Loading...
Thesis
Browse
15 results
Search Results
ThesisItem Open Access STUDY OF COMPARATIVE PERFORMANCE OF WEPP AND USLE MODEL FOR PREDICTION OF SOIL LOSS FROM KARLI RIVER CATCHMENT(COLLEGE OF AGRICULTURAL ENGINEERING AND TECHNOLOGY DR. BALASAHEB SAWANT KONKAN KRISHI VIDYAPEETH,DAPOLI, 2018-09-26) Mr. Bandgar Nitin Namdev; Dr. B.L.Ayare; Prof. dilip MAHALE, Dr. S. B. NandgudeABSTRACT "STUDY OF COMPARATIVE PERFORMANCE OF WEPP AND USLE MODEL FOR PREDICTION OF SOIL LOSS FROM KARLI RIVER CATCHMENT" By Nitin Namdev Bandgar Department of Soil and Water Conservation Engineering, College of Agriculture Engineering and Technology, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli Dist- Ratnagiri, Maharashtra 2018 Research Guide : Dr. B. L. Ayare Department : Soil and Water Conservation Engineering Soil erosion is a serious problem that seems from a combination of agricultural intensification, soil degradation and intense rainstorms. Erosion may also be exacerbated in the future in many parts of the world because of erotic climatic change results into more vigorous changes in hydrologic cycle. The different management theories, formulae, equations and models have been developed to predict the soil loss from the catchment. In recent decades, models have been built (empirical, conceptual, or physically based) in order to represent and to quantify the processes of detachment, transport, and deposition of eroded soil, with the aim of implementing assessment tools for educational, planning and legislative purposes . Among the different models being used to predict the soil loss along with other important parameters the Water Erosion Prediction Project (WEPP) and Universal Soil Loss Equation (USLE) model are being widely used for the purpose therefore the study, “Study of comparative performance of WEPP and USLE models for prediction of soil loss from Karli river catchment” was under taken to predict the soil loss. The present research work was conducted at Karli river catchment is located between the longitude 73.92 0 to 74.01 0 E and latitudes 16.04 0 to 16.10 0 N on the xviwestern coast of India in the southern part of Maharashtra state. The total geographical area of study location is 4247.12 ha. Data collected through Remote Sensing (RS) and Geographical Information System (GIS) was used in the study of land-use pattern and analysis relating the soil loss with loss of yield. With the help of RS and GIS data, Comparative performance of Water Erosion Prediction Project (WEPP) and Universal Soil Loss Equation (USLE) model were used for prediction of soil loss from Karli river catchment. The WEPP model computed soil loss for 7 channels and 18 hill slopes of Karli river catchment. The GeoWEPP model run for Karli river catchment with contributing total area to outlet was 3978.75 ha. The average annual soil loss from hill slopes and channels was found to be 42.89 t/ha/yr and 8.78 t/ha/yr respectively, totally to 51.67 t/ha/yr. The WEPP model also calculated the sediment yield of Karli river catchment that is 17.92 t/ha/yr. The Water Erosion Prediction Project (WEPP) model predicted the 9.01 t/ha/yr more soil loss than the Universal Soil Loss Equation (USLE) model. It also overestimates the sediment yield than the government data by 9.80 t/ha/yr (Ahmadi et al., 2011) The Universal Soil Loss Equation (USLE) was used for estimation of soil loss from the watershed. The different parameters including soil loss and related were determined by using Remote Sensing data and Geographical Information System tools. The predicted soil loss by using USLE in the Karli river catchment was found to be 42.66 t/ha/yr and it is 9.01 t/ha/yr less than predicted by WEPP. The R factor values were calculated using relationship between the daily rainfall and erosivity index of Wakawali region by developing regression equation. The average annual erosivity obtained for Dukanwadi station was 6635.65 MJ- mm/ha-hr-yr. Soil erodibility factor values were estimated using sand (%), silt (%), clay (%), organic matter content (%), structural code and permeability code of each village. Weighted soil erodibility factor for Karli river catchment was ranging between 0.040 to 0.041 t-ha-hr/ha-MJ-mm. The value of LS factor for Karli river catchment was found in the range of 1.81 to 4.53. The crop management factor associated with erosion losses is site specific. Detailed information on land use land cover was obtained by LANDSAT imageries and field survey. Crop management factor (C) values of Karli river catchment were ranging from 0.024 to 0.12. Considering support conservation practice factor value as 1, soil loss was estimated for Karli river catchment and its micro watersheds using USLE. xviiAlong