lin Li-Ling Weng, Chih-Cheng, Wang Chao-Wen, Shen Che- Wei and Chen-Fa Wu, SVM and BPN models for predicting Soil Erosion Degree in 921 Earthquake Slopeland Region in Taiwan, Disaster Advances, 5(4), 278-284 (2012)
36. Tehagehishvili R.G., Sediment transport and small watershade with different degrees of forest cover in the mountains of Georgia in Lesovedenie No.5, Tbilishi, Georgia, 72-77 (1977)
37. Tian Y.C., Zhou Y.M., Wu B.F. and Zhou W.F., Risk assessment of water soil erosion in upper basin of Miyun Reservoir, Beijing, China, Environ Geol., 57, 937942 (2009)
38. Vijith H., Suma M., Rekha V. B., Shiju C. and Rejith P. G., An assessment of soil erosion probability and erosion rate in a tropical mountainous watershed using remote sensing and GIS, Arab J Geosci., DOI 10.1007/s12517-010-0265-4 (2011)
39. Zhao X.L., Zhang Z.X., Liu B. and Wang C.Y., Method of monitoring soil erosion dynamic based on remote sensing and GIS, Bull Soil Water Conserv., 22, 2932 (2002).
(Received 15 th September 2013, accepted 20 th December 2013)