The architecture of the SITUMET-system consists of four components

a) data server,

b) situation finder,

c) situation processor, and

d) product platform.

Additionally, external components, such as forecast systems, situation provider and terminal devices of the users, are integrated.

General concept of SITUMET-system. The main information fluxes follow from top to down and from left to right, thus from the forecast systems through the data server to the technical devices of the end users.

1 Data server

The data server is the basis of the SITUMET weather services. It processes the results of the prepended forecast systems and provides it in form of high resolution raster data. An access control layer inside the data server controls the data access of external applications.

2 Situation finder

The situation finder processes raster data of the data server to location-based and time-related situations. Situations are status descriptions for multidimensional intervals in time and space. This is comparable with vectorisation methods, which find arcs and nodes in images (e.g. photographs). That means, connected temporal and spatial areas with the same conditions (e.g. temperature > 24°C) can be recognised from a bunch of point information (e.g. hourly temperature in Austria). Due to the temporal dimension it is possible to track changes of the location or extension of areas with specific temperature values.

3 Situation processor

Several conditions or state descriptions can be used for the definition and recognition of situations. That way, for instance temperature based situations (cold, warm, hot, ... by indicating a threshold values), or precipitation based situations (no precipitation, rain, heavy rain, or hail) can be defined. Even situations that are only dependent on a position (e.g. field, forest, meadow, town) are possible. By combination of such situations, that means by recognising spatial and temporal overlapping, new situations can be derived. This is the task of the situation processor. It combines weather based situations with information coming from other sources (situation provider), e.g. GIS and personal user profiles.

An example is the recognition and modelling of forest fire risk. Regions with the following conditions (situations) have an increased forest fire risk: „current evapotranspiration < 0,8 * potential evapotranspiration“, „rain = 0“and „area type = forest“.

4 Product platform

The product platform is a framework, which helps to built weather based information products, such as weather online portals or weather based subscription services. The needed flexibility and extensibility is obtained by a modular concept. For instance, several presentation and allocation modules care for the preparation of the information (e.g. in form of diagrams, texts, images or videos) respectively for the publication or submission of information. In this process the data, the products as well as the supply can be adapted to different user groups and information channels. A personalisation for individual users is possible, too. Via a configuration tool the products can be simply and flexibly composed according the requirements of the particular service provider and target group. The raster data, delivered by the data server, and the situation, delivered by the situation processor, provide the data basis for the product platform.
Depending on the aim of the particular weather product different terminal devices can be served. The spectrum ranges from the PC via mobile telephones and eBooks to telematic systems (e.g. telematic blind control in a house or automated irrigation system).