The European Commission (EC) initiated a special RIBAMOD expert meeting on 18 May 1998 for scientists and policy-makers from the three countries affected, together with additional experts from various EU member states; in all, over 30 specialist attended the event. The meeting was jointly organised by the EC, the Potsdam Institute for Climate Impact Research, the Polish Academy of Science, and the Czech Hydrometeorological Institute. A total of 20 papers were presented and discussed at the expert meeting and the contributions will be published by the EC (see below). These proceedings will cover four topics: Overview, Hydro-Meteorological Conditions, Hydrologic and Hydraulic Situation, and Management and Policy Issues. Pre-prints are available from PIK (PIK-Report No. 48).

Conclusions
The discussion at the expert meeting of experiences of the Oder flood indicates that the following protection strategies could be of benefit.

Modification of flood propagation

• A policy of "keeping the water where it falls" should be an important aspect of flood risk management. Watershed management and environmental conservation practice in flood risk areas includes appropriate land-use planning in order to store floodwater and to control the inundated areas. The dead forest in the headwaters of the Odra was considered to produce faster runoff than before and thus higher peak water levels.
• River engineering measures (reinforcement of levees, dikes, dams, reservoirs etc.) can modify the depth, volume, and velocity of the flood. The influence of linear infrastructure, such as roads, railways or pipelines and the discharge capacity of bridges needs to be examined.
• Effects of reservoirs in upstream areas can be important. Flood storage in the Czech Republic was shown to be effective in reducing the downstream peak flow of the first Odra flood wave.

• Effective flood warning is essential to preserve human life. This requires a robust warning system, public awareness of flood risk areas and preparedness for emergency evacuation.
• Flood forecasting and operation centres should coordinate forecasting, warning, dissemination and evacuation and thus reduce the high economic and social losses in a major flood.

• Building public awareness is one step towards a culture of living with risk.
• Flood insurance should compensate for flood losses and guarantee finance for post-flood recovery.
• However, natural disasters are unavoidable and in some extreme cases the overall community interest may be best served by the resettlement away from certain areas.

Research needs
The discussions raised the following research needs for improving flood protection and flood management systems, and these coincide with the several of those identified in previous RIBAMOD meetings.

• A holistic view of floods: inter-disciplinary across the hydrological, meteorological and human dimensions of the overall problem of flooding.
• Development of advanced forecasting methods (including real-time operation, now-casting) with international collaboration and exchange of data across borders.
• Creation of hydro-meteorological databases with long-term and homogeneous data records.
• Design of the hydrometeorological data network with sufficient redundancy to ensure the required accuracy and the security of information for forecasting in the most severe conditions.
• Integration of data of diverse types, accuracy and source to determine the state of the river catchment and the flood defence system.
• Understanding the effects of land-use and land-cover on flood formation, including reliable models for the quantification of water retention of different vegetation types, soils, landscape structures, agricultural practice etc. How do deforested areas and dead forest influence flood generation?
• Investigation of the impact of possible changes in the climate of Central Europe on flood generation in the Odra basin. Is the frequency of low-pressure systems changing in Central Europe with consequent changes in precipitation?
• Development of an integrated hydraulic model of flood generation and propagation in the river basin and a catchment-wide hydrological model. Combination with visualisation components for all areas and development of GIS-based software, visualising flood levels for supporting decisions in real-time operation.
• Creation of flood hazard maps.
• Quantitative precipitation forecasting, in particular for small and mountainous catchments.
• Risk assessment and management.

Proceedings of the RIBAMOD Events

©  December 1998