Evaluation of Soil Loss and Identification of Dominant Factors in the Semi-Arid Region Using GIS: A Case of Bhavnagar District, Gujarat, India

Abstract

The loss of valuable topsoil worldwide has led to agricultural land degradation and a reduction in crop yields. Soil erosion is mainly caused by both natural phenomena and human interference with the ecosystem. The efficiency of spatial information systems like GIS and RS has been effectively developed. From 2015 to 2021, soil loss estimation was conducted using the Revised Universal Soil Loss Equation (RUSLE) model, with remote sensing and geographic information systems (GIS) assistance. We have produced the RUSLE model\'s five essential potential parameters (R*K*LS*C*P) pixel-by-pixel. We generated the R factor map through the Indian Meteorological Department\'s (IMD) daily rainfall data, and the K factor map using the FAO\'s digital soil series global map. For the LS-factor map, we used the digital elevation model data (DEM) of SRTM. Landsat 8 dataset was used to generate LULC, and NDVI maps to derive C and P factors. The highest mean annual soil loss in 2021 was estimated to be 209.16 tons per hectare per year. The riverbank area of the Shetruji River near Palitana Taluka in this district had an extremely high risk of soil loss. The coastal area from Bhavnagar to Dholera was classified as a high and very high-risk area for soil loss, which contained barren land. The results revealed that barren land is the most susceptible to soil erosion. As per statistical analysis, the C factor is the dominant factor in this region which is most influential in soil erosion. The soil erosion maps from this study will provide policymakers with the necessary information to implement suitable conservation measures in this region.

Introduction

I. INTRODUCTION

Soil erosion is one type of land degradation in which soil particles separate from their parent materials through erosive agents like water and wind, and overland flow occurs. In this phenomenon, soil nutrients lose their fundamental originality to maintain agricultural productivity due to the loss of top cover of soil. Land degradation is one of the global issues nowadays. Water is the prime erosion-causing factor. In India, the soil erosion problem is multiplied day by day due to water. An increase in population is also one of the dominant reasons for soil erosion because of the increase in demand for agricultural products to feed the huge population, so haphazard use of land increase soil loss. In addition to overlooking the fact that land degradation is fundamentally a physical process, popular theories of human-induced soil erosion and land degradation also do a disservice to adaptive ecosystem management by the local population[1]. Basically, this study includes types of erosion sheet erosion because of straightforward estimation methods available in the numerous kinds of literature. Physical models, physical-experimental models, and empirical models are used for the assessment of soil erosion. Conventional methods used for the assessment of soil erosion in semi-arid regions are complex and time-consuming exercises while the study area is about like a district. The universal soil loss equation is a widely accepted phenomenon under the empirical models’ category which is suggested by [2]. A lot of innovation and revision is done under this USLE equation as per local meteorological conditions and parameters. The USLE equation is popular worldwide for estimating soil loss from croplands [3]. Improved estimates of global soil erosion and soil organic carbon pools, including the effects of land-use changes and conservation agriculture in past, present, and future scenarios, can be made possible by a USLE-based model that is applicable globally [4]. The MUSLE equation was developed in this family but, this equation is not used to predict the spatial distribution of the study area for the soil loss [5]. A revised universal soil loss equation was suggested by [6] for the soil loss assessment in small/large areas. This revised USLE model is very helpful to predict soil loss on the verge of fewer data and is highly compatible with Remote sensing (RS) and geographical information systems (GIS) [7], [8]. To get the best results from the RUSLE model used with a GIS application [9]–[11].

The digital maps and data which were prepared using GIS and RS to get accurate and significant results from the RUSLE model are more useful in planning, management of natural resources, and deciding policy about conservation measures [12], [13].

In this study, the RUSLE model integrated with the GIS platform is used for the determination of annual soil loss with an estimation of its factors using rainfall data, FAO soil map, SRTM DEM, and LULC map after studying numerous kinds of literature.

