Home Articles Landuse-Land cover mapping through digital image processing of satellite data- A case...

Landuse-Land cover mapping through digital image processing of satellite data- A case study from Panchkula, Ambala and Yamunanagar districts, Haryana state, India

Arvind.C.Pandey
Haryana State Council for Science & Technology, Chandigarh.
Email:[email protected]

M.S.Nathawat
Birla Institute of Technology, Mesra, Ranchi (Jharkhand).
Email:[email protected]

Introduction
The landuse- land cover pattern of a region is an outcome of both natural and socio-economic factors and their utilization by man in time and space. Land is becoming a scarce commodity due to immense agricultural and demographic pressure. Hence, information on landuse-land cover and possibilities for their optimal use is essential for the selection, planning and implementation of land uses schemes to meet the increasing demands for basic human needs and welfare.

Increasing human interventions and unfavorable bio-climatic environment has led to transformation of large tracts of land into wastelands. Satellite remote sensing play an important role in generating information about the latest landuse-land cover pattern in an area and its temporal changes through times. The information being in digital form can be brought under Geographical Information System (GIS) to provide a suitable platform for data analysis, update and retrieval.

The present study was carried out to evaluate the present status of landuse-land cover in the districts of Panchkula, Ambala and Yamunanagar on 1:50,000 scale by using digital satellite data of IRS- 1D.

Location
The Panchkula (areal extent 893 sq. km.), Ambala (1472 sq. km.) and Yamunanagar (1779 sq. km.) form the northernmost districts of Haryana bounded by States of Punjab in the west, Himachal Pradesh in north and Uttar Pradesh in east. The districts lies within longitude 760 33′ to 770 36′ (N) and latitude 290 55′ to 300 50′ (E) covering an area of about 4145 sq. km. They are covered under topographical sheet No. 53/B, 53/F and 53 G of Survey of India (1:250,000).

Materials for the Study
The Indian Remote Sensing Satellite (1RS-1D) digital data of LISS-III sensor, which provide a resolution of 23.25 meters in multispectral mode was utilised in the study. To estimate the spatial distribution pattern of landuse-land cover in Rabi and Kharif seasons, the satellite data of two seasons was acquired The details of satellite data used in the study is given below:

S.No. Path/Row for IRS- 1D, LISS -III
Digital Satellite Data
Date of pass
RABI KHARIF
1 95-49 February 2000 October. 2000
2 95-50 February 2000 October. 2000
3 96-50 February 2000 October. 2000

Survey of India (SOI) topographical sheets of the said district on 1:50,000 scale were utilised for registration of satellite data, selection of ground control points and locating training sets as well as to identify and authenticate the various cultural features on the satellite image.

Administrative maps of three districts on 1:50,000 scale were also utilised to demarcate the study area on the satellite images.

Landuse-land cover (LU/LC) classification scheme
In the present study the Landuse-land cover (LU/LC) categorisation in the said districts is envisaged based on the classification scheme developed by National Remote Sensing Agency (NRSA, 1995), table-1. The modification in the categories at level II is done keeping in view the area under investigation.

Table 1: Landuse-land cover (LU/LC) classification scheme
(NRSA, 1995).

S.No. LEVEL-I LEVEL-II LEVEL-III SYMBOLS
1. Built up Land 1.1 Towns/Cities
1.2 Villages
  01
02
2. Agricultural Land 2.1 Crop land + 2.1.1 Kharif
2.1.2 Rabi
2.1.3Kharif + Rabi
(Double Crop)
03
04
05
2.2 Fallow $
2.2 Plantation #
  06
07
3 Forest a 3.1 Evergreen/Semi-evergreen
3.2 Deciduous (Moist & Dry)
3.1.1 Dense
3.1.2 Open
3.2.1 Dense
3.2.2 Open
08
09
10
11
3.3 Scrub Forest
3.4 Forest Blank
3.5 Forest Plantations
3.6 Mangrove
  12
13
14
15
4. Wastelands 4.1 Salt Affected Land
4.1 Waterlogged Land
4.2 Marshy / Swampy Land
4.3 Gullied / Ravenous Land
4.4 Land with Scrub
4.5 Land without Scrub
4.6 Sandy area (Coastal and desertic)
4.7 Mining / Industrial Wasteland
4.8 Barren Rocky / StonyWaste / Sheet Rock Area
  16
17
18
19
20
21
22
23
24
5. Water Bodies 5.1 River/ Stream
5.2 Canals
5.3 Lake * /Reservoirs /Tank


25
6. Others 6.1 Shifting Cultivation
6.2 Grass Land / Grazing land
6.3 Salt Pans
6.4 Snow covered / Glacial Area
6.2.1 Dense
6.2.2 Degraded
26
27
28
29
30

+ It includes land under agricultural crops during Kharif Rabi (both irrigated + un irrigated) and
the area under double crop (cultivated during Kharif as wall as Rabi seasons).
$ It is the vacant land without crop during both Kharif and Rabi seasons.
# It includes all agricultural plantation like tea, coffee, rubber, coconut, areca nut, citrus woodland and others
a It includes those which occur within the notified forest boundary as shown on the Survey of India topographic maps on 1:50.000 scale. Those occurring outside the notified areas are also included under forest class, but the area estimated of the two will be shown separately.
* It includes inland fresh water lakes, salt lakes, coastal lakes and lagoons.

