GISdevelopment.net ---> AARS ---> ACRS 1992 ---> Agriculture

Analysis of Rangeland vegetation using Remote Sensing

Shigeru Takahata
Tropical Agriculture Research Center (Japan)


Introduction
Based in human history, hunting/collection and grazing had originated in Eurasia. Even now, it is estimated that there are about 20 million km2 of drylands where approximately 500 million people. Sheep and goat grazing which originated in the Middle east has a long history of more than 7,000 years. Old stone mosaic pictures in ruins showed the ancient fauna. The inhabitants drew many kinds of animals including lion, wolf and elephants which dies out a few thousand years ago in the Middle East. The existence of these animals suggested that fertile grasslands were abundant in this area, with presently, a poor fauna and floors persist. Although the change of climate can be ascribed to this degradation, prolonged unsuitable land use resulted in the depletion of the rich grasslands. Generally speaking, there is a growing awareness that the global environment is being threatened, with the spread of desertification. The development of measures for the prevention of desertification should be promoted for the continuation of pastoralism and for the conservation of the environment. The currently study aimed at evaluating the conditions of the steppe using Remote Sensing, to monitor the process of desertification.

Outline of the test site :
This study was conducted in the Maragha National Range (Syria) which is located at 35 33’ N, 37 41’ E. (fig. 1). Annual rainfall is estimated at about 200mm, and rainfall is mainly distributed in winter (December – March). The growing season of vegetation occurs during the period February to April after rainfall.


Fig 1 Map of Syria

A model diagram of yearly supply of food for sheep is shown in fig. 2. The animals depend on the steppe for about 42% of the total supply. Recently, cultivation for crop fields has expanded in the northwest around the Maragha Nat. Range.


Fig 2 Model diagram of yearly supply of food for sheep

Methods :
Landsat data. Path 173, Row 35 TM
1989/3/21, 1989/4/22, 1990/3/8, 1990/7/14, 1991/4/28
Balloon Aerial Photo system
Photographs taken from a balloon at a height of 200m (fig. 3) were digitized by color image scanner, Epson GT4100.


Fig 3 Balloon aerial photograph system

Global Positioning system
The exact position of the area studies was determined using a Sony PYXIS G.P.S. which is a receiver of G.P.S. Satellite ultra high frequency signal (L band 1.57GHz). The accuracy of the measured position is about 35m on the ground.

Image processing and GIS software :
ILWIS (Integrated Land and Watershed Management Information System), produced by ITC (The Netherlands), PC-based geographic information system with the following main functions.

Remote Sensing (Visualization, Enhancement, Classification, Geographical Correction, Image Arithmetic)

Geographic information (Digitizing, Polyorganization, Vector to Raster Conversion)

Internal database and interface (Table Manipulation, Database operation).

Results:

1. Interpretation of satellite imageries

1-1. General Survey of land – use in North Syria
Landsat 5 TM data of 1990/3/8 were combined to infrared false color imageries which were useful for the interpretation of the land use and vegetational conditions. Compared with old land use data which were interpreted by Landsat MSS in 1975, many crop fields were found to encroach upon grazing lands (Photo 1)


Photo 1 Landsat 5 TM 1990. 3. 8. North Syria

1-2 Analysis of shrub plantations
Range lands are being improved by the plantation of fodder shrubs, for example, Attriplex harimus, A. Canescens and Salsola vermiculata. We can analyze the plantation areas based in enlarged false color imageries in the following combinations; 4 band = red, 3 band = green and 2 band = blue. Shrub areas give a dark color, and can be defined clearly in the Adamy and Maragha areas. Plantation map, scale 1:60,000, was constructed from Landsat imagery.

2. Analysis of large aerial photographs

2.1 Species and density of planted shrubs
Balloon photo shows the difference in species. Canopy of Atriplex halimus is larger than that of Salsola vermiculata. Plant coverage percentage was estimated at 20.4% for A. halimus and 4.9% for A. Vermiculata using aerial photos (Photo 2).

2.2 Vegetational patterns natural grassland.
Photo 3 shows the vegetational pattern on the natural plot 1. Dark color indicates the area with dense vegetation and whitish color those with poor vegetation, due to the micro topography where hills appear whitish and valleys dark. (Photo 3).

2.3 Line intercept observation of depressed areas
In the northern part of plot 3, there is a round depressed area, with a diameter of about 90m. The details of the vegetation were obtained from the ground survey. The dominant species were Peganum harmala, Hordeum bulbosum, Carex sp. And Plantago lagoplus. Hordeum predominated on hill sides and Peganum, Carex and plantago predominated indepressed areas.


Photo 2 Maragha Natural Plot 1


Photo 3 Improved Plot 11

Discussion
Grazing intensity should be controlled according to the grazing capacity, which can be estimated based on near infra-red color imageries. Other factors involved include soil type, topography and climate. After acquisition of these data, a bio-mass map will be processed to obtain a grazing capacity map.

Acknowledgements
This study was carried out at the International Center for Agricultural Research in the Dry Areas (ICARDA, Aleppo, Syria) as a collaborative research project. The author is greatly indebted to Dr. A.E Osman (Pasture, Forage and Livestock program) for providing the test site for grassland trial and ground truth data. Assistance of Mr. Fahim Ghassaly (P.F.L.P) for the operation of the balloon system and identification of plant species in the test field is gratefully acknowledged.