II. STUDY AREA

The study area is a Bhavnagar district located in the Saurashtra Region of Gujarat. The total geographic area of Bhavnagar district is 7034 km2. The latitude and longitude of this area are situated at 21.7645° N, and 72.1519° E, and having average elevation is 24 m from the Mean Sea level. Fig.1 shows the location of the study area.

Bhavnagar district is one of the most developed districts of the Saurashtra region of Gujarat State. Its district headquarter is located in Bhavnagar City. Bhavnagar was founded by Bhavsinhji Gohil (1703-64 AD) in 1723 AD near the Gulf of Khambhat, in a carefully chosen strategic location having the potential for maritime trade. Bhavnagar is bordered by Ahmedabad, Surendranagar, and Botad districts in the North, the Gulf of Cambay in the East and South, and Amreli and Rajkot districts in the West. Bhavnagar District has 9 talukas like Gariyadhar, Ghogha, Jesar, Mahuva, Palitana, Sihor, Talaja, Umrala, and Vallabhipur. Among these talukas, Palitana and Talaja territories have mountainous regions, and Ghogha and Mahuva territories have coastal regions. Most of the area is underlain by the Deccan Trap and alluvium. The district is characterized by a tropical climate with general dryness, except in the coastal areas, and falls under the semi-arid region category. Shetruji River is the prime source of water for the territory of the study area which is situated in Palitana Talukas of the Bhavnagar District.

There are four seasons in a year, viz., the hot season from March to May, the monsoon season from June to September, the post-monsoon season from October to November, and the cold season from December to February. The average annual rainfall of this region is 519 mm as revealed by rainfall data collected from the Indian Meteorological Department (IMD), Pune. Soil Characteristics of this area are Shallow medium black soil and coastal alluvial soil.

III. Data Used

For Bhavnagar District Rainfall data of 0.250 x 0.250 resolution were used from the Indian meteorological department (IMD), Pune. For the present study Soil data were obtained from the FAO global soil map of 1:5000000 resolution. SRTM DEM 30 m resolution is used in this study. For Land Use and Land cover features MODIS Land cover images were used. NDVI Map obtained from the LANDSAT-8 images from USGS earth explorer. Table.1 depicts data collection and its source used in this study.

The region with conservation efforts is shown by the low P factor. Their biggest value is seen in non-conservation regions like developed land and plantation areas with strip and contour cropping. In this study, the P factor value ranges from 0.29 to 1.

F. Annual Soil Loss (A)

For estimating the annual soil loss, the Rainfall Erosivity Factor (R), Soil Erodibility Factor (K), Slope Length and Steepness Factor (LS), Cover and Management Factor (C), and Support Practice Factor (P) parameters of the Revised Universal Soil Loss Equation (Renard et al., 1991) were integrated (A). The annual soil erosion rate is then categorized into risk classes based on its magnitude: Low risk (0–20 t ha-1 yr-1), Medium Risk (20–50 t ha-1 yr-1), High Risk (50-100 t ha-1 yr-1), Very High Risk (100 – 500 t ha-1 yr-1), Extreme Risk (Above 500 t ha-1 yr-1). The mean of annual soil loss for the year of 2015 to 2021 is shown in fig.7(a).

In the Study area, the Mean Soil Erosion rate is 155.19 tonnes/ha/yr, 164.82 tonnes/ha/yr, 155.30 tonnes/ha/yr, 131.70 tonnes/ha/yr, 176.90 tonnes/ha/yr, 203.43 tonnes/ha/yr, 209.6 tonnes/ha/yr for 2015, 2016, 2017, 2018, 2019, 2020 and 2021 respectively. The average of seven years erosion rate is 170.99 tonnes/ha/yr. The greatest mean soil erosion of 209.6 tonnes/ha/yr in 2021 was due to high rainfall events occurring in this region. According to reports, solely precipitation erosion causes 56% of soil loss, and rivers account for 29% of losses [28], [29]. In this study, the Result shows that the Extreme Risk erosion category falls into the streamlines of the river at all locations. In this study, it is observed that river bank areas (Fig.7c) have seen severe Average yearly soil loss above 1500 tonnes/ha/yr from 2015 to 2021. Due to stronger water forces, less vegetation, steeper slopes, and sandy loam soils, riverbanks have seen extreme soil loss. Mainly Shetruji River and its tributaries which is perennial river up to some extent of Palitana taluka have severe soil loss as per the observed result of the annual soil erosion map (Fig.7b & 7c). In Fig.6, blue coloured pixels showed very high-risk category which is shown near shetruji river area and region between Bhavnagar to Dholera.

