land cover classification around

International Journal of Civil Engineering and OF Technology ISSN 0976 – AND 6308 (Print), INTERNATIONAL JOURNAL CIVIL(IJCIET), ENGINEERING ISSN 0976 – 6316(Online), Volume 5, Issue 11, November (2014), pp. 79-96 © IAEME

TECHNOLOGY (IJCIET)

ISSN 0976 – 6308 (Print) ISSN 0976 – 6316(Online) Volume 5, Issue 11, November (2014), pp. 79-96 © IAEME: www.iaeme.com/Ijciet.asp Journal Impact Factor (2014): 7.9290 (Calculated by GISI) www.jifactor.com

IJCIET ©IAEME

ANALYSIS ON LAND USE/LAND COVER CLASSIFICATION AROUND MYSURU AND CHAMARAJANAGARA DISTRICT, KARNATAKA, INDIA, USING IRS-1D PAN+LISS-III SATELLITE DATA Basavarajappa H.T, Dinakar .S, Manjunatha M.C Department of Studies in Earth Science, Centre for Advanced Studies in Precambrian Geology, University of Mysore, Manasagangothri, Mysore-570006, India

ABSTRACT Land is a non-renewable resource and mapping of LU/LC is essential for planning and development of land and water resources in a region of engineering projects under progress. Land is an area of the earth surface, which embraces all reasonable stable or predictably cyclic, attribute of the biosphere including the atmosphere, soil and underlying geology. Hydrology, plant and animal population are the results of the past and present human activity to the extent that significantly influences on present and future LU/LC system. Proper management and development of these lands should be initiated to increase the land productivity, restoration of soil degradation, reclamation of wastelands, increase the environmental qualities and to meet the needs of rapidly growing population of the country. Remote Sensing (RS) satellite data with its synoptic view and multispectral data provides essential information in proper planning of LU/LC conditions of the larger areas. An attempt have been made to delineate the level-1, level-2 and level-3 LU/LC classification system through NRSC guidelines (1995) using both digital and visual image interpretation techniques by Geographical Information Systems (GIS) software’s. The classification accuracy is found to be more in case of digital technique as compared to that of visual technique in terms of area statistics. Efforts have been made to classify the LU/LC patterns using False Color Composite (FCC) data of IRS-1D PAN+LISS-III (Band: 2,3,4) through MapInfo v7.5, ArcView v3.2, Erdas Imagine v2011 and ArcGIS v10. The final results highlight the potentiality of geomatics in classification of LU/LC patterns around Chamarajanagara district, Karnataka, in natural resource mapping and its management which is a suitable model for application to similar geological terrain.

79

International Journal of Civil Engineering and Technology (IJCIET), ISSN 0976 – 6308 (Print), ISSN 0976 – 6316(Online), Volume 5, Issue 11, November (2014), pp. 79-96 © IAEME