with the prediction of the soil loss by using WEPP model and its comparison with USLE model the sediment yield predicted by WEPP and observed data by Government department was also considered and for the purpose, The sediment data of Dukanwadi station from 2001-2010 was collected from Hydrology Project, Water Resource Department, Government of Maharashtra (India). According to the observed data, the average annual sediment yield was 8.12 t/ha/yr whereas the predicted sediment yield from the WEPP model was 17.92 t/ha/yr. which is 120 % more predicted than the observed data. The sedimentation also 15.71 % and 34.68% of predicted soil loss by WEPP and for the observed sediment yield and predicted sediment yield respectively. The similar statement is made and that is the sediment yield is 30 to 60% of soil erosion loss (Fernandez et al., 2003; Vemu et al., 2012; Richarde et al., 2014). According to comparative performance of Water Erosion Prediction Project (WEPP) model overestimates the soil loss value and sediment yield value than the Universal Soil Loss Equation (USLE) model and Government data respectively. WEPP model was best suitable model for Karli river catchment due to its less input files, less time consumption, ease to operate and understand, and less data requirement with minimum pre-processed data. xviiiThesisItem Open Access FORECASTING OF WATER TABLE FLUCTUATIONS FOR PRIYADARSHINI WATERSHED USING ARTIFICIAL NEURAL NETWORK(COLLEGE OF AGRICULTURAL ENGINEERING AND TECHNOLOGY DR. BALASAHEB SAWANT KONKAN KRISHI VIDYAPEETH,DAPOLI, 2019-09-06) Mr. Prakash Basavanni Hittanagi; Dr. H. N. Bhange; Dr. M. S. Mane, Dr. B. L. AyareABSTRACT “FORECASTING OF WATER TABLE FLUCTUATIONS FOR PRIYADARSHINI WATERSHED USING ARTIFICIAL NEURAL NETWORK ” By Prakash Basavanni Hittanagi Department of Soil and Water Conservation Engineering, College of Agriculture Engineering and Technology, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli Dist- Ratnagiri, Maharashtra 2019 Research Guide : Dr. H. N. Bhange Department : Soil and Water Conservation Engineering Groundwater is an important natural resource essential for sustenance of life. Over 98% of the freshwater on the Earth lies below its surface. It is located below the soil surface and largely contained in interstices of bedrocks, sands, gravels, and other interspaces through which precipitation infiltrates and percolates into the underground aquifers due to gravity. The total amount of water in the world is 1.4 billion km 3 . 97.5% of these waters are in the oceans and the seas and 2.5% is in fresh water. Sweet waters; 0.3% is in lakes and rivers, 30.8% in ground water, soil necropsy and marsh, 68.9% in the form of ice and permanent snow. Groundwater is one of the major sources of supply for domestic, industrial and agricultural purposes. The weekly Rainfall data, Temperature data, Solar data, Water level data and Permeability data of 9 years were used. Artificial Neural Network is an information processing paradigm that is inspired by the way biological nervous systems, such as the brain, The network is composed of a large number of highly interconnected processing elements called as neuron. They typically consist of hundreds of simple processing units which are wired together in a complex communication network. Each unit or node is a simplified model of real neuron which sends off a new signal or fires if it receives a sufficiently strong input signal from the other nodes to which it is connected, learning in this system involves the adjustment between neurons through synaptic connection. In this study feed-forward neural networks architecture has been used in predicting weekly water table depths. In this study, sensitivity analysis has 3been done to measure relative importance of each input variable for precisely predicting groundwater table fluctuations. Sensitivity analysis is done by removing one input parameter at a time from the model and testing its performance by comparing with original model. Considering training, validation and testing period and all the statistics, it is difficult to say which algorithm is better among the two selected for study. Because there was a lot of variation in all the statistics among the two selected algorithms for training, validation and testing period. But considering the testing period of all the nine wells it was found that LM algorithm was better than CG for wells i.e., well 1 (2-9-1), well 2 (2-9-1), well 3 (1-8-1), well 4 (1-6-1), well 5 (2-9-1), well 6 ( 1-9-1), well 8( 2-9-1) while CG algorithm was better than LM for wells i.e., well 7 (2-5-1) and well 9 (3-5-1) So these algorithms for particular well were selected for sensitivity analysis. As the results found