Growing demands for sustainable management of tropical forests have resulted in the acquisition of ArcForest, a forest management decision support product developed by Esri-Canada, by the Sabah state in Malaysia and by Ecuador. The application of the ArcForest product to tropical forest management has required adjustments to the forest inventory and planning modules to reflect tropical forest data, planning activities, and practices. Based on the Sabah and Ecuador experiences, the purpose of this paper is to present the conceptual Tropical ArcForest shell that will enable the ArcForest product to support tropical forest management globally.
Managers of scenic landscapes have employed a range of simulation techniques to graphically illustrate planning decisions that might cause changes to a region's scenic quality. New and expanded computer techniques have assisted this effort, supplying image processing tools that articulate scenic change through the manipulation of scanned photographs. Typical strategies merge ground photographs with non-graphic data extracted from plan view maps, charts, and lists. Combining this diverse information has proved challenging as a mechanism to reference ground view/plan view information is required. The research presented here explored this merging to material, utilizing an ArcInfo data structure, global positioning (GPS) technology, and a digital elevation model (DEM), to produce a series of photographic-quality simulations of a forested landscape undergoing change. A multi-year photographic survey of a coniferous landscape in southern Utah was initially conducted using GPS referencing to coordinate all camera/target positions. Within this view shed, 5000 individual trees were surveyed to record their locations, the species, their height, trunk diameter, age and their estimated ground elevation. Using the TIN module within ArcInfo, a surface model was generated using (1) contour coverage for surface delineation outside the study area, and (2) ground-height attributions from the 5000 surveyed trees for surface delineation inside the study area. Next, utilizing the species attribution, a specific tree type (Englemann spruce) was isolated to create an additional 3D surface. This surface showed these selected trees extended to their estimated heights, using that attribution. These models, teamed with the GPS camera positions, were then used to create a series of perspective wireframes, and camera/target relationships equal to those of the site photographs. The scanned photos, which were subsequently processed in ERDAS to simulate color changes indicative of a beetle infestation, were then merged with these perspectives, using the GPS registrations to correctly position the wireframes with the imagery. Using these composite images, the photographic positions (the elevated portions of the DEM) where the desired simulations would be developed were identified. This section of trees was extracted from the photographs, using Adobe Photoshop, and then repositioned in other unedited photographs, using Adobe Photoshop, and then repositioned in other, unedited photographs. In this way, color/change relationships were isolated and repositioned in other imagery, using the GIS/DEM structure to transfer the attribute information with a high degree of positional accuracy.
The use of Geographical Information Systems has been spreading faster in always more fields. People not accustomed to dealing with geographical information come in touch with data processed with a GIS. Many of them are decision-makers who rely on GIS operators to make their choices. Mainly because of the lack of the third dimension in map sheets, they often find difficulty in understanding data. A 3D physical model helps displaying data, but is complex to build, it is cumbersome and allows displaying only scarce information. Instead a 3D numerical model is much easier to deal with, mainly because relief information is already available within the GIS itself and the Digital Terrain Model may be used to display all necessary data. Each point of the model already has its elevation datum associated, in the format (closely) needed to build the 3D numerical model. These considerations are at the basis of a graduation theses in Terrain Imaging at the Faculty of Architecture of the University of Florence. A study on Elba Island, 222 sq. km (86 sq. mi.) in the Tyrrhenian Sea, 9.7 km (6 mi.) W of the Italian peninsula, was used to develop an example of how data may be displayed by the means of a 3D digital model. Elba was chosen because: