Local early-warning system grabs the attention of the UN, Nasa.
A locally-developed early-warning system is being reviewed by GDIN, Nasa, in the US, and the United Nations (UN), in order to create a global atlas making use of this home-grown technology. It is said that the early-warning system is one of the most advanced in the world. Developer Dusan Sakulski will be demonstrating the system to the UN in Geneva, Switzerland, this month, after having just returned from Nasa. Sakulski works at the South African National Disaster Management Centre (SANDMC), which provides worldwide users with a single, integrated view of all potential natural disaster hazards, vulnerability and risks, thanks in large part to the computational ability of Mathematica. The centre, housed in the old Y2K management centre at the National Department of Provincial and Local Government, in Pretoria, draws together a range of natural phenomena data and publishes these to the Internet. Using webMathematica publishing tools, the centre developed a system which is able to provide on-the-fly analysis of the data directly to the web-interface which is precisely what drew the UN's and Nasa's interest.

Mathematica is a numeric and symbolic programming language that gives users cross-sector computational abilities and is used by users from engineering and science, through to design and education. WebMathematica is an integration technology that allows Mathematica users to publish customised applications and functions to the Internet using Java servlets, called mathlets. WebMathematica is built around open standards, allowing for the integration of Mathematica applications into existing websites where users can gain access using any standard Java-capable browser. Heading up the SANDMC project is system integrator and coordinator Sakulski. He has built up the monitoring systems of the centre over a number of years and today operates what is probably one of the most information-technology-integrated disaster management centres in the world. It is not that we manage more data than any other centres, but we offer more tightly integrated tools to manage potential hazards, vulnerability and risks everything from floods and droughts to offshore storms, says Sakulski. He says the SANDMC plays an important role in early warning, and in assisting government and civil society organisations to plan and prepare for possible natural disasters. The centre is not here to replace individual services such as the South African Weather Service, but rather to play an overarching role in drawing together a range of information and drawing structure from the data that may be of use to other users. Mother Nature is a very complex thing to understand and it takes enormous calculation power to begin to understand her. With webMathematica the task is greatly simplified, he adds. Sakulski offers an example, saying that some of the drought analysis available on the site, for example, could take up to an hour to do by hand. With Mathematica the same analysis can be done in a few seconds and, because it is available via the web, anyone can access these results or run their own computations, he explains. The system setup also allows users to build simple indicators of complex problems, making the information available to a much wider audience. To produce this level of information and analysis you need a complex and powerful mathematical language, and Mathematica is exactly that, continues Sakulski.

SANDMC accesses data from around the country and internationally, including analysis and data on rainfall patterns and catchment areas, radar images, as well as monitoring offshore storm conditions for possible storms that are likely to hit the South African coast. The centre is also the Southern African distributor of the US's Nasa satellite images. There has been a perception that Mathematica is simply a tool for mathematicians. This could not be further from the truth, says Clemens Dempers of Blue Stallion Technologies, South African reseller of the Mathematica range. The centre's most powerful tool is in all probability its ability to publish its collated data to the Internet, making it available to all agencies and people that require the information. Using webMathematica, Sakulski is able to publish the analysis functions he has developed directly to the Internet. So, for example, given the rainfall data collected over the last 80 years, the SANDMC site is able to calculate and graph patterns of rainfall by individual catchment area. Users are then able to input their own requirements, and webMathematica takes care to prepare the data in a usable and manageable form. No-one else in the world has been able to perform on-the-fly live data analysis when it comes to disaster management.

With Mathematica and webMathematica we are the first to be able to do this, says Sakulski.