were based on trial and error methods Levenberg- Marquardt (LM) algorithm provides better results than Conjugate Gradient algorithm. Levenberg- Marquardt (LM) best results for ANN network architecture of model for well 1 (2-9-1), well 2 (2-9-1), well 3(1-8-1), well 4(1-6-1), well 5(2-9-1), well 6 (1-9-1), well 7 (3-5-1), well 8 (2-9-1), well 9 (2-9-1).The predicted water level trend followed the observed trend closely, showing the accuracy of the network. In present study, results were found and based on sensitivity analysis models selected and their statistics for all the nine wells. It was observed that selected algorithms predicted the water table depths in a better way in terms of its performance statistics. The values of R for LM and CG were found to be 0.836 and 0.743, respectively. The observed values of RMSE for LM and CG were found to be 0.100 and 0.101, respectively. Similarly, the value of E for LM and CG were found to be -376.40 and -20.88, respectively. The above result concludes Levenberg-Marquardt predicts the water table depth better than Conjugate Gradient. (Keywords : ANN, Sensitivity analysis, Levenberg-Marquardt, Conjugate Gradient)ThesisItem Open Access ESTIMATION OF CARBON STOCK IN AGRICULTURAL AND FOREST LAND IN SOUTHERN PART OF RATNAGIRI DISTRICT(COLLEGE OF AGRICULTURAL ENGINEERING AND TECHNOLOGY DR. BALASAHEB SAWANT KONKAN KRISHI VIDYAPEETH,DAPOLI, 2019-09-23) Mr. Paulzagade Pavan Mahadeo; Dr. H. N. Bhange; Dr. S. B. Nandgude, Dr. B. L. Ayare; Dr. P. M. IngleABSTRACT “ESTIMATION OF CARBON STOCK IN AGRICULTURAL AND FOREST LAND IN SOUTHERN PART OF RATNAGIRI DISTRICT ” By Paulzagade Pavan Mahadeo Department of Soil and Water Conservation Engineering, College of Agriculture Engineering and Technology, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli Dist- Ratnagiri, Maharashtra 2019 Research Guide : Dr. H. N. Bhange Department : Soil and Water Conservation Engineering One of the environmental problem arising to human beings is increasing amount of Green House Gases (GHG) in atmosphere specially CO 2 . The human activity such as burning of fossil fuel for transportation, heating, cooking, electricity generation and manufacturing have increasing concentration of CO 2 in atmosphere that lead to global warming and climate change due to which now a days earth is getting warmer and warmer. It has been found that earth average temperature has risen by 1.4 0 F (0.72 0 C) over past century and it is projected that it will further rise up to 2 to 11.4 0 F over next century. Therefore it is required to understand how much carbon can be stored in ecosystem. Forest and soil are two major carbon pools who store maximum amount of carbon therefore study entitled “Estimation of carbon stock in agriculture and forest land in Southern part of Ratnagiri district”, has been undertaken. The study area selected was Southern part of Ratnagiri district of Konkan region in Maharashtra. Southern part of Ratnagiri district of Maharashtra state, situated in the western coast of India, which is located between 16 0 31' and 17 0 12' N latitude and 73 0 25' and 73 0 33' E longitude. The total geographical area of Southern part of Ratnagiri district is 426369 ha, it is divided into four tehsils namely Sangameshwar, Ratnagiri, Lanja and Rajapur. The regression equation was used to estimate carbon stock invegetation land. The carbon stock in soil was calculated by using wet oxidation method of Walkley and Black. Total forest carbon stock in Southern part of Ratnagiri district was 22.94 M tonnes with an average carbon stock rate of 220.0 t C/ha. The highest rate of forest carbon stock was found in the Sangmeshwar tehsil because Sangmeshwar tehsil has large forest area. The carbon stock in paddy crop was ranges from 0.6 to 5.55 tonnes of C/ha with an average rate of 1.84 t C/ha. The total carbon stock value of paddy crop was 63479.80 tonnes. The lowest average rate of carbon stock in paddy was found in Lanja tehsil because of low cultivation area of paddy in Lanja tehsil. The total carbon stock in horticultural crop in Southern part of Ratnagiri district was found 8.03 M tonnes of carbon with an average carbon stock rate of 74.95 t C/ha. Carbon stock rate of arecanut in Southern part of Ratnagiri district was ranging from 9.68 to 26.83 t C/ha with an average carbon stock rate of 17.06 t C/ha. For mango crops, carbon stock rate was ranging between 6.76 to 398.35t C/ha with an average carbon stock rate of 142.66 t C/ha. For coconut crops the carbon stock rate was ranging between 28.14 to 117.16 t C/ha with an average carbon stock rate of 84.87 t C/ha and for cashewnut crops carbon stock rate was ranging between 27.35 to 74.01 t C/ha with an average carbon stock rate