Use of a Smartphone for Collecting Data on River Discharge and Communication of Flood Risk.

Abstract:

Although many developed countries have well-established systems for river monitoring and flood early warning systems, the population affected in developing countries by flood events is unsettled. Even more, future climate development is likely to increase the intensity and frequency of extreme weather events and therefore bigger impacts on the population can be expected.

There are different types of flood forecasting systems, some are based on hydrologic models fed with rainfall predictions and observed river levels. Flood hazard maps are also used to increase preparedness in case of an extreme event, however these maps are static since they do not incorporate daily changing conditions on river stages. However, and especially in developing countries, data on river stages are scarce. Some of the reasons are that traditional fixed monitoring systems do not scale in terms of costs, repair is difficult as well as operation and maintenance, in addition vandalism poses additional challenges. Therefore there is a need of cheaper and easy-to-use systems for collecting information on river stage and discharge. We have developed a mobile device application for determining the water stage and discharge of open-channels (e.g. rivers, artificial channels, irrigation furrows). Via image processing the water level and surface velocity are measured, combining this information with priori knowledge on the channel geometry the discharge is estimated.

River stage and discharge measurement via smart phones provides a non-intrusive, accurate and cost-effective monitoring method. No permanent installations, which can be flooded away, are needed. The only requirement is that the field of view contains two reference markers with known scale and with known position relative to the channel geometry, therefore operation and maintenance costs are very low. The other advantage of using smartphones, is that the data collected can be immediately sent via SMS to a central database. This information can be easily gathered for its use within models and redistributed, using the same channels, among interested stakeholders and the community.