Use of IRS LISS-II data for Urban Land Use Mapping - A case study of Ludhiana city, Punjab, India

Use of IRS LISS-II data for Urban Land Use Mapping - A case study of Ludhiana city, Punjab, India

R Chaurasia, P K Sharma and Gurjit Singh Gill
Punjab Remote Sensing Centre, Ludhiana - 141 004 (India)

Abstract
The IRS-1A LISS-II multispectral data enlarged on 1:50,000 scale for Ludhiana city was visually interpreted to evaluate their suitability for urban land use mapping. Standard false color composites (FCC) in the form of diapositives )1:1M scale) and paper print (1:50,000 scale) was visually interpreted for mapping urban landuse. Different categories of land use identified on imagery were transferred to a base map prepared from survey of India toposheet their re calculated. A comparison of land use map toposheet and there are calculated. A comparison of land use map based on Remote Sensing data with conventional maps of Town and Country Planning Department for the year 1971 was also made to study the growth of the city. The analysis of the data revealed that most of the agricultural land within municipal boundary is fast converting to other activities and practically no vacant land exists in central area for future development. On the basis of this study, appears that IRS LISS-II data is suitable for identification of urban land use features and in mapping all the level I and II categories and some of the level III land use sub classes of NRSA classification scheme.

Introduction
Land is one of the important natural resources and as such it needs proper evaluation for sound planning and management (Singh and Roy, 1989). since the piece of land under different uses in nature is fixed, the increase in area under one category is possible only at the cost of other. Up to date information about the existing land use of one of the basic requirement for the preparation of integrated development plan and economic development porgramme of a region. Remote Sensing techniques can provide such information within short time at less cost and efforts (Gautam and Narayan, 1983; Gautam and Channaich, 1985). Irrational demands of land for rural and urban growth, transport and industrial development is putting huge pressure on agricultural and forest lands to shrink at the fast rate. In the city unplanned growth of urban settlements, appearance of slums and industrial units due to excessive increase in population is deteriorating local environment and making it unfit of human health and dwelling. At some places, immediate attention is called for to control the degradation of the living standard in the urban centres.

There liable and up-to-date land use data for Ludhiana city area is not available. Therefore, this study was undertaken with the aim to generate information about different land use categories and impact of human settlement and industrial growth on agricultural land. The study also evaluates the potential of the Is-1A, LISS-II data for delineation of various land use classes. The development of the Ludhiana city in context with the master plan 1971 and 1984 has also been studied.

Methodology
The IRS-1A LISS-II (FCC) imagery in the form of diapositives on 1:1M scale taken on March 5, 1989 was optically enlarged with Procom-2 to the scale of 1:50,00 and super imposed on the IRS-1A LISS-II Geocoded product (Paper print) of the same date and scale. The super imposition of diapositive enlargement on the hard copy imagery increased sharpness of the features which made recognition, location and delineation of different landuse classes through visual interpretation easy. The use classes delineated from the imagery were transferred on the standard base maps prepared from the survey of India (SOI) toposheet. Landuse categories of classification scheme proposed by National Remote Sensing Agency (NSA), Hyderabad, India, were finalized on the basis of Agency (NRSA), Hyderabad, India, were finalized on the basis of ground verification using the Ludhiana city map on 1:2,500 scale for the year 1985. The area of different land use categories was calculated from the map using Planix digital planimeter.

Results and Discussions
On the basis of the information obtained by the identification of the physical characteristics from the imagery and their verification in the field, four level I categories, 14 level II as well as level III categories were identified and mapped. The tone, texture pattern, shape and size were the photo elements used in the interpretation of imagery and identification of land use units. The scale and spatial resolution of the IRS LISS-II data (36.25m) were kept in view while interpreting the data. The detailed classification is given in table-1.

Table 1 : Urban land use classes identified and mapped form IRs-1A LISS-II data

The major level I land use categories mapped are built up, agricultural and water bodies. Under built up category seven sub-classes viz., residential, industrial/commercial, transportation, recreational, public/semi-public, mixed built-up land and others were identified and mapped. In the agricultural category, tow sub-classes namely crop land and plantation were derived from the imagery. Under water bodies three sub-classes: canals, ponds and nala were mapped. Under the residential category, three sub-classes namely dense, medium and thin were identified. The densely populated areas are concentrated around main railway station in the heart of the city. These area lack open spaces like park, play ground and greenery. The houses also lack court yards. The roads and streets are also narrow.