of 54.98 t C/ha in Southern part of Ratnagiri district. Total soil organic carbon stock in Southern part of Ratnagiri district was 6.13 M tonnes at 0 to 15 cm depth and 5.53 M tonnes at 16 to 30 cm depth. The average carbon stock rate was 14.28 t/ha and 12.98 t/ha at 0 to 15 cm and 16 to 30 cm depths, respectively, and the total carbon stock at total depth of soil from 0 to 30 cm was 11.67 M tonnes with an average carbon stock rate of 27.26 t C/ha. Total carbon stock value of Southern part of Ratnagiri district was 42.71 M tonnes of carbon which include that soil carbon stock was 11.67 M tonnes of carbon and vegetation carbon stock was 31.04 M tonnes respectively. It was found that vegetation has higher carbon storage capacity than soil. Value of CO 2 sequestered from vegetation was 113.83 M tonnes of CO 2 . Amount of CO 2 sequestered by soil was 43.14 M tonnes of CO 2 . Total amount of CO 2 sequestered from Southern part of Ratnagiri district was 156.97 M tonnes of CO 2 . Keywords: Carbon stock, Carbon sequestration, Soil organic carbon.ThesisItem Open Access STUDY OF COMPARATIVE PERFORMANCE OF MMF AND RUSLE MODEL FOR PREDICTION OF SOIL LOSS FROM KARLI RIVER CATCHMENT(COLLEGE OF AGRICULTURAL ENGINEERING AND TECHNOLOGY DR. BALASAHEB SAWANT KONKAN KRISHI VIDYAPEETH,DAPOLI, 2019-09-09) Miss. Khot Sonal Bharat; Dr. B.L.Ayare; Prof. dilip MAHALE, Dr. S. B. Nandgude; Dr. M. S. ManeABSTRACT "STUDY OF COMPARATIVE PERFORMANCE OF MMF AND RUSLE MODEL FOR PREDICTION OF SOIL LOSS FROM KARLI RIVER CATCHMENT" By Miss. Sonal Bharat Khot Department of Soil and Water Conservation Engineering, College of Agriculture Engineering and Technology, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli Dist- Ratnagiri, Maharashtra 2019 Research Guide : Dr. B. L. Ayare Department : Soil and Water Conservation Engineering Soil erosion is a serious problem arising from agricultural intensification, land degradation and other anthropogenic activities. A large area suffers from soil erosion, which in turn, reduces productivity. For protection of land and to meet the increasing demand of food, it is necessary to understand soil formation and erosion process. The different management theories, formulae, equations and models have been developed to predict the soil loss from the catchment. Soil erosion models are classified into three groups viz. Empirical, Conceptual (partly empirical/mixed) and Physically-based. Among the different models being used to predict the soil loss along with other important parameters the Morgan-Morgan-Finney (MMF) and Revised Universal Soil Loss Equation (RUSLE) model are being widely used for the purpose, therefore the study, “Study of comparative performance of MMF and RUSLE models for prediction of soil loss from Karli river catchment” was under taken to predict the soil loss. The present research work was conducted at Karli river catchment is located between the longitude 73.92 0 to 74.01 0 E and latitudes 16.04 0 to 16.10 0 N on the western coast of India in the southern part of Maharashtra State. The total length of the Karli River is 56 km. The total geographical area of study location is 4247.12 ha. Data collected through Remote Sensing (RS) and Geographical Information System (GIS) was used in the study of land-use pattern and analysis relating the soil loss with loss of yield. With the help of RS and GIS data, comparative performance of Morgan-Morgan-Finney (MMF) and Revised Universal Soil Loss Equation (RUSLE) model were used for prediction of soil loss from Karli river catchment.The average annual soil loss (t/ha/yr) was estimated for Karli river catchment by MMF model. Soil, climate, crop management and slope were used as input in the MMF model, which gave average annual soil loss of the study area. MMF model involve input maps of kinetic energy of rainfall, mean rain per rainy day, crop cover management factor, soil moisture storage capacity. These were used to generate output maps like volume of overland flow, rate of soil detachment by raindrop impact and transport capacity of overland flow. Annual soil loss estimated by comparing two maps of soil detachment rate and transport capacity and then taking the minimum value from them. The estimated soil loss by using Morgan- Morgan and Finney (MMF) model was 38.36 t/ha/yr. The Revised Universal Soil Loss Equation (RUSLE) was used for estimation of soil loss from the watershed. The different parameters including soil loss and related were determined by using Remote Sensing data and Geographical Information System tools. The estimated soil loss by using RUSLE in the Karli river catchment was found to be 44.38 t/ha/yr. Average annual soil losses were estimated with the help of average annual R factor