An Internet Atlas Prototype of the Sri Lankan Central Province

Marco Sieber
Department of Geography, GIS Division, University of Zürich
Winterthurerstr. 190, 8057 Zürich, Switzerland
Phone: +411 635 51 91; E-mail: msieber@access.unizh.ch


Keywords: Web mapping, example of client-side SVG atlas, Sri Lanka.

Background

In contrast to other Provinces in Sri Lanka, digital spatial and statistical data on the lowest administrative unit exist for the whole Central Province within the so-called Regional Information System (RIS). The statistical data comprises a wide range of socio-economic information and data on private sector enterprises, which are stored within a Microsoft Access database. The spatial data set, digitised in MapInfo data format, includes layers of all administrative units, the transportation infrastructure, and the water resources. So far, RIS is unique in Sri Lanka.

The Enterprise Information Project (EIP) is one of more than 40 projects of the German Agency for Technical Cooperation (GTZ) currently operating in Sri Lanka. The objective of the EIP is to establish an information system used by private enterprises that help overcoming the lack of business related information. Among different information services GIS based Business Mapping Services (BMS) were introduced to its partner organisations using the RIS data sets.

Objectives

The EIP felt the need to establish an Internet atlas of the Central Province in order to overcome current limitations and to create a broader awareness of the existing RIS data sets as well as resulting possibilities for planning and analysing tasks using GIS.

This project was started in early 2000 as a joint-initiative between Geoconcept , the Department of Geography (University of Zürich), and the EIP. The objective was to implement a first prototype that accomplishes the following requirements. The prototype should:

The intended audience for the prototype are interested laypersons as well as experts (private sector entrepreneurs, corporate clients, international projects, policy makers, etc.) and employees of EIP partner organisations.

Implementation

Already in the early stage of the project, after the evaluation of current Web mapping technologies, it became evident that a client-side realisation - using SVG, HTML, JavaScript, DOM and CSS - satisfies the requirements best.

The SVG base maps were generated with self programmed Perl scripts and embedded within the HTML based GUI. The statistical data were transformed into a JavaScript array structure. All interactions and functionalities within the prototype were programmed using JavaScript and the DOM.

The current prototype is designed for a screen resolution of 1024 * 768 pixels and the use of Internet Explorer (5.0 and higher) together with the Adobe SVG Viewer (version 2 and 3), running on Windows operating systems. Other browsers are not supported so far.

Link to the current prototype

Fig. 1: Link to the current prototype

Results

The current prototype contains more than 60 socio-economic variables subdivided into four subject groups (population, housing, health & education and land use). Each variable can be chosen via an easy-to-use thematic content navigation system. After a variable is selected, it can be drawn as thematic map (choropleth or circle diagrams) by one mouse click. The map legend is adjusted automatically each time a new map is drawn. Choropleth maps can be tailored to the user's preferences by choosing different administrative unit levels, type of classification method, colours, and number of classes.

Zooming and panning is supported via the overview map, which is linked to the main map in order to keep control over the current position. The main map consists of eight layers that can be switched on and off. Mouse events have been added on the lowest administrative unit. By moving the mouse over such a region, the names of all appropriate administrative units and the current value of the selected theme are displayed on an info panel. By clicking on one of these totally 2162 regions, a detailed data profile is generated, containing a list of the values of all socio-economic variables.

First evaluation tests of the current prototype will soon be launched by the EIP. It can be anticipated that further improvements will be implemented after having received feedback from the target audience.



15.01.2002