Linear features like roads, railways, canals etc., were quite distinct and identifiable by naked eyes from IRS-1A LISS-II Geocoded products/ enlargements of 1:1M films diapositives because of their contrast with the surrounding features. The big institutions like Punjab Agricultural university is visible with its boundary lines marked by thick plantation. With the super imposition of the two imageries i.e enlarged film diapositive over 1:50,000 scale hard copy, the other land cover / landuse classes like park, stadium, railway yard, dense, medium n low built up areas in the city could be identified. In the sub urban area, the villages merged with urban areas, however, in the agricultural area isolated villages are clearly seen. For the identification and mapping of land use categories, local knowledge of the areas was helpful. Only the major network of the road system in the city area could be identified.

The distribution of various landuse categories delineated through visual interpretation of IRS-1A data are shown on Map 1. For reference the urban maps prepared by Department of Town and Country Planning and survey of India are also given (Map 2 & 3).

Map 1

Map 2

Map 3
At the existing resolution of IRS-1A LISS-II data the identification and detection of the some of the land use classes for urban areas are not possible. This results in the merging of minor classes into the adjoining classes (Pathan et al., 1988). The areas falling just under the range of the resolution of the IS-1A data required detailed ground survey for its correct delineation. The discrepancy in the data of the various groups of the landuse classes obtained from IRS-1A LISS-II and that from resulting in commissioning and omission of classes during visual interpretation.

The comparison of land use map prepared through Remote Sensing with conventional maps prepared by the Town & Country Planning Department for the year 1971 and 1985 suggests that in the recent years, the residential and built up land is increasing very fast. The low lying areas have also been acquired and utilized for residential use. The lands fro agricultural purposes are being put to both industrial and residential uses.

Conclusion
The IRS-1A LISS-II data can provide the information required for mapping urban land use through visual interpretation. The level III land use classes can be identified by enlarging the IRS-1A imagery to 1:50,000. the analysis of the data in the form of diapositive and hard copy of the same date when super imposed, can increase the sharpness of the boundary of the various landuse classes and thus enhance the accuracy.

References

Gautam, N.C. and L.R.A Narayan 1983. Landsat MSS data for land use and land cover inventory and mapping. J. Indian Soc. Of R.S.11(3) : 15-27.
Gautam N.C and C.H Channiah. 1985. Land use and land cover Mapping and change detection in Tripura using satellite Landsat data. J. Indian Soc.of R.S. 6(3&4) : 517-528
Pathan, S.K. P. Jothimani, K.V., Sitarama Rao, H.D., Patel and V.A. Varmaji. 1988. Evaluation of IRS-1A, LISS-II, Landsat TM and SPOT data for urban land use mapping. - A case study of Ahemdabad City. Remote Sensing application using IRS-1A data. scientific note. Space application centre ISRO, Ahemedabad pp. 141-148.
Singh, B.M. and A.K. Roy 1989. Remote Sensing for integrated survey of urban environment. J. Indian Soc. of R.S. 17 (3): 109-114.
Singh, B.D. and A.K. Goel, 1984. fluviomorpholgical investigation in flood plain area of river satluj I the north east of Ludhiana (Punjab State) using photo interpretation technique. J. Indian Soc. of R.S. 12(2): 48-52.

Table 2: Area of major land use categories in Ludhiana city obtained through connectional (1985) and Remote Sensing (1989) technique.

Sl. No.	Land use category	Conventional		Remote Sensing
Sl. No.	Land use category	Sq. km	%age	Sq. km	%age
1	Residential	35.5	32.3	36.8	33.0
2	Builtup land (Commercial, Institutional, Public and semi-public)	1.6	1.5	0.9	0.8
3	Industrial	8.9	8.1	9.8	9.0
4	Recreational	0.2	0.2	0.2	0.2
5	Transportation	6.5	5.9	4.0	3.7
6	Public/semi-public	8.8	8.0	5.7	5.3
7	Agricultural	48.3	43.9	41.9	38.6
8	Water bodies	0.2	0.2	0.3	0.2
9	Waste land	-	-	-	-
10	Unclassified	-	-	8.9	8.2