Floods of the Upper Danube River and Its Tributaries and Their Impact on Urban Economies (c. 1350–1600): The Examples of the Towns of Krems/Stein and Wels (Austria)

This paper examines the impact of disastrous and ‘ordinary’ floods on human societies in what is now Austria. The focus is on urban areas and their neighbourhoods. Examining institutional sources such as accounts of the bridge masters, charters, statutes and official petitions, it can be shown that city communities were well acquainted with this permanent risk: in fact, an office was established for the restoration of bridges and the maintenance of water defences and large depots for timber and water pipes ensured that the reconstruction of bridges and the system of water supply could start immediately after the floods had subsided. Carpenters and similar groups gained 10 to 20 per cent of their income from the repair of bridges and other flood damage. The construction of houses in endangered zones was adapted in order to survive the worst case experiences. Thus, we may describe those communities living along the central European rivers as ‘cultures of flood management’. This special knowledge vanished, however, from the mid-nineteenth century onwards, when river regulations gave the people a false feeling of security.

Combining Wireless Sensor Networks and Machine Learning for Flash Flood Nowcasting

This paper addresses an investigation with machine learning (ML) classification techniques to assist in the problem of flash flood now casting. We have been attempting to build a Wireless Sensor Network (WSN) to collect measurements from a river located in an urban area. The machine learning classification methods were investigated with the aim of allowing flash flood now casting, which in turn allows the WSN to give alerts to the local population. We have evaluated several types of ML taking account of the different now casting stages (i.e. Number of future time steps to forecast). We have also evaluated different data representation to be used as input of the ML techniques. The results show that different data representation can lead to results significantly better for different stages of now casting.