Home Articles Mapping Water-logged Area

Mapping Water-logged Area

M. Sai Sudarsanam1, Ranvir Singh1 and P. K. Garg2
1Department of Hydrology
2Centre for Remote Sensing, Department of Civil Engineering, University of Roorkee
Roorkee 247 667 Uttaranchal, India
Email: [email protected]    In India, water logging in irrigated fields was first noticed during 1850 in Punjab State. By 1855, rise in water table and salinity problems developed in Haryana State under Western Yamuna Canal System. Later in 1892, problems were also noticed in lower Chenab Canal System, followed by other project areas in Punjab. Similar problems were also noticed in Maharashtra after the introduction of irrigation in Nira valley. Drainage problems during 1870s and 1880s have been tackled with moderate canal re-alignment and improvement of drainage systems. But present-day scenario is much more complex with vast extension of irrigated areas, intensive irrigation practice with multiple cropping, adoption of high water consuming crops and the irrigation management over-burdened with socio-politico-economical considerations.

Why mapping?
In the thrust for rapid irrigation development, drainage of irrigated areas in the country has not been paid adequate attention in the planning process, partly due to lack of requisite data and partly due to resource crunch. Over the years, the problems of water logging, hitherto neglected, have surfaced noticeably in different proportions of magnitude and intensity in various parts of the country. The possible causes of water logging are presented in Fig. 2. The Working Group, set up by the Ministry of Water Resources (1986) for study of water logging soil salinity and alkalinity adopted the norms of water logged areas for the irrigated lands as follows: (1) Water logged areas due to rise in water table (Water table within 2m of land surface), (2) Potential areas for water logging (Water table between 2m to 3m below) and (3) Safe areas (Water table below 3m of land surface).

Area Mapped
The study area falls in Survey of India toposheet o. 53D/10 at 1:50000 scale, covering approximately 677 km2 of area, between latitudes 28030/ to 28045/ N and longitudes 76030/ to 76045/ E of Jhajjar District (Haryana). Jhajjar District was carved out from District Rohtak in July 1997 with five Community Blocks namely Jhajjar, Bahadurgarh, Berikhas, Matanhail and Sahlawas. The climate of the area under study is sub-tropical, semi-arid and monsoon type. The area is covered by the Western Yamuna Canal command and partly under Sahibi basin. Sahibi is only a seasonal stream flowing through the south-eastern part of Jhajjar Tahsil. During rains, the water instead of flowing into the river/streams, follows a course along inland depressions in the eastern and southern parts of the study area and these parts get flooded. According to the studies made in 1991 by HARSAC (Haryana State Remote Sensing Application Centre), Hisar, six categories of wastelands exist in the study area, covering an area of 7200 ha. (10.63%) of total study area (Table 1).

Table 1: Wastelands categories in the study area
(based on maps prepared by HARSAC, Hisar, 1997)
S. No. Landuse/Land Cover Category Area HARSAC Study (1991-92) % Area Area (ha) Present (2000) % Area Change (1991- 2000)
1 Built-up area 1240.25 1.83 1306.99 1.93 + 66.7
2 Agricultural land 58287.89 86.1 58813.70 86.88 + 525.81
3 Forest land 212.67 0.32 0.00 0.00 – 212.67
4 Wasteland 6349.14 9.38 5600.54 8.27 – 748.60
5 Water bodies 283.87 0.42 116.77 0.17 – 167.10
6. Water logged area 1321.00 1.95 1856.82 2.75 + 535.82
Total 67694.82 100.00 67694.82 100.00 0.00

Way of Identifying the Areas

Interpretation from satellite image: IRS LISS III satellite image of May 2000, procured from NRSA, Hyderabad, has been used and visually interpreted for land use and land cover classes including water bodies and water logged area. The areal extent of these classes is presented in Table 2. The interpreted water logged area map is shown in Fig. 1. The topographical map is also used for the extraction of features viz. canal/drainage network, road network and villages, including other land features, contours and spot levels. Some of these features have been updated using IRS image.

Analysis of ground water table data: Ground water data from pre-monsoon 1989 to pre-monsoon 2000 have been collected from Ground Water Cell, Department of Agriculture, Chandigarh and Haryana State Minor Irrigation Tube Well Corporation, Karnal and Rohtak. The depth to ground water table contour map and isohyetal maps has been prepared using Kriging option of SURFER software. The gradual variation in the sub-soil water logging conditions from 1989 to 1999/2000 has been studied.

The sub-soil water logged area which was concentrated in the north-western region of study area has been progressively increased towards the central north and extended up to the central portion up to Beri Khas/Jahajgarh villages during the above mentioned period. Based on the ground water table conditions, water logged area increased from 2360 ha. during pre-monsoon, 1989 to an area of 6854 ha during pre monsoon, 2000. The net water logged area during pre-monsoon 2000, based on the surface and sub-surface conditions are found to be 8059 ha. which works out to be about 12% of the study area. This is a serious problem of land degradation as the soils in these areas are gradually turning water-logged and saline, and therefore unfit for cultivation of conventional local crops.

Use of GIS: ARC/INFO GIS has been used to digitise the (i) water logged map, as identified from IRS image, (ii) soil map with eight soil series, as prepared by Haryana Agricultural University, Hisar (1991-92), (iii) geomorphological map, prepared by HARSAC, Hisar, and (iv) land use/land cover map, prepared by HARSAC, Hisar (Wetlands of Haryana, 1997). The area of land use/land cover categories identified by HARSAC as well as from IRS images of 2000 used in the present study, is presented in Table 2.

