While the occurrence of floods is sometimes thought of as a stationaryrandom process, periodic climatic processes may lead to long-term variability of floods over decades and centuries. The relative role of the synoptic atmospheric situation, catchment conditions and river system processes are difficult to unravel here. These attribution studies may be affected by observation biases and it may be difficult to account for the non-linear interactions of drivers. To make progress in flood change research there is a need to go beyond the existing practice of trend analyses and formally attribute observed flood changes to their drivers and anticipate future changes in a process based way.
Key science questions
- What were the long-term dynamics of river floods in the past?
- What were their causes and, in particular, the relative role of atmospheric, catchment and river processes?
- What is the role of feedbacks between anthropogenic activities and processes in the atmospheric, catchment and river systems?
- What is the sensitivity of river floods to changes in their controls and how will river floods evolve in the future?