The Integration of Remote Sensing System and GIS for Forest Landuse Planning
Dr. Suwit Ongsomwang
Forest Resources Assessment Division
Forest Research Office, Royal Forest Department
Bangkok 10900. Thailand
Abstract
In general, the applications of remote sensing to forestry in Thailand were mostly based on visual interpretation and the thematic data that were produced from the remote sensing data were still separated. The accuracy map acquired from the interpretation of aerial photographs at 1: 15,000 scale for forest and non-forest area in the past cannot be used as a base map for the interpretation of satellite imageries of scale 1: 250,000 at present. In fact, it leads to the inconsistency of the extraction data from Landsat imageries and also increases the time and cost for the ground truth survey. Actually, the integration of the thematic data with remote sensing data is limited because of using visual interpretation. method. The main object of this study is to develop reliable method for integration of remote sensing system and the GIS for forest land use planning.
The method was developed by using the multistage remote sensing concept and the integration of remote sensing system and the GIS. In this study, Landsat images were used as the first stage for existing forest area classification. In the secondary stage, aerial photographs at the scale of 1: 15,000 were used for forest land use classification. In the third "stage, ground verifications were conducted for collecting and generating the nieces information for forest land use planning. In addition, the geographic information databases that were designed for handling and processing spatial data of remote sensing and ancillary data were established under the exchangeable format of the image processing system and the GIS.
The main forest land types in the forest land use plan of the Ngao Demonstration Forest, Ngao District, Lampang Province are (I) non-forest area with no limiting conditions, (2) non-forest area with limiting conditions, (3) mining area, (4) community forestry area, (5) productive reforestation area, (6) productive forest area, (7) protective. reforestation area and (8) protective forest area.
The results from this study shows that the integration of remote sensing and the GIS appears to have potential for providing the necessary information for forest land planning, particularly in classification of existing forest land uses and their status. In addition, forest land use planning which was conducted by using the GIS shows the new methodology to the manager and planner for making a forest land use plan.
I. Introduction
The application of remote sensing data in forestry in Thailand was mostly based on visual interpretation and the thematic data that were produced from the remote sensing data were still separated. The accuracy map that would be interpreted for forest and non-forest area in the past by using aerial photographs of scale 1: 15,000, for example, cannot be used as a basic map for the intetrpretation of satellite imageries of scale 1: 250,000 at present. In fact, it leads to the inconsistency of the extraction data from Landsat imageries and also increases the time and cost for the ground truth survey. .Actually, the integration of the thematic data with remote sensing data is limited because of the visual intetrpretation. In addition an efficient geographic handling and processing system that transforms the thematic data into usable information does not exist.
At present the major tool for the geographic handling and processing system of the thematic data is the Geographic Information System (GIS). The GIS is a set of tools for collecting, storing, retrieving at will, transforming, and displaying spatial data from the real world for a particular purpose (Burrough 1986). The GIS evolved as a means of assembling and analyzing diverse data pertaining to specific geographic areas, using the spatial locations of the data as the basis for the information system (Shelton and Estes 1979). Kalensky (1992) stated that remote sensing and the GIS are closely related. They are complementary information technologies which are both needed and operationally used for natural resources mapping and inventories, natural disaster assessment and risk forecasting and for environmental monitoring. If they operated separately in institutes for management of natural resources, it would be harmful. It would result in costly duplication of various activities, training and some equipment and negatively affect the efficiency of project implementation. Goodenough (1988) also suggested that for the integration of remote sensing and the GIS, the remote sensing data must use the digital image analysis system. The main objects of this study are to develop reliable method for integration of remote sensing system and GIS for forest land use planning.
2. Study Area
Ngao Demonstration Forest area was chosen as the study area, The Ngao Demonstration Forest area extends over the headwaters of Ngao river in the north-west of Lampang province, Thailand down to its confluence with the Yom river on the Phrae province boundary, the boundaries being those of Ngao district, Most of the Ngao Demonstration Forest area that is in Ngao district, Lampang province lies between 18° and 21,5 north, 99° and 45 east The total area is ""; about 1,751,59 sq, kIn,
3. Methodology
The basic concept for the development study method is employed from the "multi II remote sensing concept which was , enunciated by Colwell (1975). Multistage remote sensing concept is used in this study. This concept encompassed the ;i: variety of remote sensing platforms and sensors that can be used in various combination, to acquire remote sensing data and also considered the processing, analysis and utilization methodologies available to practitioners in the field of remote sensing. In the multistage remote sensing approach, satellite data may be analyzed in conjunction with high and low altitude data, and ground observation. Each successive data source might provide more detailed information about
geographical areas. Information extracted at any lower level of observation may then be extrapolated to higher levels of observation (LiIlesand and Kiefer 1979). In addition, the geographic information databases that were designed for handling and processing spatial data of remote sensing and ancillary data were established under the exchangeable format of the image processing system and the GIS. McFarland (1982) stated that the establishment and maintenance of geographic databases in computer-readable form has great .potential for supplying resource management information.
To fulfill this methodology, a scheme of pract1cal work on this study 1S shown m Figure 1
Figure 1 Schematic flowchart of the main tsks in this study