In the Study region, the Soil loss risk category is divided into 5 classes, Low risk, Medium Risk, High risk, Very High risk, Extreme risk, and No Erosion class. As per Table.5 result showed that 45.06% area of Bhavnagar district is under the No soil erosion category which is made up of Urban and built-up lands and soil with strong bonding with its roots with good conservational measures. After that, the average percentage of the high-risk and very high-risk categories is 20.73%, and 19.40% respectively, which indicates almost 40% area is under a severe risk zone, which needs considerable action plans. While, Sum of the mean of the percentage of Low risk and medium-risk categories is 10.889%, which indicates less worried for that area. Table.5 shows an area covered in hectares as per annual soil erosion risk category for the year 2015 to 2021.

Table.5 Area covered in Hectares as per annual soil loss risk category for Bhavnagar District

???????G. Effect of Soil Erosion on Various LULC Classes

In this study, for the semi-arid regions, results are mentioned here for Soil erosion as per Land use and Land cover of the study area. As per the Result, data derived from the MODIS land cover image Bhavnagar district has 8 classes of land cover which are open shrublands, savannas, grasslands, Permanent wetlands, croplands, urban and built-up lands, barren lands, and water bodies, and highest area contributed to Croplands has 478551 hectares, and the Least area contributes to Savannas has 243 hectares. Table.6 shows area covered by different LULC classes of the study area.

The result showed in fig.7d indicates that the Croplands Cover has a major area covered under soil erosion in all classes from 2015 to 2021, While the least area covered under soil erosion occurred in Savannas Cover from all the LULC Cover classes from 2015 to 2021. In fig.5(a) pink coloured area shows barren land, and in fig.6, blue pixel shows very high risk soil erosion occured in bareen land cover class which is placed between Bhavnagar city to Dholare village towards coastal direction.

VI. STATISTICAL ANALYSIS

To comprehend the degree to which each element impacts soil erosion, some common statistical techniques have been applied. Also, an effort has been made to determine if there are some factors that have a more significant impact on erosion than others. Sensitivity analysis was done to examine how the local geo environment's effects compared to the components used to quantify soil loss. For sensitivity analysis, RUSLE five parameters and soil loss data were taken and analysed by Bivariate Relationships and multivariate relationships in an Excel spreadsheet. The result showed that the C factor (cover management factor) is the most influencing for soil erosion, after that LS factor (slope length and steepness factor) is regulate soil erosion up to some extent than R, K, and P factors.

???????A. Bi-Variate Relationships

First, it was determined that the bivariate relationship has been between soil loss and the five parameters of rainfall erosivity (R), soil erodibility (K), slope length-gradient (LS), and crop (C), and Support practice factor (P) with the results shown in Table.7

Table.7 Bi-variate Relationships between Annual Soil loss and RUSLE Parameters

The result depicts in Table 7, shows that among other factors, the C factor plays a dominant role in influencing soil erosion in this study area. It is having greater correlation (R2 = 0.79) with annual soil loss. Fig.8 shows graphical representation of the relationships between the Soil loss and R, K, LS, C and P factor respectively. On other hand, the LS factor (slope length and steepness) is also well correlated (R2 = 0.56) to annual soil loss, but it is less sensitive than the crop management factor. While other factors like R, K, and P have R2 values of 0.09, 0.07, and 0.12 respectively, which are less sensitive to soil erosion.

???????