obtained from 24 years rainfall data, K, LS, C and P. The average annual erosivity for Dukanwadi station was 13888.52 MJ- mm/ha-hr-yr. Soil erodibility factor values were estimated using sand (%), silt (%), clay (%), organic matter content (%), structural code and permeability code of each village. Weighted soil erodibility factor for Karli river catchment was ranging between 0.040 to 0.041 t-ha-hr/ha-MJ-mm. The value of LS factor for Karli river catchment was found in the range of 0 to 5.54. Crop management factor (C) values of Karli river catchment were ranging from 0.968 to 1.271. Considering support conservation practice factor value as 1, soil loss was estimated for Karli river catchment and its micro watersheds using RUSLE. Using MMF model, about 70 % area and using RUSLE model, about 80 % area comes under moderately severe to extremely severe class for Karli river catchment. This proves the high need of soil and water conservation measures in the watershed for the sustainable management of natural resources. The Revised Universal Soil Loss Equation (RUSLE) model required less input data, ease to understand and operate, which took less time to run than the Morgan- Morgan and Finney (MMF) model. According to comparative performance, Revised Universal Soil Loss Equation (RUSLE) model estimated 6.02 t/ha/yr more soil loss than Morgan–Morgan–Finney (MMF) model. RUSLE model required less input data, less time consumption, less data requirement, easy to operate and understand.ThesisItem Open Access SITE SELECTION FOR WATER HARVESTING STRUCTURES IN TETAVALI WATERSHED USING REMOTE SENSING AND GIS(COLLEGE OF AGRICULTURAL ENGINEERING AND TECHNOLOGY DR. BALASAHEB SAWANT KONKAN KRISHI VIDYAPEETH,DAPOLI, 2021-09-30) MISS. SURVE RASIKA RAGHUNATH; Dr. H. N. Bhange; Dr. B. L. Ayare, Dr. P. M. IngleABSTRACT “SITE SELECTION FOR WATER HARVESTING STRUCTURES IN TETAVALI WATERSHED USING REMOTE SENSING AND GIS ” By Rasika Raghunath Surve Department of Soil and Water Conservation Engineering, College of Agriculture Engineering and Technology, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli Dist- Ratnagiri, Maharashtra 2021 Research Guide Department : Dr. H. N. Bhange : Soil and Water Conservation Engineering Water plays very important role not only in fulfilling basic needs but also in socio economic development. Increasing needs of high population and increasing water demand in all sectors creates a water scarcity problem. The basic unit of water resources assessment and planning is watershed. The present study is aimed to assess the potential for water harvesting and to identify suitable sites for water harvesting of Tetavali watershed, which is one of the research block of central experimental station, wakavali, Dr. BSKKV, Dapoli. The Tetavali watershed is located in Dapoli tehsil of Ratnagiri district from Maharashtra is selected as study area, which is situated between 16 0 70 ̓ 33” N. latitude and 73 0 52 ̓ 70”E.longitude. Average rainfall of Dapoli is 3500 mm. Area occupied by watershed is 382 ha. SRTM DEM is downloaded and used for watershed delineation. LANDSAT-8 image was used for preparation of land use land cover map of watershed in Arc GIS 10.2.Data was procured from Bhukosh –geological survey of India portal was used to understand and study geomorphology and lithology of study area. Geomorphological parameters were determined to understand watershed characteristics. Drainage density map was created using DEM. The area under the study was very poor, poor, moderate, good, very good drainage density are 298.44 ha (78.13 %), 10.73 ha (2.93%) 25.9 ha (6.80%), 35.65 ha (9.33%) and 11.19 ha (2.93%) respectively. Drainage density of study area represents permeable subsoil and dense vegetation is present in Tetavali watershed. The land use land cover of study area is classified as forest, agriculture, barren land and build up area covering 192.03 ha (50.27%), 99.70 ha (26.4%),37.73 ha (9.67%) and 52.54 ha (13.66 %) respectively showing most of the land covered with forest and agriculture. Geomorphology of Tetavali watershed is covered by three classes moderately dissected plateau, pediment pediplain complex and waterbody and covers 122.98ha (32.19%), 256.65ha (67.19%) and 2.37 ha (0.62%) of total area of watershed. Lithology covers total area by laterite 44.15 ha (11.56%) and basalt 337.85ha (88.44%). Study used drainage density map, stream order map, slope map, soil texture map, land use land cover, geomorphology map and lithology map for selection of suitable water harvesting sites. Weightages were assigned with rank 1 to 5 and all maps were integrated in GIS. Integrated Mission for Sustainable Development (IMSD) guidelines were used for selecting water