Use of Landsat TM data in location of sites for afforestation programmes in India

Use of Landsat TM data in location of sites for afforestation programmes in India

T.S. Kachhwaha
Remote Sensing Applications Centre, U.P.
Sector G, Jankipuram, Kursi Road,
Lucknow 226 007, India

Abstract
The forest cover of the country and specially that of Uttar Pradesh as much less than recommended in the National Forest Policy approved by Government of India in 1952. According, vigorous afforestation and social forestry programmes have been launched to increase the tree cover. One of the important aspects of these programmes is the selection of suitable location which can be acquired with ease without necessarily changing the present landuse which may there wise lead to socio-economic problems. Therefore, the best location for afforestation programmes in the first instance, would be those degraded lands which be not suitable for agiculature and are currently lying waste, followed by the degraded forest lands. This would not only increase the productivity of such land but would also lead to qualitative improvement of such degraded lands.

In the present study, Landsat - 5 Thematic Mapper (TM) data acquired on February 23, 1986 has been used. False Colour Composite (FCC) on 1:50,000 sale has been visually interpreted to identify and delineate degraded lands which could be utilized for afforestation and social forestry programmes.

Introduction
The forest wealth of the country is known to have depleted to a large edent in the past. Deforestation has resulted in a series of degradation processes philately leading to various kinds of environmental hazards, which not only depletes the biomass reserve but also degrades the land, affects the climate and disturbs the water balance of the area. Frequent periods of floods and droughts are the resultant effects of deforestation which ultimately influences the human and animal population to a great extent.

NRSA (19823), based on the satellite data of 1972-75 and 1980-82, has imported that forest cover in Uttar Pradesh was depleted by 1.65% in intervening 7 years and in the year 1982 it was only 7.14% of the total geographical area. Mishra et al. (1985), based n satellite data of 1981-83, have reported 11.29% first cover in Uttar Pradesh. In the Forest Statistics of U.P. (1983), State forest Department has however, reported 17.41% forest cover in Uttar Pradesh. These figures, although, mutually contradictory, do indicate an alarming situation of forest degradation. Evidently all these estimates are much less than recommended in the Nation Forest Policy approved by Government of India in 1952 which states that on an average 33% of total geographical area of the country should be under permanent forest cover.

Until about a few years ago, there was a lack of comprehensive awareness of the degradation of forest cover and the resultant consequences. During the lat decade, however, vigorous efforts have been made to rehabilitate the first cover through launching of afforestation programmes. These programmes can be implemented well only if the problems related to these programmes are well understood. One of the important aspects of such programmes is the location of those areas where such schemes can be implemented without adversely affecting the present landuse. This is particularly necessary because most of the forest land, encroached in the past, is currently being used for agriculture.

Since it is not possible to lead to status quo ante, it is imperative that such schemes are either taken up in degraded forest lands or on those lands which are currently not being put to any use but lying waste. It is possible to identify and map such lands where social forestry and afforestation programmes can be undertaken more effectively. This would not only increase the wooded area to meet ever increasing demand of timber, fuel wood etc., but would also simultaneously improve these lands. Perhaps even more important aspect of these programmes is the continuous monitoring of the afforested area. While the programme of tree plantation is generally executed satisfactorily, the rate of survival and plant growth which play an important role in the evaluation of these schemes is generally ignored in computing the statistics. It is thus essential that, apart from monitoring afforested areas, temporal assessment of wooded area should also be undertaken.

Satellite remote sensing technique is very effective in obtaining an accurate information on both the aforesaid aspects with minimum of turn-around time. It thus provides an excellent monitoring system in a cost effective manner. In fact the conventional methodologies can not provide a matching monitoring system. In the present study, an area of approximately 620 sq km has been selected in Fatehpur district, Uttar Pradesh, as a test site to demonstrate utility of remotely sensed satellite data for mapping those areas where afforestation programme can be undertaken. Monitoring capabilities of satellite remote sensing have already been amply demonstrated in eastern U.P. (Kachhwaha, 1985).

Test site
The study area lies between 25030' to 25045' N lat. And 81015' E long. And covers approximately 620 sq km of southern half of the Khaga Tahsil of Fatehpur district, and a small part of Allahabad district, Uttar Pradesh, and occupies most part of topographic map sheet no. 63 G/2 (Fig. 1).

The river Yamuna and its tributary Sasur Khaderi Nadi (=river) traverse through this area. The ravine and gully formation along these river systems has led to land degradation. Furthermore, land degradation has also taken place due to salinisation and waterlogging.

Fig. 1 Location Map of the study area

Materials and methods
The study area falls under Landsat - 5 Scene path-rows No. 143-042. Multispectral Thematic Mapper TM data of 23.2.1986 in the form of False Colour Composite (FCC) of part of two sub-scenes covering the study area with frame No. 2 & 3 on approximately 1:50,000 scale were interpreted. Survey of India topographical maps on 1:50,000 scale and Tahsil Maps on approximately 1:20,0000 scale were also used for base map preparation.

