Automation of hydrological drought typology to study drought propagation
in a tropical catchment
Understanding how droughts propagate through the hydrological cycle from
precipitation to streamflow and groundwater is important for improving
water and risk management policies. At the catchment scale, the analysis
of drought propagation and classification into drought types is usually
done manually, which can be time consuming and difficult to replicate.
Here, we developed an automated, objective procedure for classification
of different drought types with the aim to study drought propagation in
the tropics. The method was applied to the Savegre catchment in Costa
Rica as a proof-of-concept. We first confirmed that drought events in
the catchment could be classified into the process-based typology from
the literature: classical rainfall deficit drought, wet-to-dry season
drought, and composite drought. The automation algorithm was able to
replicate the classification obtained with the manual typology with the
exception of two events, and thus it is a development towards objective
and time efficient hydrological drought analysis in tropical catchments.
Most of the detected hydrological droughts (80% and 76% of all river
discharge and baseflow droughts, respectively) were classical rainfall
deficit droughts, which suggests that climate plays a more important
role in drought development than catchment characteristics in this
catchment. However, the importance of catchment characteristics was
revealed by the presence of severe composite drought events and by the
attenuation of significant precipitation droughts.