harvesting structures like check dam, farm pond, percolation ponds, cement nala bund and earthen nala bund. Study found six farm pond sites, three check dam sites, two percolation pond sites and one each cement nala bund and earthen nala bund site. 278.36 ha (72.86%) area of total area found suitable for water harvesting structure construction and 103.64 ha (27.14%) of watershed not suitable for water harvesting structures. Keywords: RS and GIS, DEM, Water harvesting structures.ThesisItem Open Access DELINEATION OF GROUNDWATER POTENTIAL ZONES AND SITE SELECTION FOR ARTIFICIAL RECHARGE USING GEO SPATIAL TECHNOLOGIES(COLLEGE OF AGRICULTURAL ENGINEERING AND TECHNOLOGY DR. BALASAHEB SAWANT KONKAN KRISHI VIDYAPEETH,DAPOLI, 2021-11) Mr. PATIL DEEPAK RAVINDRA; Dr. H. N. Bhange.; Dr. B. L. Ayare, Dr. P.M.Ingale; Prof.M.H.TharkarABSTRACT“DELINEATION OF GROUNDWATER POTENTIAL ZONES AND SITE SELECTION FOR ARTIFICIAL RECHARGE USING GEO SPATIAL TECHNOLOGIES” By Patil Deepak Ravindra Department of Soil and Water Conservation Engineering, College of Agriculture Engineering and Technology, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli Dist- Ratnagiri, Maharashtra 2021 Research Guide Department : Dr. H. N. Bhange : Soil and Water Conservation Engineering Groundwater is emerging as the critical issue all over the world. Due to overuse of groundwater, water level from deep aquifers that have taken decades to accumulate have been diminishing. The study of groundwater has remained difficult, as there is no direct method to facilitate observation of water below the surface. Its presence or absence can only be inferred indirectly by studying the geological and surface parameters. Various technique can be used to provide information about potential occurrence of groundwater. Remote sensing has been practiced for about three decades in groundwater exploration. An integrated approach of remote sensing (RS) and GIS technique is very useful in demarcation of various groundwater potential zones. Remote Sensing and GIS are one of the advanced technologies which understands sub-surface water condition with their advantages of spectral, spatial, temporal nature of data and analyse of data Asond is small watershed situated in Dapoli tehsil of Ratnagiri district, Maharashtra. The total area of Asond watershed is 510.76 ha. The average annual rainfall of nearby rain gauge station Wakavali is 2860 mm. Even though in every summer, it faced severe water scarcity due to improper water management. Therefore, there wasneed to delineate groundwater potential zones and suitable sites for artificial recharge to increase the groundwater level. Five thematic maps such as lithology, geomorphology, drainage density, slope and land use/land cover were generated to demarcate different groundwater potential zones of the Asond watershed. Each thematic map plays a significant role in delineation of groundwater potential zone. All the thematic maps were integrated using weighted overlay method in spatial analyst tool. In weighted overlay analysis, the ranking has been given to each individual parameter of each thematic map and weights were assigned according to the influence characteristics such as geomorphology – 30%, Lithology – 20%, slope – 15%, Land use/land cover-15% and drainage density – 20% to find out the potential zones of study area . The groundwater potential zone map was delineated and classified into three zones, ‘Poor’, ‘Moderate’ and ‘Good’. About 15.18% of the study area falls under poor potential zone category; 62.80% area falls under moderate category and 22.01% area falls under good potential category. For the sustainable groundwater resource management, artificial recharge structures play a vital role in hard rock regions. By considering this aspect, zones favourable for artificial recharge of groundwater were to be demarcated. Ranking was assigned to each class of thematic map and all the thematic maps were integrated using weighted overlay method. Based on the artificial recharge potential zone map, the favourable artificial recharge sites were selected in the watershed. Three types of recharge structures, namely percolation tanks, check dams and recharge ponds were suggested to augment the groundwater. Keywords: RS and GIS, Thematic maps, Groundwater potential zone, Artificial recharge.ThesisItem Open Access LAND COVER MAPPING WITH CHANGE DETECTION OF SOUTH KONKAN REGION FOR SOIL EROSION RISK ASSESSMENT USING REMOTE SENSING AND GIS(COLLEGE OF AGRICULTURAL ENGINEERING AND TECHNOLOGY DR. BALASAHEB SAWANT KONKAN KRISHI VIDYAPEETH,DAPOLI, 2021-09-27) Mr. Suraj Dattatraya Dhaigude; Dr. H. N. Bhange; Dr. B. L. Ayare, Dr.S.T.Patil; Dr.P.B.BansodeABSTRACT “ LAND COVER MAPPING WITH