Visual interpretation of TM data based on image characteristics such as texture, pattern, colour, hue, shape etc. with limited field checks was carried out to delineate broad landuse categories with special emphasis on degraded lands. Base map information i.e. important habitations, means of transport, canals etc. were transferred on the thematic map from 1:50,000 topographic maps using Planvariograph 'Mapmaker'. Area under each category was computed using a Digital Electronic Planimeter.

Results
The thematic map showing sites of degraded land n 1:50,000 scale prepared form TM data indicates that most of the land in the study area is being utilised for agriculture (Fig. 2). There is no notified forest land in this area. The land has been classified into five major categories viz., agriculture had, scrubland, ravenous or gullied land, salt affected land and waterlogged or marshy land. Table 1 shows the broad landuse categories and the computed area under each category.

Table 1.: Area under each landuse category in the study area

Categories	Area(ha)	Percentage to study area
Agriculture Land	55,800	90.00
Scrub Land	273	0.44
Ravinous or Gullied Land	1,313	2.12
Salt affected Land	1,073	1.73
Waterlogged or Marshy Land	202	0.33
Other Miscellaneous Categories	3,339	5.38

Agriculture Land
About 90% of the total study area is being used for agricultural crops. It forms a part of Yamuna Flats comprising of low lands having dark clay soil. The main Kharif crop is Sorghum while paddy is also grown where irrigation facilities are available. The main Rabi crops are Gram and Wheat.
Ravinous or Gullied Land
Nearly 1313 ha of land is ravinous or gullied which is about 2.21% of the study area. Most part of this land is confined along Sasur Khadari Nadi (river). Some ravenous or gullied lands are also seen along Yamuna river mainly around Kishanpur, Garha and Kot village. This class represent only those ravenous gullied which are devoid of any vegetation cover.
Salt Affected Land
Salt affected land in the area is mainly confined to north of Dhata-Nodki (via Ghazipur) Road. It occupies an area of about 1073 ha which is 1.73% of the study area.
Scrub Land
In the study area, the vegetation cover is generally poor and is reprinted only by low scrub. Scrub vegetation, which is mainly confined to ravenous or gullied lands, have sparse and bushy growth with less than 30% crown density. It occupies an area of about 273 ha which is only 0.44% of the study area.
Waterlogged or Mashy Land
About 202 ha of land has been classified as surface waterlogged which is only 0.33% of the study area.

Discussions
The results of the satellite data interpretation shows that 90% of the land is under cultivation and is generally not available for afforestation or social forestry programmes. Obviously, it will not be desirable to change the state of this agriculture land, particularly when about 5% of the study area lies degrade and is not being put to any use. It would be a proper strategy to use such degraded lands to undertake afforestation and social forestry programmes. There are two very important advantages in selecting such lands for tree plantations. Firstly, acquisition of these land for tree plantation is relatively easier. Secondly it will arrest the process of further land degradation in the neighbouring area which may also otherwise lose otherwise lose their productivity due to ecological depredates. Tree plantation would make these lands productive, check further degradation thereby restoring ecological balance so that there is no drop of agricultural potential in the surrounding land.

Periodic monitoring of plant growth is important to assess the out-come of plantation schemes. Satellite remote sensing has the potential to assess the results of such programmes in view of synoptic coverage and repetivity of data acquisition at a regular interval of time and provide excellent monitoring system even in the inaccessible areas. Kachhwaha (1985) has shown utility of Landsat data for change detection in forest land and monitoring plant growth over a period of 8 years from 1975 to 1983 in Gorakhpur district of Uttar Pradesh.

Conclusion
Remotely sensed data obtained from satellite can be effectively used for quick and accurate identification and mapping of sites for afforestation and social forestry programmes. Repetivity of satellite data on a regular time interval provides excellent monitoring system which has potential to temporally assess the result of these programmes.

Acknowledgement
Author is thankful to the Director, Remote Sensing Application Centre, Uttar Pradesh, for granting permission to publish this paper.

Reference

Forest Department, Uttar Pradesh ( 1983). Forest Statistics; Uttar Pradesh, Lucknow 1983. A report, 266 p.
Kachhwaha T. S. (1985). Temporal Monitoring of forest land for change detectics and forest cover mapping through satellite remote sensing techniques, Proceedings of the 6th Asian Conference on Remote Sensing. November 21-26, 1985, Hyderabad, India. PP. 276-281.
Mishra , D. B., S. L. Dabral and M.K. Sharma, (1985). Mapping of vegetal cover in India -A case study of Uttar Pradesh. Proceedings of the 6th Asian Conference on Remote Sensing . November 21-26, 1985, Hyderabad India. PP. 207-516.
National Remote Sensing Agency , India ( 1983). Mapping of forest cover in India from Satellite imagery of 1972-75 and 1980-82. A project report.

Fig. 2. Map showing degraded lands- Potential sites for afforestation programmes.
(Based on satellite data interpretation)

* Present at 11th Asian Conference on Remote Sensing held at Zhongshan university, Gaungzho, China, November 15-21, 1990.