The emergence of large-scale hydrologic datasets, data analytic techniques, and mechanistic hydrologic models has supported the advancement of flood risk analysis. Traditionally, the study of hydrologic extremes has centered on observations made within the stream channel. Recent research has highlighted the importance of understanding the aspects of the atmosphere, land surface, and river network that impart their unique fingerprint on floods. In this dissertation I first review the critical aspects of sub-daily precipitation in the context of a stochastic weather generator. Next, we consider the direction of influence of the land surface on hydrologic extremes in conjunction with changes to climatic forcing. Finally, we propose a framework for flood risk analysis from the perspective of flood inducing meteorological events.