Extreme hydrological events disasters such as droughts, floods and storms lead to the most devastating natural in terms of casualties and economic losses. In the context of current global warming, there is a high uncertainty on the observed trends and projected changes in extremes at a global scale. Extreme events that occurred in the past play here an important role as they enable us to investigate the dynamics of extremes under natural climate variability beyond the instrumental period. The main goal of this session is to bring together scientist, scholar and engineers that explore the variability and controlling mechanisms of past hydrological extremes on decadal to millennial time-scales based on different historical and natural archives such as tree-rings, speleothems, lacustrine and marine sediments and ice cores. We also welcome contributions that integrate both, proxy data and climate modelling to understand the external and internal forcing controlling the hydrological cycle. We also invite contributions that explore new statistical modelling approaches aiming to quantitatively assess the climate drivers of the non-stationary behaviours of extreme events frequency and intensity.