Effective work flows in SMHI water-information production.


Hydrology at SMHI consists of five activity areas which correspond to components in a production chain (Figure 1).

This chain is used to deliver decision support to authorities and companies considering, for instance, warnings (water level, flow, fire risk, droughts), water power (security, regulations), nutrient flow, water status, environmental monitoring,  dimensioning of infra-structure (Slussen, Västlänken), flood risk (cities, watercourses), irrigation demand, climate effects and mitigation.

In addition, the general public receives information on the water cycle, as warnings and for recreational purposes.

Work flow - production chain at SMHI
Figure 1. Components in the hydrological production chain at SMHI. Enlarge Image

The results are distributed as interactive web pages, reports, software, files or scientific articles dependent on the receiver. Most parts of the operational production is automated, such as storage of real time data and computation of prognosis. Computations are made on office PCs, SMHI servers, at the super computer center (NSC) or in the cloud. The products are maps and time series, usually based on a model setup for a specific area.

Daily hydrological prognoses are produced for the full area of Sweden and Pan Europe. In addition, there are large scale model applications for the Arctic area, India, the La Plata river, the Niger river and The Middle East/North Africa together with customer specific model applications for the power industries and municipalities.

Every model application requires input data from a large number of data bases that are preprocessed and adjusted for the computations. The combination of observations and model calculations result in cost efficient decision support. The hydrological activities are dependent on input from meteorological observations and models.

The hydrological production chain is constantly developed based on input from external users.

Scientific questions

  • How can we ensure quality of the data management through the production chain?
  • A computed result is always dependent on the underlying input data and choices in the hydrological modelling. How can we increase transparency in our computation chains to make these dependencies visible?
  • How does the user perceive the presented information? How can we increase the user’s understanding on how the information can be used in decision making or for research purposes?


  • To clarify the modelling process by making results, software and tools openly available and usable for different target groups.
  • Quality assurance of the processes in the production chain through the development from research prototype to production.
  • Development of effective methods and tools for transparency and quality assurance of the dataflow through the production chain.
  • Methods for presentation of results to increase the understanding of the end users. The focus for this work is interactive web applications to allow the users to explore the results.