Keywords: LU/LC Classification; Visual & digital interpretation; Chamarajanagara; Mysuru; Satellite data. 1. INTRODUCTION Land use systems need thorough systematic monitoring and management to maintain food security, to minimize deforestation, conservation of biological diversity and protection of natural resources. It is necessary to enhance human occupation to the changing social, economic and natural environmental conditions. Rapid increase in population demands for more food, fodder and fuel wood have led to large scale environment degradation and ecological imbalance. In order to use land optimally, it is necessary to have firsthand information about the existing land use/land cover (LU/LC) patterns. Jacks (1946) reviewed land classification as it relates to the grouping of land according to their suitability for producing plants of economic importance. Land use refers to man’s activities and the various uses which are carried on land (Clawson and Steward, 1983). Land cover refers to natural vegetation, water bodies, rock/soils, artificial cover and other resulting due to land transformation. Land use describes how a parcel of land is used such as agriculture, settlements or industry, whereas land cover refers to the material such as vegetation, rocks or water bodies that are present on the surface (Anderson et al., 1976). The term LU/LC is closely related and interchangeable. LU/LC exposes considerable influence on the various hydrological aspects such as interception, infiltration, catchment area, evaporation and surface flow (Sreenivasalu and Vijay Kumar., 2000; Kumar et al., 1999). LU/LC provides a better understanding on the cropping pattern and spatial distribution of fallow lands, forests, grazing lands, wastelands and surface water bodies, which is vital for developmental planning (Philip and Gupta, 1990). 2. STUDY AREA 2.a Mysuru district: The study area include Cauvery and Kabini riverine plains flowing towards south easterly and both joins at Tirumalakudu Narasipura. Most parts of Nanjungudu and Mysuru taluks show gentle slope and plains with both cultivated seasonal crops such as irrigated and dry seasonal crops. The southern parts of Mysuru district is traversed by 3 sets of joints-trending in N-S, NE-SW and E-W direction & 4 sets of lineaments are noticed towards NNE-SSW, NNW-SSE, NESW & E-W. The study area is subjected to F1, F2, and F3 Sargur type of structure, deformational folds and joints formation in the past. 2.b Chamarajanagara district: The study area represents a part of Biligiri-Rangan Hill Ranges which belong to an oldest Precambrian hard rock terrain in southern Karnataka (Basavarajappa H.T., 1992). The eastern portion of the study area forms a hilly terrain with lofty mountains (BiligiriRangan hill ranges) raising about 1677m above MSL, run approximately towards N-S direction with thick vegetation. The western parts form a plain country with an average elevation of 686.25m with minor undulations. Honattikal, Chikkangiri betta and Honnamatti betta are some of the important tracts. The north western region is drained by major river Cauvery & Kabini which flow from west to east and both the river conflicts at Tirumalkudalu Narsipura. Suvarnavathi and its tributary, Hebba halla flows from south to north in the central part of the study area, in turn drain into river Cauvery (Azadhe T. Hejabi and Basavarajappa H.T., 2011).

80


The land adjoining the banks of meandering course of the river forms the most fertile and rich tracts of land, which is cultivated intensively for paddy and coconut. The paleo-channels of the study are also mapped using satellite data which gives additional information regarding water bearing zones like hidden aquifers, old river course, fractures and valley fills (Basavarajappa et al., 2008; 2013; Dinakar S and Basavarajappa H.T., 2005; Satish et al., 2008). 3. LOCATION The study area lies between 11°45’ to 12°15’N latitude and 76°45’ to 77°15’E longitude with total areal extent of 3,011 Km2 (Fig.1). The study area includes parts of 9 taluks of Karnataka state namely Yelandur, Kollegal, Chamarajanagara, Malavalli, Mysuru, Gundlupet, T. Narsipura, Nanjungudu and small patches of Tamil Nadu region (Sathyamangalam) in the southern and southeastern parts. Cauvery and Kabini are the two major rivers flowing in the study area in which Kabini is one of the tributary of River Cauvery.

81


4. CLIMATE & RAINFALL Generally weather is pleasant and the climate is divided into four seasons viz., pre-monsoon (Jan-Feb), south-west monsoon (May-Sept), north-east monsoon (Oct-Dec) and summer (MarApril). The average annual rainfall is 786.8mm (2004) with a major contribution south-west monsoon (44.45%). The annual minimum rainfall is recorded as 558.07mm (Kavalande rain-gauge station) while the maximum is 1455.43mm (2010) (Biligiri-Rangan temple rain-gauge station). There is a continuous rise in temperature attaining a maximum in the month of April and minimum during December. Wind speed is moderate during south-west monsoon and relative humidity is high (Dinakar S and Basavarajappa H.T., 2005).

6. METHODS & MATERIALS 6.1 Methodology: LU/LC maps are prepared using satellite image in conjunction with collateral data like SoI topomaps on 1:50,000 scale by taking permanent features such as road, tanks, co-ordinates, etc. Visual interpretation of IRS-1D PAN+LISS-III FCC of Band 2,3,4 on 1:50,000 scale is carried 82


out and various LU/LC categories are delineated. The satellite data of two seasons is acquired (Rabi in December 2002 and Kharif in October 2003) to estimate the spatial distribution of LU/LC pattern. These categorizations are done based on the classification scheme developed by National Remote Sensing Agency (NRSA, 1995). 6.2 Materials used: a. Topomaps: 57D/16, 57H/4, 58A/13 and 58E/1. Source: (SoI, Dehradun). b. Satellite Data: IRS-1D LISS-III of 23.5m Resolution (March & Nov-2001) and PAN+LISS-III of 5.8m, Date of pass 10-March-2003. Source: (National Remote Sensing Agency (NRSA), Hyderabad. c. GIS software’s: Mapinfo v7.5, Arc Info v3.2, Erdas Imagine v2011 and Arc GIS v10. d. GPS: Garmin 12 is used during Ground Truth Check (GTC).