Conclusion

In this study we found that the lowest annual R factor of 252.40 MJ mm ha-1 h-1 yr-1 was recorded in Gariyadhar Taluka for the year 2018. The highest annual R factor of 388.50 MJ mm ha-1 h-1 yr-1 was recorded in the Palitana and Mahuva talukas for the year 2021. The results showed that the least K factor of 0.13 t ha h ha-1 MJ-1 mm-1 in Talaja Taluka and Palitana taluka, whereas the greatest value of 0.17 t ha h ha-1 MJ-1 mm-1 was seen in coastal regions like, Ghogha taluka, and Mahuva taluka and Bhavnagar taluka. The slope length and steepness factor (LS) was ranging between 0 to 1000 and higher values are observed at the bank of the river stream and hilly terrain. The least value was observed at cropland cover and urban & built-up land cover of the Bhavnagar district. The greatest LS factors were seen on the bank of the Shetruji river in Palitana taluka of Bhavnagar district. The mean cover management factor (C) was 0.82, 0.80, 0.80, 0.85, 0.74, 0.85, and 0.85 for the years 2015, 2016, 2017, 2018, 2018, 2019, 2020, and 2021 respectively. It is a unitless parameter and it depends on the vegetation cover of the land. The support practice factor (P) is also a unitless parameter and ranges between 0.29 to 1 for different Land use and land cover class. The highest values are provided to built-up land and water bodies and the least value is provided whereas some amount of conservation practices is done. We conclude that the mean of annual soil loss ranges between 131.70 to 209.6 tonnes/ha/year in the study region for the years 2015 to 2021. The highest mean value of soil loss was 209.6 tonnes/ha/year was observed in the year 2021, and the lowest value of soil loss 131.70 was observed in the year 2018. The extreme risk for soil loss locations located at the bank of the shetruji river of Palitana Taluka (Latitude 21027’36” N, Longitude 71050’44” E) and very high-risk soil loss occur nearby the above location and another site which is in the surrounding area of Ghogha taluka have latitude 21040’10” N, 72013’01” E. The lower vegetation area of Bhavnagar district observed from the NDVI map is most of the barren land between Bhavnagar and Dholera towards the coastal direction. But due to the clayey soils in this land, high erosion cannot take place in this region. After a detailed analysis of the study region from the result it has been concluded that lack of vegetation cover and conservation practices very high risk and high-risk potential soil loss occurred in the Palitana taluka and Ghogha taluka mostly. As the bare land is vulnerable to soil erosion, proper vegetation cover should be adopted in that area to avoid the risk of erosion. The study is based on remote sensing and geographic information systems (GIS), which may be monitored in the future to assess the important changes over time. Farmers and policymakers will find the research to be very useful in implementing the necessary actions to reduce soil losses.