Table 2: Landuse/Land cover details of the study area
S. No. Landuse/Land Cover Category Area HARSAC Study (1991-92) % Area Area (ha) Present (2000) % Area Change (1991- 2000)
1 Built-up area 1240.25 1.83 1306.99 1.93 + 66.7
2 Agricultural land 58287.89 86.1 58813.70 86.88 + 525.81
3 Forest land 212.67 0.32 0.00 0.00 – 212.67
4 Wasteland 6349.14 9.38 5600.54 8.27 – 748.60
5 Water bodies 283.87 0.42 116.77 0.17 – 167.10
6. Water logged area 1321.00 1.95 1856.82 2.75 + 535.82
Total 67694.82 100.00 67694.82 100.00 0.00

The waterlogged area has increased by 535.82 ha in a period of about 9 years. In both the studies, however, smaller areas of less than 2.25 ha could not be identified and included mainly due to limitation of visual interpretation of satellite image. Forest has disappeared in year 2000 and therefore the forestland has decreased. Wasteland has decreased by 748.60 ha, which has been perhaps replaced by built-up area and agricultural land. Water bodies cover smaller area in year 2000, as these are identified based on pre-monsoon image, whereas HARSAC has used pre and post images to delineate the water bodies and therefore the area covered by this class is more.

Rainfall data of stations in and around the study area from 1985 to 1999 have been collected form the Directorate of Land Records (Haryana), Chandigarh. There are 3 rain gauge stations falling in the study area namely Dujana, Beri and Jhajjar and 9 stations in the vicinity of area. The time series of rainfall data of the stations falling within the study area were studied. The rainfall data have been used to prepare mean isohyetal map using SURFER package. The isohytals are digitised in GIS environment and bifurcated into five zones ranging from 470 mm to 590 mm with an interval of 40 mm.

A buffer map for the existing network of canals has been prepared in ARC/INFO with 5 different classes ranging from nearest area to the canal (i.e. within 150 m.) to 600m with gradual rise at 150m. The assumption behind the preparation of a buffer map is that the area nearer to a canal is more prone to water logging than an area far away from the canal. Non-spatial data are also created and linked to the themes of above maps. These maps, including land use map prepared from IRS satellite image have been registered together in GIS environment for their integration and analysis.

Major findings
The water-logged areas are identified in several regions, adjacent to canals, in agricultural lands, in natural depressions, etc. An area of 1857 ha. is found to be water-logged as identified through IRS remote sensing imagery of year 2000, whereas an area of 6854 ha is found to be water-logged based on the subsoil water conditions (within 3m below ground level). According to a study made by HARSAC, Hisar (Wetlands of Haryana, 1997) using November, 1991 and May, 1992 satellite imagery, the water-logged area has been found to be 1039 ha. during pre-monsoon (May 1992) and 1321 ha. during post monsoon (November, 1991). Therefore an increase in water-logged area of about 536 ha is observed in a period of 9 years (1991-2000). The net area affected due to both surface and sub surface water logging is found to be 8059 ha., about 12 % of study area.

The digitised map of water-logged area (based on the analysis of IRS satellite image of May 2000) was found to have 55 polygons. Representative polygons covering a total area of 1393.5 ha. (75% of total identified surface water-logged area) have been selected and analysed in GIS for finding out the association of water-logged areas with the spatial pattern of soils, geomorphology landuse, rainfall and ground water conditions. The association of the water-logged areas with the soils, slope, land use, rainfall in addition to the ground water conditions is given in Table 3.

Table 3: Association of water logged areas with the spatial parameters

Soils series:    Kal – Kalanaur; Lad – Ladhaut; Kar – Karauntha; Ism – Ismaila; Has – Hasanpur 
Land use:    WL – Wastelands; WB – Water bodies; AGRI – Agricultural land; BU – Built-up land

The LINEST command of Microsoft Excel package has been used for assessing the sequence of influential parameters causing water logging conditions. The present study is concentrated on investigating a relationship between the dependent variable i.e. water logging and the six parameters being the independent variables. The input for the LINEST command is the representative data covering 75% of the total water logged area. Among the five categories of rating for each polygon, the weightage covering the highest magnitude of area constituting each parameter is considered and the corresponding weightage has been selected as an input for the LINEST command. Out of the several outputs of the LINEST command, the standard error of each parameter has been selected for estimating the sequence of the most influential parameter, which obviously will have a least standard error of estimation.

The surface water-logged area has been delineated from IRS LISS III satellite imagery of May 2000 and is found to be 1857 ha. An increase in waterlogged area of about 536 ha is observed in 9 years period, when compared with the results of the study carried out by HARSAC. Satellite image provided update information about the ground condition and hence reduced the quantum of fieldwork. Repetitive satellite images can further provide information on areal changes in waterlogged area with time. The spatial and non-spatial database using various parameters like ground water, rainfall, soil, geomorphology, land use has been created using ARC/INFO GIS. The association of water logged area with the terrain and hydrological parameters have been studied using GIS. The sequence of influential parameters causing waterlogging conditions in the study area has been determined. Such information could be used to prioritise zones in the study area for reclamation work. It is a known fact that Haryana belt is mostly falling in the alluvial terrain and most of the land has flat topography which is more congenial for water logging if precautions for optimal use of irrigation water and planning for vertical/horizontal drainage are not opted for. Remote Sensing and GIS are powerful tools, which could be used effectively to study the dynamic behaviour of water-logged area.