Satellite data IRS-1D, LISS-III & PAN+LISS-III of 2 Season Geocoded

Data Source

Collateral data • SoI toposheet • Forest Map

Base Map

Classification System

Image Analysis

Image Interpretation

Preliminary Interpreted Map

Ground Truth Check

Post Field Correction/Modification

Final Land use/Land cover Map

Fig.3. Flow chart showing the methodology adopted in the preparation of Land use/Land cover map

Classification analysis using Geomatics: Information on land use/land cover is of utmost importance in hydrogeological investigation as the groundwater regime of a region is influenced by the type of land use/land cover. Hence the satellite based data is very much useful in preparing the precise land use/land cover maps in a very short time period as compared to the conventional methods using Geomatics. LU/LC classes such as built-up land, agricultural land (crop land), fallow land, plantation, forest (evergreen, deciduous, scrub, etc), wastelands (salt affected land, waterlogged 83


land, gullied/ravinous land, barren rock/stony waste etc), water bodies (rivers, streams, canals, lakes, etc) are delineated based on the image characteristics like tone, texture, shape, association, background, etc. The level-1 classification consists of 5 major categories such as built-up land, agricultural land, forest, wastelands, water bodies and others. These 5 major classes of level-1 are further divided into sub-categories of level-2; keeping the area under investigation. Level-3 classification has been done in detail on agricultural and forest lands to study the cropping pattern. Geomatics are the advent high-tech tool that can be used more effectively in natural resources management using Survey of India (SoI) toposheet, Satellite image with limited Ground Truth Check (GTC) using GIS software’s (Tiwari A and Rai B., 1996). This helps in analyzing, mapping and integrating the information database to generate thematic maps for development and management of natural resources (NRSA, 1995). Digital interpretation and post classification comparison techniques are adopted to find out the changes among various land uses over a period (Rubee and Thie, 1978; Likens and Maw, 1982; Priyakant et al., 2001). Lithological formations and geomorphological landforms are derived by visual image interpretation of IRS-1D PAN+LISS-III of False Color Composite (FCC) based on the image interpretation elements such as association, pattern, shadow, shape, size, tone, texture etc., and verified during the field visits. Drainage and slope maps are digitized using Survey of India (SoI) toposheets of 1:50,000 scale. 7. LEVEL-1 CLASSIFICATION 7.1 Built-up land: These are the land surfaces of man-made constructions due to non-agricultural use including buildings, transportation network, communication, industrial, commercial complexes, utilities and services in association with water, vegetation and vacant lands. Collectively, cities, towns and habitations are included under this category. The total aerial extent of built-up land is 61.71 Km2 (2.05%). 7.1.1 Urban (Towns and Cities): Land used for human settlement of population more than 5000 of which more than 80% of the work forces are involved in non-agricultural activities is termed as urban land use. Most of the land covered by building structures is parks, institutions, playgrounds and other open space within built up areas. The major urban settlements are noticed in Chamarajanagara, Kollegal and Yelandur taluks. Urban land occupies an area of 9.2 Km2 (0.31%). 7.1.2 Rural (Villages): Land used for human settlement of size comparatively less than the urban settlement of which more than 80% of people are involved in agricultural activities. Though the total number of rural settlements in the study area is 601 as per the toposheet information, only 483 villages can be clearly noticed from the satellite data due to less number of houses (less than 10 houses) in a village, inter spread with trees and agriculture fields especially in South Eastern parts of study area occupied by thick forest with hilly region. The area occupied by this class is about 52.4 Km2 (1.74%). 7.2 Agricultural land: These are the land primarily used for farming, production of food, fiber, other commercial and horticultural crops. It includes land under crops (irrigated and unirrigated), fallow, plantations, etc. The area under this category is 2043.21 Km2 (67.85%). 7.2.1 Crop Land: It includes those lands with standing crops as on the date of the satellite data acquisition. The crops may be either Kharif/Rabi or Kharif and Rabi seasons or double cropped. It includes land under crops (irrigated and unirrigated), fallow, plantation, etc (NRSA, 1989). The area under crops have been identified in both during Kharif (June to September) & Rabi seasons (October to February) are mapped. The land under double crop (land cultivated both during Kharif and Rabi seasons) have also been mapped and digitized.

