Understanding the Role of Climate Characteristics in Drought Propagation

Abstract

In this study, we use numerical experiments with a simple water balance model to understand the roles of key climate characteristics in hydrologic drought propagation and the consequence of human responses to drought events under different climates. The experiments use climate inputs from a range of places with a hypothetical catchment of fixed properties to study drought propagation under different climates. Three drought propagation mechanisms are identified that produce hydrologic droughts with differing characteristics. The first mechanism involves seasonal groundwater recharge cycles, which persist during low rainfall periods, resulting in shorter hydrologic droughts compared to meteorological droughts. The second is characterized by seasonal groundwater recharge cycles that are suppressed during low rainfall periods, resulting in longer hydrologic droughts than meteorological droughts. The third is exemplified by a lack of seasonality in groundwater recharge and a strong control of precipitation over groundwater recharge, resulting in hydrologic droughts of similar duration as meteorological droughts. The roles of seasonality, climate aridity, and timing of precipitation in producing these different drought propagation mechanisms are studied. The timing of precipitation is found to have the most significant impact. Furthermore, modeling experiments are performed to understand the role of climate in the interaction between short and long time‐scale human activities in response to droughts and the effect of the common practice of groundwater pumping during drought events on long‐term groundwater depletion. Interestingly, climates with high interannual variability of precipitation are found to be associated with less groundwater depletion than the climates with low interannual variability.

A 250‐Year European Drought Inventory Derived From Ensemble Hydrologic Modeling

“…Precipitation characteristics, especially for recent droughts (after 1990), can be related to these differences between precipitation drought rank class and runoff or soil moisture drought rank class within a given year. Features such as the number of wet days, seasonality and the fraction of heavy versus light rainfall are known to influence drought propagation (Apurv et al, ; Zolina et al, ) and characteristics (Van Loon et al, ). Light and moderate rainfall soaks into the soil and potentially feeds the groundwater tables, whereas heavy, short‐term rainfall often cannot fully infiltrate the soil column and generally leaves the system quickly through overland flows or (near‐)surface runoff.…”

Analysis of Drought Progression Physiognomies in South Africa

“…Two clusters are concentrated in the Eastern parts of South Africa (comprised of the Limpopo, Mpumalanga, North-West, Gauteng and KZN provinces), one cluster is mainly concentrated in central interior (consisting of the Free State, Northern Cape and some parts of North-West Province) and Northwestern parts of South Africa, while two clusters dominate in the Western and coastal areas of South Africa (mainly in the Western Cape and Eastern Cape provinces). Overall, the resulting spatial drought transition zones subtly correspond to the broader climatic zones of South Africa, i.e., the Savanna, Grassland, Karoo, Fynbos, Forest and Desert zones [29].…”

Hydrological Drought Simulations: How Climate and Model Structure Control Parameter Sensitivity

“…In their study on drought propagation, Apurv et al (, p.3) state that “Simple conceptual models contain fewer numbers of parameters, which makes it easier to track the influence of each hydrologic process on the final outcome.” Although this statement in itself is true, the results of this study imply that no process‐based conclusion can be drawn concerning hydrological drought based upon a single model. We showed that parameters representing comparable mechanisms (or processes) in different models had differing sensitivity to hydrological drought indicators; that is, it differs per model which parameters, representing different processes, are triggered to simulate hydrological drought.…”

“…In our adopted definition, hydrological drought is a negative deviation from “normal” flow conditions. There are two frequently used methods to identify hydrological drought; the fixed threshold method (e.g., Apurv et al, ; Laaha et al, ) and the variable threshold method (e.g., Sheffield et al, ; Van Huijgevoort et al, ; Yevjevich, ). Both methods are based on a threshold level below which a hydrological drought is defined.…”