This project addresses three technical issues, to be resolved through three major tasks:

1. To establish a regional data warehouse for natural habitat data, current Web technology can provide the mechanisms for data exchange and download by File Transfer Protocol (FTP) or HyperText Transfer Protocol (HTTP). The design of a data warehouse will focus on the implementation of Web-based user interface and methods to organize data sets in this data warehouse. For example, natural habitat data sets can be organized by theme-based layers or location-based tilts. Different types of data schemes can be applied in the design of the data warehouse to facilitate the sharing and exchange of geo-spatial data sets and remote sensing images.

2. To establish an Internet map server, this project will adopt a leading GIS package, ArcIMS, developed by ESRI, currently the leading GIS vendor. Advantages of adopting ArcIMS are its compatibility of with GIS/Remote Sensing data sets and its comprehensive mapping functions. ArcIMS can access ESRI shapefiles directly or ARC/INFO coverages via ArcSDE (Spatial Database Engine). Currently, most GIS datasets for our prototype application are in coverage or shapefile format. Also, ArcIMS can provide comprehensive functions for on-line mapping and customizable tools for adding new GIS functions.

3. To develop on-line analytic functions and software, there are two possible approaches: server-side applications or client-side applications. The server-side application will access data directly from the GIS or remote sensing database. The client-side application will access data remotely via a Web browser, or access the local data stored in a Web cache. Available program languages on the client side (Web browser) include Java or Active X controls. A Java platform is preferred because of its interoperability with different client environments (such as Windows 2000 or UNIX). The server side programming language might be more flexible, depending on the configuration of server platforms. With a UNIX server, the programming language could be C++ or Java, while with a Windows platform, the language could be Visual Basic or Java. One consideration in choosing the development platform is that the language should provide database connectivity and network communication to access GIS databases and remote sensing images.