References

[1] L. M. Kiage, “Perspectives on the assumed causes of land degradation in the rangelands of Sub-Saharan Africa,” Progress in Physical Geography: Earth and Environment, vol. 37, no. 5, pp. 664–684, Oct. 2013, doi: 10.1177/0309133313492543. [2] W. H. Wischmeier and D. D. Smith, Predicting Rainfall Erosion Losses. USDA, 1978. [3] M. Kouli, P. Soupios, and F. Vallianatos, “Soil erosion prediction using the Revised Universal Soil Loss Equation (RUSLE) in a GIS framework, Chania, Northwestern Crete, Greece,” Environ Geol, vol. 57, no. 3, pp. 483–497, Apr. 2009, doi: 10.1007/s00254-008-1318-9. [4] M. Xiong, R. Sun, and L. Chen, “Global analysis of support practices in USLE-based soil erosion modeling,” Progress in Physical Geography: Earth and Environment, vol. 43, no. 3, pp. 391–409, Jun. 2019, doi: 10.1177/0309133319832016. [5] G. Wang, P. Hapuarachchi, H. Ishidaira, A. S. Kiem, and K. Takeuchi, “Estimation of Soil Erosion and Sediment Yield During Individual Rainstorms at Catchment Scale,” Water Resour Manage, vol. 23, no. 8, pp. 1447–1465, Jun. 2009, doi: 10.1007/s11269-008-9335-8. [6] Renard K.G, Foster G.R, Weeisies G A, and Porter J P, “Revised Universal Soil loss equation,” Jornal of soil and water conservation, vol. 46, no. 1, pp. 30–33, 1991. [7] D. Agarwal, K. Tongaria, S. Pathak, A. Ohri, and M. Jha, “SOIL EROSION MAPPING OF WATERSHED IN MIRZAPUR DISTRICT USING RUSLE MODEL IN GIS ENVIRONMENT,” ijsrtm, vol. 4, no. 3, pp. 56–63, Dec. 2016, doi: 10.18510/ijsrtm.2016.433. [8] V. J. Markose and K. S. Jayappa, “Soil loss estimation and prioritization of sub-watersheds of Kali River basin, Karnataka, India, using RUSLE and GIS,” Environ Monit Assess, vol. 188, no. 4, p. 225, Apr. 2016, doi: 10.1007/s10661-016-5218-2. [9] J. Pan and Y. Wen, “Estimation of soil erosion using RUSLE in Caijiamiao watershed, China,” Nat Hazards, vol. 71, no. 3, pp. 2187–2205, Apr. 2014, doi: 10.1007/s11069-013-1006-2. [10] B. Pradhan, A. Chaudhari, J. Adinarayana, and M. F. Buchroithner, “Soil erosion assessment and its correlation with landslide events using remote sensing data and GIS: a case study at Penang Island, Malaysia,” Environ Monit Assess, vol. 184, no. 2, pp. 715–727, Feb. 2012, doi: 10.1007/s10661-011-1996-8. [11] V. Prasannakumar, R. Shiny, N. Geetha, and H. Vijith, “Spatial prediction of soil erosion risk by remote sensing, GIS and RUSLE approach: a case study of Siruvani river watershed in Attapady valley, Kerala, India,” Environ Earth Sci, vol. 64, no. 4, pp. 965–972, Oct. 2011, doi: 10.1007/s12665-011-0913-3. [12] M. K. Jat, D. Khare, P. K. Garg, and V. Shankar, “Remote sensing and GIS-based assessment of urbanisation and degradation of watershed health,” Urban Water Journal, vol. 6, no. 3, pp. 251–263, Sep. 2009, doi: 10.1080/15730620801971920. [13] P. Thapa, “Spatial estimation of soil erosion using RUSLE modeling: a case study of Dolakha district, Nepal,” Environ Syst Res, vol. 9, no. 1, p. 15, Dec. 2020, doi: 10.1186/s40068-020-00177-2. [14] S. Chatterjee, A. P. Krishna, and A. P. Sharma, “Geospatial assessment of soil erosion vulnerability at watershed level in some sections of the Upper Subarnarekha river basin, Jharkhand, India,” Environ Earth Sci. [15] L. Jiang, Z. Yao, Z. Liu, S. Wu, R. Wang, and L. Wang, “Estimation of soil erosion in some sections of Lower Jinsha River based on RUSLE,” Nat Hazards, vol. 76, no. 3, pp. 1831–1847, Apr. 2015, doi: 10.1007/s11069-014-1569-6. [16] S. R. Kashiwar, M. C. Kundu, and U. R. Dongarwar, “Soil erosion estimation of Bhandara region of Maharashtra, India, by integrated use of RUSLE, remote sensing, and GIS,” Nat Hazards, vol. 110, no. 2, pp. 937–959, Jan. 2022, doi: 10.1007/s11069-021-04974-5. [17] M. Nakil and M. Khire, “Effect of slope steepness parameter computations on soil loss estimation: review of methods using GIS,” Geocarto International, vol. 31, no. 10, pp. 1078–1093, Nov. 2016, doi: 10.1080/10106049.2015.1120349. [18] D. S. Pai, M. Rajeevan, O. P. Sreejith, B. Mukhopadhyay, and N. S. Satbha, “Development of a new high spatial resolution (0.25° × 0.25°) long period (1901-2010) daily gridded rainfall data set over India and its comparison with existing data sets over the region,” MAUSAM, vol. 65, no. 1, pp. 1–18, Jan. 2014, doi: 10.54302/mausam.v65i1.851. [19] S. Gupta and S. Kumar, “Simulating climate change impact on soil erosion using RUSLE model ? A case study in a watershed of mid-Himalayan landscape,” J Earth Syst Sci, vol. 126, no. 3, p. 43, Apr. 2017, doi: 10.1007/s12040-017-0823-1. [20] K. K., B. F., and O. O., “ASSESSMENT OF SOIL EROSION BY RUSLE MODEL USING GIS: A CASE STUDY OF CHEMORAH BASIN, ALGERIA,” Malays. j. geosci., vol. 4, no. 2, pp. 70–78, May 2020, doi: 10.26480/mjg.02.2020.70.78. [21] N. K. Trambadia, D. P. Patel, V. M. Patel, and M. J. Gundalia, “Comparison of two open-source digital elevation models for 1D hydrodynamic flow analysis: a case of Ozat River basin, Gujarat, India,” Modeling Earth Systems and Environment, vol. 8, no. 4, pp. 5433–5447, Nov. 2022, doi: 10.1007/s40808-022-01426-2. [22] H. Mitasova, J. Hofierka, M. Zlocha, and L. R. Iverson, “Modelling topographic potential for erosion and deposition using GIS,” International journal of geographical information systems, vol. 10, no. 5, pp. 629–641, Jul. 1996, doi: 10.1080/02693799608902101. [23] P. K. Shit, A. S. Nandi, and G. S. Bhunia, “Soil erosion risk mapping using RUSLE model on jhargram sub-division at West Bengal in India,” Model. Earth Syst. Environ., vol. 1, no. 3, p. 28, Oct. 2015, doi: 10.1007/s40808-015-0032-3. [24] X. Yue-Qing, S. Xiao-Mei, K. Xiang-Bin, P. Jian, and C. Yun-Long, “Adapting the RUSLE and GIS to model soil erosion risk in a mountains karst watershed, Guizhou Province, China,” Environ Monit Assess, vol. 141, no. 1–3, pp. 275–286, Jun. 2008, doi: 10.1007/s10661-007-9894-9. [25] I. Gaubi, A. Chaabani, A. Ben Mammou, and M. H. Hamza, “A GIS-based soil erosion prediction using the Revised Universal Soil Loss Equation (RUSLE) (Lebna watershed, Cap Bon, Tunisia),” Nat Hazards, vol. 86, no. 1, pp. 219–239, Mar. 2017, doi: 10.1007/s11069-016-2684-3. [26] A. Pandey, V. M. Chowdary, and B. C. Mal, “Identification of critical erosion prone areas in the small agricultural watershed using USLE, GIS and remote sensing,” Water Resour Manage, vol. 21, no. 4, pp. 729–746, Feb. 2007, doi: 10.1007/s11269-006-9061-z. [27] V. Ferro, G. Giordano, and M. Iovino, “Isoerosivity and erosion risk map for Sicily,” Hydrological Sciences Journal, vol. 36, no. 6, pp. 549–564, Dec. 1991, doi: 10.1080/02626669109492543. [28] S. S. Biswas and P. Pani, “Estimation of soil erosion using RUSLE and GIS techniques: a case study of Barakar River basin, Jharkhand, India,” Model. Earth Syst. Environ., vol. 1, no. 4, p. 42, Dec. 2015, doi: 10.1007/s40808-015-0040-3. [29] D. V. V. Narayana and R. Babu, “Estimation of Soil Erosion in India,” J. Irrig. Drain Eng., vol. 109, no. 4, pp. 419–434, Dec. 1983, doi: 10.1061/(ASCE)0733-9437(1983)109:4(419).

Copyright

Copyright © 2023 Akshaykumar R Desai, V. M. Patel, N. K. Trambadia. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.