Methodology for the Study
The methodology followed in the digital classification of satellite data is as under:

  1. Digitization of administrative base maps of three district and bring them to a projection system.
  2. IRS-1D, LISS-III, digital data procured for Landuse-land cover mapping was geo-referenced with Survey of India (SOI) topographical sheets. The two satellite scenes covering the said districts were mosaiced.
  3. Digital Image Processing (DIP) was done using EASI-PACE software to prepare landuse land cover maps. For this purpose training sets were demarcated on the False Colour Composite (FCC) image of the study area belonging to both the season. Based on the training sets the digital image was utilised for supervised classification. Select training sets were checked in the field to authenticate the landuse category assigned to each training set in the FCC. The accuracy of classified image was checked through confusion matrix. The classification was further refined till the desired accuracy is reached. After proper classification of the images of both Kharif and Rabi season, these images were aggregated based on the truth table prepared by the resource scientist.
  4. Normalised Difference Vegetation Index (NDVI) image was generated separately for forest classification. The NDVI technique allows for precise identification of cropped areas, dense and open forest areas and water bodies. The Rabi season data was utilised for generating NDVI image. It was observed that higher NDVI values are associated with larger leaf area and larger green biomass of canopy.
  5. The extracted forest and water bodies from NDVI image were draped over the classified image.
  6. The classified raster image was converted to vector layer and then brought under GIS environment for generation of Landuse-land cover maps with proper labels. Finally the area calculation for various categories of landuse-landcover was done.

    The various categories of landuse and land cover observed in the three districts can classified into five major groups, viz., built-up land, agricultural land, forestland, wasteland, and water bodies. The spatial distribution of landuse-land cover categories in the three districts are described below.

Spatial distribution of Landuse- land cover

  1. Panchkula District

    Agricultural land
    Landuse-land cover map shows that nearly 36 percent of the district area is used for agricultural of which Kharif crops covers the largest area (19.34%) followed by double-crop land (8.92). The fallow land occupies only 1.36 percent of the area. The main crops are the paddy and wheat in the Rabi and maize and pulses in the Kharif season.

    Forest The forest area is mainly confined to Siwalik hill ranges and parts of Piedmont zone. An area of nearly 272 km2 (30%) is marked under reserved forest of which 105 km2 & 130 km2 area is under good & moderate forest respectively. An area of 147 km2 is marked outside reserve forest of which only 72 km2 has been considered as good forest.

    Wasteland
    It occupies an area of about 16 percent of the geographical area of district and mainly consists of land with or without scrub (112.71% ) and sandy area (1.07% ).

    The remaining area is covered by built-up land and water bodies. The built up land constitute about 28 km2 areas, which is mainly comprised by Panchkula township.

Table No. 2: Landuse-landover categories and their aerial coverage in Panchkula district.

Level I Level II Area in sq. km. % of district area
1. Built-up land City / Village
City plantation
28.02
0.78
3.14
0.09
2. Agricultural land Kharif crop
Rabi crop
Double crop
Fallow land
Plantation
177.15
46.58
79.69
12.16
5.93
19.84
5.22
8.92
1.36
0.66
3. Forest Reserve forest(Poor)
Reserve forest (Moderate)
Reserve Forest (Good)
Outside forest (Poor)
Outside forest (Moderate)
Outside forest (Good)
36.45
105.36
130.60
44.13
30.23
72.75
4.08
11.80
14.62
4.94
3.38
8.15
4. Wasteland Land without scrub
Sandy area
113.51
9.54
12.71
1.07
5. Water Bodies Lakes/ tanks/river 0.22 0.02
Total 893.10 100.00
  1. Ambala District

    Agricultural land
    About 85% of the total geographical area of district is occupied by agricultural lands of which double crop area covers 653 km2 (44%) and Rabi is cropped in 196.71km2 area (13%) suggesting availability of adequate water for irrigation. The Kharif crop occupies 25% of total geographical area of district. The fallow land occupies only 7% whereas agricultural plantation is only 1% of the total area. The occurrence of 7% fallow land is appreciable considering the good irrigation facilities and types of soils, which are non-calcareous and well drained. The principal crops in the district are wheat, rice, sugarcane and barley, which is grown in irrigated areas whereas maize, groundnut, bajara and pulses are the main Kharif crops.

    Wasteland
    Wastelands occupy about 8.7% of the district area, which mainly includes land with or without scrub and sandy areas.

    There is no forest area demarcated in the district. The built up land constitutes 5.91% of the district area. It is interesting that plantation within the city covers an appreciable area of about 13.8 km2 , which is mainly concentrated in the Ambala cantonment.