CHANGE DETECTION OF SOUTH KONKAN REGION FOR SOIL EROSION RISK ASSESSMENT USING RS AND GIS” By Suraj Dattatraya Dhaigude. Department of Soil and Water Conservation Engineering, College of Agricultural Engineering and Technology, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli Dist- Ratnagiri, Maharashtra 2021 Research Guide : Dr. H. N. Bhange Department : Soil and Water Conservation Engineering The land and water are the most important natural resources which are essential for existence of life on the Earth. The soil and water has natural and man caused problems including soil degradation, soil erosion, water pollution, water logging, water salinity etc. Among the above mentioned problems, the soil erosion is the major problem in developing countries. The changes in the land use and land cover (LULC) significantly affect the volume and extent of the soil erosion. Therefore, the attempts were made for land cover mapping with change detection of South Konkan xvi iiregion for soil erosion risk assessment for the years 2013 and 2018. The Landsat 8 satellite images with 30 meter resolution, Digital Elevation Model (DEM), daily rainfall data and soil data was used along with Arc GIS 10.3 software, available with Dr. BSKKV, Dapoli, to estimate the soil loss and assess the erosion risk of Ratnagiri and Sidhudurg district. The Universal Soil Loss Equation (USLE) was used to calculate the annual soil loss from the study area. It was found that, the area under forest and barren land in Ratnagiri district was decreased by 0.22 % and 2.43 % from 2013 to 2018, respectively. At the same time area under agriculture and residence or built-up was increased by 0.81 % and 1.86 %, respectively. The overall accuracy of land use mapping of Ratnagiri district for the years 2013 and 2018 was found to be 90.84 % and 90 %, respectively. The agricultural area and residential area in Sindhudurg district was found to be increased in the year 2018 by 3.44 % and 2.65 %, respectively. However, the area under forest land and barren land was decreased by 4.08 % and 2.33 %, respectively. The overall accuracy of the land mapping was found to be 88.89 % and 88.65 % for the years 2013 and 2018, respectively. The area under slight, moderate, moderately severe, severe, highly severe and extremely severe for Ratnagiri district in the year 2013 was found to be 9.99 %, 12.81 %, xix23.84 %, 24.30 %, 17.91 % and 11.15 %, respectively. In the year 2018, the area under slight, moderate and moderately severe soil erosion was increased to 14.95 %, 15.33 % and 25.92 %, respectively. At the same time, the area under severe, highly severe and extremely severe soil erosion risk was decreased to 23.47 %, 13.94 % and 6.39 %, respectively. The area under slight, moderate and moderately severe soil erosion risk of Sindhudurg district was increased from 11.99 %, 10.75 % and 18.49 % to 16.02 %, 13.07 % and 21.73 % over the study period, respectively. At the same time the area under very severe and extremely severe soil erosion risk was decreased from 21.89 % and 10.40 % to 16.96 % and 5.08 %, respectively. Keywords: RS and GIS, DEM, Landsat, Soil erosion risk, LULC. xxThesisItem Open Access PLANNING AND DESIGNING OF SOIL AND WATER CONSERVATION STRUCTURES IN DAPOLI TEHSIL USING REMOTE SENSING AND GIS(COLLEGE OF AGRICULTURAL ENGINEERING AND TECHNOLOGY DR. BALASAHEB SAWANT KONKAN KRISHI VIDYAPEETH,DAPOLI, 2021-09-09) Miss DARA ROOHA BLESSY; Dr. B.L.Ayare; Dr.H.N.Bhange, Dr.S.T.Patil; Dr.P.R.KolheABSTRACT "PLANNING AND DESIGNING OF SOIL AND WATER CONSERVATION STRUCTURES IN DAPOLI TEHSIL USING REMOTE SENSING AND GIS" by Miss Dara Rooha Blessy Department of Soil and Water Conservation Engineering, College of Agricultural Engineering and Technology, Dr. Balasaheb Sawant Konkan Krishi Vidyapeeth, Dapoli Dist.- Ratnagiri, Maharashtra 2021 Research Guide : Dr.B.L.Ayare Department : Soil and Water Conservation Engineering Soil and water are two important natural resources and basic need for agricultural production. Though the nature has been gifting us with modest amount of soil and rainfall, the unscientific management of these resources leads to scarcity in several parts of the country. Therefore, efficient management of soil and water has to be performed by planning and designing suitable structures at appropriate locations. In recent times, Remote sensing and GIS have gained higher attention in water resource planning and management. Thus, a study was conducted in Dapoli tehsil of Ratnagiri district in Maharashtra, India to calculate the geomorphological characteristics and to plan, locate and design soil and water conservation structures using remote sensing and GIS. Thematic maps like land use/land cover, stream