全球变化知识资源中心

Despite the strong societal impact of natural hazards, their documentation remains incomplete, with only a few inventories exceeding the past two centuries. Surprisingly enough, this also applies to Europe, a densely populated territory, and to floods, which along with storms are the most common and damage-causing natural hazard in Europe. In addition, existing inventories have often been compiled by scientists and technicians and are used for risk management in a top-down manner, although the participation of all parties concerned has been recognized as a key factor for disaster reduction. To address this double paradox, the present article presents the regional flood risk observatory ORRION for the Alsatian region, north-eastern France, and its very rich data content. Stemming from two successive interdisciplinary and transnational French-German research projects, ORRION was designed as a participative online platform on which information is shared between individuals, stakeholders, engineers, and scientists. This original approach aims at maximizing knowledge capitalization and contributes to building a common knowledge base for flood risk. ORRION is organized by events including all river floods that have likely arisen from a single synoptic situation. For each event, it documents information sources, date of occurrence, causes, and consequences in terms of damage and affected river basins and municipalities. ORRION has contributed toward renewing our knowledge of flood hazard and risk in the target area. Notably, here, long chronicles of floods are derived for 13 rivers, the Rhine and most of its main Alsatian tributaries and for all Alsatian municipalities, most of them since the end of the 15th century but over more than one millennium for the Rhine. Their main characteristics according to various typologies (seasonality, causes, severity, etc.) are analysed. Major developments over the study period related to sources, land use, and/or climate change are identified. The advantages and limitations of the approach are discussed, and the potential to expand both data exploitation and build common flood risk knowledge is outlined.