84


7.2.1.a Kharif: Kharif includes standing crops from June to September in associated with rainfed crops under dry land farming and limited irrigation. Kharif crops include Jowar, Ragi, Horsegram and others in the study area. The prospect of Kharif crops mainly depends upon the regularity of monsoon to some extent on irrigation facilities. The cultivated land of Kharif season on FCC shows bright red tone. The areas in single crop system with moderately deep to deep soil on nearly level to very gently sloping with good to moderate groundwater potential/accessible surface water resources or both can be put into intensive cropping system. This land occupies an area of 1160.64 Km2 (38.54%). 7.2.1.b Double Cropped: This category has been identified and mapped using the two season satellite images. Most of the double crop areas are concentrated adjacent to the river Cauvery and Kabini flowing in north-western parts of the study area. The cropping intensity is very high due to physical factors such as flat terrain, fertile soil and irrigated from canal system. Paddy, Sugarcane, Groundnut, Sunflower and others are grown in this region. On FCC, the double crop show a dark red tone with square pattern representing soil covers with higher amount of moisture near the streams. The water table is found to be at shallow level, indicating the good groundwater prospects. Higher the growth of natural vegetation; higher will be the groundwater availability. The cultivated land at elevated zones show bright red tone generally representing the less amount of moisture and deeper 85


levels of groundwater prospect indicates the moderate groundwater prospect zones. Intensive agriculture is seen in north-western and central parts growing multiple crops in sequence on same land. They are mostly confined to valleys, low lands, alluvial tracts where the groundwater potential is good. The soils are deep, provide good groundwater yield with maximum nutrient holding capacity. This category covers an area of 650.35 Km2 (21.60%).

86


7.2.2 Fallow land: The agricultural land which is taken up for cultivation but is temporarily allowed to rest, uncropped for one more season, but less than one year. These are particularly devoid of crops at the time; when the imagery is taken from both seasons. On FCC, fallow land shows yellow to greenish blue tone, irregular shape with varying size associated with amidst crop land as harvested agriculture field. Fallow land are noticed near the villages of Triambakapura, Tavarugottemole, Mallaipura and Galipura. The total area under this category is 6.01 Km2 (0.2%). 7.2.3 Plantations: These are agricultural land with tree plantation or fruit orchards; planned by adopting certain agricultural management techniques. It includes mainly Coconut, Mango, Arecanut, Banana, Mulberry and other horticultural nurseries are noticed in south-western parts around Chamarajanagara taluk and Southern region. These plantations are undoubtedly considered to be lucrative as compared to agriculture crops; further no tedious maintenance is required for the plantation. Huge number of eucalyptus plantation is noticed in south-eastern parts which are covered by denudation hills. Differentiation of plantation from cropland is possible by multi-temporal data of period matched harvesting time of inter-row crop/flowering of the plantation crops. Overall, Rabi season data is found to be better discrimination of plantations from croplands. The total area under this category is 226.21 Km2 (7.51%). 7.3 Forest: It is an area (within the notified forest boundary) bearing an association predominantly of trees, other vegetation types capable of producing timer and other forest products. Satellite data has become useful tool in mapping the different forest types and density classes with reliable accuracy through visual as well as digital techniques (Madhavanunni, 1992; Roy et al., 1990; Sudhakar et al., 1992). Forest cover with 40% or move vegetation density (crown cover) is called dense or closed forest; while between 10-40% of vegetation density is called as scrub whereas