order, slope, soil, lineament, geomorphology and runoff were mapped and analyzed through weighted overlay analysis in GIS environment to obtain suitable locations for conservation structures. Water budget calculation was also done to find out the inflows and outflows. The study resulted in reflection of 120 potential sites for designing soil and water conservation storage structures in the study area which comprises check dams 71 in number followed by farm ponds with lining 36 in number and percolation tank 13 in number. Therefore, the study was found to be successful indemonstrating the use of remote sensing and GIS for locating suitable soil and water conservation structures at watershed level.ThesisItem Open Access FORECASTING OF WATER TABLE FLUCTUATIONS FOR PRIYADARSHINI WATERSHED USING ARTIFICIAL NEURAL NETWORK (Accession No. T06734)(dbskkv., Dapoli, 2019) Hittanagi, Prakash Basavanni; Bhange, H. N.Groundwater is an important natural resource essential for sustenance of life. Over 98% of the freshwater on the Earth lies below its surface. It is located below thesoil surface and largely contained in interstices of bedrocks, sands, gravels, and otherinterspaces through which precipitation infiltrates and percolates into the underground aquifers due to gravity. The total amount of water in the world is 1.4 billion km3. 97.5% of these waters are in the oceans and the seas and 2.5% is in fresh water. Sweet waters; 0.3% is in lakes and rivers, 30.8% in ground water, soil necropsy and marsh, 68.9% in the form of ice and permanent snow. Groundwater is one of the major sources of supply for domestic, industrial and agricultural purposes. The weekly Rainfall data, Temperature data, Solar data, Water level data and Permeability data of 9 years were used. Artificial Neural Network is an information processing paradigm that is inspired by the way biological nervous systems, such as the brain, The network is composed of a large number of highly interconnected processing elements called as neuron. They typically consist of hundreds of simple processing units which are wired together in a complex communication network. Each unit or node is a simplified model of real neuron which sends off a new signal or fires if it receives a sufficiently strong input signal from the other nodes to which it is connected, learning in this system involves the adjustment between neurons through synaptic connection. In this study feed-forward neural networks architecture has been used in predicting weekly water table depths. In this study, sensitivity analysis has been done to measure relative importance of each input variable for precisely predicting groundwater table fluctuations. Sensitivity analysis is done by removing one input parameter at a time from the model and testing its performance by comparing with original model. Considering training, validation and testing period and all the statistics, it is difficult to say which algorithm is better among the two selected for study. Because there was a lot of variation in all the statistics among the two selected algorithms for training, validation and testing period. But considering the testing period of all the nine wells it was found that LM algorithm was better than CG for wells i.e., well 1 (2-9-1), well 2 (2-9-1), well 3 (1-8-1), well 4 (1-6-1), well 5 (2-9-1), well 6 ( 1-9-1), well 8( 2-9-1) while CG algorithm was better than LM for wells i.e., well 7 (2-5-1) and well 9 (3-5-1) So these algorithms for particular well were selected for sensitivity analysis. As the results found were based on trial and error methods Levenberg- Marquardt (LM) algorithm provides better results than Conjugate Gradient algorithm. Levenberg- Marquardt (LM) best results for ANN network architecture of model for well 1 (2-9-1), well 2 (2-9-1), well 3(1-8-1), well 4(1-6-1), well 5(2-9-1), well 6 (1-9-1), well 7 (3-5-1), well 8 (2-9-1), well 9 (2-9-1).The predicted water level trend followed the observed trend closely, showing the accuracy of the network. In present study, results were found and based on sensitivity analysis models selected and their statistics for all the nine wells. It was observed that selected algorithms predicted the water table depths in a better way in terms of its performance statistics. The values of R for LM and CG were found to be 0.836 and 0.743, respectively. The observed values of RMSE for LM and CG were found to be 0.100 and 0.101, respectively. Similarly, the value of E for LM and CG were found to be -376.40 and -20.88, respectively. The above result concludes Levenberg-Marquardt predicts the water table depth better than Conjugate Gradient. (Keywords : ANN, Sensitivity analysis, Levenberg-Marquardt, Conjugate Gradient)