    Table No. 7.3: Landuse-land cover categories and their aerial coverage in Ambala district.

    Level I Level II Area in sq. km. % of district area
    1. Built-up land City / Village
    City plantation
    87.08
    13.83
    5.91
    0.94
    2. Agriculture Land Kharif crop
    Rabi crop
    Double cropped
    Fallow land
    Plantation
    370.05
    196.71
    653.34
    7.01
    6.14
    25.13
    13.36
    44.37
    0.47
    1.10
    3. Wasteland Land with/ without scrub
    Sandy area
    118.45
    9.57
    8.04
    0.65
    4. Water Bodies Lakes/ tanks/river 0.41 0.03
    Total 1472.58 100.00

  2. Yamunanagar

    Agricultural Land
    About 77.5% of the geographical area of the district belongs to this category. Double crop and Kharif crop areas make up nearly 38.23% and 30.44% respectively whereas Rabi crop area is only 6.42%. The fallow land constitutes 0.53% whereas the plantation covers around 2.01%. The dominance of agricultural plantation in Yamunanagar district reflects the changing attitudes of farmers from conventional agricultural practices to agro forestry. The principal crops of districts are wheat, sugarcane and paddy in the irrigated areas whereas pulses and maize form main crops of Kharif season.

    Forest
    Forests occupy about 10% of the district area and are mainly concentrated in the Siwalik hills and Piedmont zone (Kandi belt). The majority of mapped forest area (9%) lies within the reserve forest boundaries and only 1% forest area is classified under outside reserve forest category. Within the reserve forests the good forests occupies the maximum area of about 96 km2.

    Table No. 3: Landuse-land cover categories and their aerial coverage in Yamunanagar district.

    Level I Level II Area in sq. km. % of district area
    1. Built-up land City / Village
    City plantation
    62.44
    5.09
    3.51
    0.28
    2. Agricultural Land Kharif crop
    Rabi crop
    Double cropped
    Fallow land
    Plantation
    541.73
    114.14
    680.27
    9.53
    5.78
    30.44
    6.42
    38.23
    0.53
    2.01
    3. Forest Reserve forest (Poor)
    Reserve forest (Moderate)
    Reserve Forest (Good)
    Outside forest (Poor)
    Outside forest (Moderate)
    Outside forest (Good)
    19.95
    45.78
    96.01
    4.91
    5.33
    12.73
    1.12
    2.57
    5.39
    0.28
    0.30
    0.72
    4. Wasteland Land with/ without scrub
    Sandy area
    107.21
    32.94
    6.03
    1.85
    5. Water Bodies Lakes/ tanks/river 5.70 0.32
    Total 1779.51 100.00

    Wasteland
    Wastelands occupy about 140.15 km2 of district area and are characterised by land with/ without scrub and sandy area.

    The rest of the area (4%) is covered under built up land, water bodies etc. The built up land comprising 62.44 km2 mainly lies within the twin township of Yamunanagar and Jagadhri. The plantation within the city constitutes about 5 km2.

Conclusion
The heterogeneous climate and physiographic conditions in Panchkula, Ambala and Yamunanagar districts has resulted in the development of different landuse-land cover types. The present status of landuse-land cover in the districts of Panchkula, Ambala and Yamunanagar as evaluated by digital analysis of satellite data indicates that majority of areas in these districts (36% in Panchkula, 85% in Ambala 77.5% in Yamunanagar) are used for agricultural purpose. The hilly regions of the Panchkula and Yamunanagar districts exhibits fair development of reserve forests (30% and 9% respectively). Wasteland occupies about 16%, 8.7% and 7.88% area in Panchkula, Ambala and Yamunanagar districts respectively. It is inferred that landuse – land cover patterns in the area are generally controlled by agro-climatic conditions, ground water potential and hosts of other factors like irrigation facilities, soil characteristics, socio-economic status and demography. Land degradation due to soil erosion is prevalent in the steeply sloping Siwalik ranges of the districts whereas soil salinization, water logging and dumping of mining/industrial waste causing deterioration of productive agricultural land in the plane areas of these districts. There is an urgent need to reverse this trend and to restore the productive capacity of land in order to meet the demands of increasing population and other development needs. The remote sensing with its multispectral, multitemporal and synoptic view has the potential to provide accurate spatial and temporal information on landuse-land cover of a region in a time and cost effective manner.

Bibliography
Drury, S.A. (1987): Image interpretation in geology, Allen and Unwin publication, London.

Manual of procedure for preparation of wastelands digital data base using remote sensing & GIS techniques- NRSA, DOS, 1997.

Data User Hand book, NRSA, DOS, 1998.

Pandey, A.C., 2001 Geology and Morphotectonics of a part of Lesser Garhwal Himalaya (Tehri-Uttarkashi districts) Using Remote Sensing and GIS techniques-Unpublished thesis submitted to the University of Delhi for the degree of Doctor of Philosophy.