Author(s): Alex Crespillo Lopez; S.M. Vicente Serrano; L. Gimeno
Linked Author(s):
Keywords: Big data hydrology; Drought propagation; Event-based analysis; Hydroinformatics; Hydrological drought
Abstract: Hydrological droughts are dynamic phenomena that evolve across space and time, propagating through river networks under the combined influence of climatic forcing, catchment characteristics, and human regulation (Van Loon, 2015)5. Despite significant advances in drought research, most existing approaches rely on aggregated indicators, statistical relationships, or large-scale metrics, limiting their ability to explicitly characterise how individual drought events propagate within connected river systems. In particular, there remains a lack of methodologies capable of reconstructing the spatiotemporal evolution of drought signals at the event scale while explicitly accounting for river network connectivity (Bruno et al., 2026)2. This study presents a novel event-based framework to analyse hydrological drought propagation along river networks. The methodology identifies drought events independently at each gauging station using the Standardised Streamflow Index (SSI) and reconstructs propagation processes by linking temporally consistent upstream and downstream events based on network topology. This approach enables the construction of drought propagation chains, defined as sequences of connected events that capture the transmission of hydrological deficits from headwaters to downstream locations. A set of propagation metrics is introduced to quantify the spatial extent, temporal dynamics, and magnitude of these chains, including propagation lag, propagation fraction, chain duration, and both standardised and volumetric severity. The framework is applied to daily streamflow data from 33 gauging stations in the Ebro River basin (Spain) over the period 1961–2020, a region characterised by strong hydroclimatic heterogeneity and significant human regulation (Batalla et al., 2004)1. The results reveal a wide diversity of propagation behaviours, ranging from rapid, basin-wide droughts affecting a large proportion of the network to highly localised events confined to individual sub-basins. Propagation timescales exhibit substantial variability, from short delays of a few days to long-lasting, multi-annual processes exceeding one year. These differences reflect the interaction between hydrological memory, storage processes, river network structure, and regulation effects.
Year: 2026