Application of Low Cost GIS and Remote Sensing for the Analysis and Evaluation of Regional Processe of Landscape Change in the Republic of Yemen [Archives:1998/46/Business & Economy]

November 16 1998


By: Abdulkhaleq A. Ahmed Al-Qusaily 
(MSc, Germany)
This study shows one way how to integrate remote sensing with the Geographic Information System (GIS) in the low cost technology level as a planning instrument to deliver reliable and accurate basic information for decision makers evaluating development strategy. This, in our country (Yemen), will help us attain greater autonomy in the planning process and in resource management.
The Republic of Yemen is located in the semi-arid and arid zone of southwestern Arabia and dominated by a rugged mountainous terrain.
This study shows several ways to integrate remote sensing data with GIS (Geographic Information System). However, the data preprocessing and the classification algorithm must be improved before operational application is feasible.
Banding effects and forward/reverse scanning differences should be corrected.
The purpose of this thesis is to survey the land-cover change in Radaa Region using digital satellite imagery.
Two Landsat. MSS scenes (From EROS Data Center in the U.S.A), recorded from 1973 to 1985, were evaluated by means of sophisticated digital image processing techniques.
In addition to the multispectral data, ancillary data such as topographical and geological maps were introduced into the classification process to improve the accuracy of the results.
Interpretations of landsat imagery using standardized criteria led to a quantitative determination of land-cover use change.
The integration of geological and topographic maps into a GIS and the subsequent interdisciplinary interpretation gives better insight into the inter-relationship of an ecosystem. This provides the basic knowledge which is necessary to simulate different scenarios depicting possible development in mountain ecosystems.
What is GIS?
GIS is an organized collection of computer hardware, software, geographic data and personnel designed to efficiently capture, store, update, manipulate, analyze and display all forms of geographically referenced information.
Certain complex spatial operations are possible with a GIS that would otherwise be difficult, time consuming, or impracticable to perform.
GISs are particularly important to government agencies and other private and public concerns dealing with land-related information. Through GIS, geographical information from paper maps, satellites, aerial photographs and land-related descriptive records are converted to digital form and fed into computers as overlays representing property parcels, political and man-made boundaries, utility distribution and transportation networks, land use patterns and an endless list of additional information describing the infrastructure and natural resources that support us.
Methods of Approach
The methodological approach of monitoring seasonal and annual land-cover changes on landsat imagery by using low-cost technology was tested in a region of approx. 1,600 located in and around Radaa.
After a description of the landsat system and the low-cost GIS IDRISI, the geometric and radiometric preprocessing of the landsat imagery is outlined. In a preprocessing step each landsat-MSS scene was geometrically corrected to correspond to the map projection used in the topographic maps (Survey Authority, Sanaa).
The multispectral data set of each scene was classified using unsupervised and post-classification techniques. The classification approach led to 26 classes subsequently merged into six final land-use and land-cover categories.
These show the spatial distribution and the change analysis for irrigated agricultural land, rain-fed land, cropland, wadi soils (alluvial sediments, sand, clay … etc.), sandstone and granite areas.
In the study area, the expansion of pump-irrigated lands mainly along the wadis and plains is easily understandable considering the availability of local ground water. Radaa plain, which is covered by alluvial deposits underlain by sandstones is the most important ground water reservoir in the region.
The agricultural setting of the study area is changing rapidly. The processes are mainly induced by ongoing development projects, governmental development policy and the opening up of the area with a dense network of roads. The main consequences are as follows:
– Increasing importation of cheap foreign grain puts the market prices of local products under pressure, making local grain production not feasible. This prevents market-oriented agricultural production and leads to enforced migration and partly to field abandonment.
– Along the roads (main asphalt road) the amount of pump-irrigated land is increasing.
– Good traffic connections between a city and its hinterland promote urbanization, reaching rural society and causing severe social changes in the ways of living, the variety and amount of commodities consumed and the rate of emigration, etc.
Most of the economic as well as the social developments in the rural areas are related to land-use and land-cover changes and have repercussions on the ecological system. These changes can be monitored regularly by the use of remote sensing and GIS techniques.
Application of an appropriate technology:
Remote sensing is the most efficient and economical method to take inventory of land-cover features and to monitor their changes in a short time by the use of multitemporal imagery. But remote sensing techniques have to be evaluated carefully with regard to their appropriateness to local conditions especially when implemented in a developing country.
Remote sensing methodology and its related technology have a modular characteristic that allows gradual development, starting if necessary at a low technological level with comparatively low costs, flexible and unsophisticated equipment and a small staff of trained personnel.
Generally three levels can be distinguished with increasing technological complexity:
1) Low technological level with simple photographic processing and visual-interpretation methods.
2) Intermediate technological level with mainly analogue interpretation methods but with more complex data handling and processing systems.
3) High technological level with automatic classification using digital and hybrid processing systems.
The three levels correspond directly with progressively increasing investments for equipment and needs for trained staff to operate and maintain it.
* This article is the summary of an MSc thesis submitted to Dresden University, Germany.