The propagation from meteorological to hydrological drought and its potential influence factors

abstract

It is important to investigate the propagation from meteorological to hydrological drought and its potential influence factors, which helps to reveal drought propagation process, thereby being helpful for drought mitigation. In this study, Standardized Precipitation Index (SPI) and Standardized Streamflow Index (SSI) were adopted to characterize meteorological and hydrological droughts, respectively. The propagation time from meteorological to hydrological drought was investigated. The cross wavelet analysis was utilized to examine the correlations between hydrological and meteorological droughts in the Wei River Basin (WRB), a typical arid and semi-arid region in China. Moreover, the potential influence factors on the propagation were explored from the perspectives of large-scale atmospheric circulation anomaly and underlying surface characteristics. Results indicated: (1) the propagation time from meteorological to hydrological drought has noticeably seasonal characteristics, that in spring and summer is short, whilst that in autumn and winter is long; (2) hydrological and meteorological droughts are primarily characterized by statistically positive linkages on both long and short time scales; (3) El Niño Southern Oscillation (ENSO) and Arctic Oscillation (AO) are strongly correlated with actual evaporation, thus strongly impacting the propagation time from meteorological to hydrological drought. Additionally, the propagation time has roughly positive associations with the parameter w of the Fu’s equation from the Budyko framework.

本研究采用了标准化降水指数（SPI）和标准化流量指数（SSI）来表征气象和水文干旱，以调查气象干旱向水文干旱的传播时间。采用交叉小波分析方法检验了中国典型干旱半干旱地区——渭河流域（WRB）水文和气象干旱之间的相关性。此外，从大气环流异常和地表特征两个角度探讨了传播过程的潜在影响因素。结果表明：（1）气象干旱向水文干旱的传播时间具有明显的季节特征，春夏季较短，而秋冬季较长；（2）水文和气象干旱在长期和短期时间尺度上主要表现为统计上的正相关；（3）厄尔尼诺-南方涛动（ENSO）和北极涛动（AO）与实际蒸发强相关，从而对气象干旱向水文干旱的传播时间产生了明显影响。此外，传播时间与Budyko框架中Fu方程的参数w呈大致正相关关系。

1. Introduction

Drought is a kind of extreme natural disasters resulting from abnormal decreases in precipitation (Oladipo, 1985; McKee et al, 1993; Wilhite et al 2000; Huang and Chou, 2008; Wang et al, 2011a; Huang et al, 2014a, 2015a,b), which can give rise to a large number of losses in the fields of economy, ecology, and environment (e.g. crop losses, degradation and desertification, urban water supply shortages, forest fires, etc.) (Flannigan and Harrington, 1988; Nicholson et al, 1998; Austin et al 1998; De Gaetano, 1999; Evans and Geerken, 2004). Compared to other kinds of natural hazards, the spatial extent of drought is extremely larger and its influencing time is commonly much longer. Thus, the damages caused by drought are expected to be highly larger than other natural hazards (Mishra and Singh, 2010; Xu et al, 2014).

Over the past century, the global climate and environment have witnessed remarkable changes, in which global warming is one of the most striking characteristics, which leads to accelerating the rate of water circulation, thereby resulting in highly frequent extreme events such as droughts and floods at the global scale (Willems, 2000; Kunkel, 2003; Roy and Balling, 2004; Beniston and Stephenson, 2004; Christensen and Christensen, 2004; Leng et al, 2015a,b). The frequency of drought tends to increase under the context of global warming. Hence, many scientists made a lot of attempts to investigate drought, which mainly consist of spatial and temporal differences of drought (Lana et al, 2001), the mitigation of drought effects (Huang and Chou, 2008), the frequency analysis of drought (Huang et al, 2014b; Mondal and Mujumdar, 2015) and drought prediction based on various atmospheric circulation indices (Cordery and McCall, 2000). Amongst these previous studies, more efforts have been focused on developing reliable drought indices, which can be applied for earlier detection of droughts including their intensity and spatial extent. For instance, Standardized Precipitation Index (SPI) is one of the widely used indices to monitor drought as sole parameter (e.g. Hayes et al, 1999) or in combination with other meteorological indices (for Iran, Morid et al, 2006), and to make a spatial and temporal 

analysis of drought (in Greece, Livada and Assimakopoulos, 2007).

However, studies focused on studying the propagation from meteorological drought to hydrological drought and their associations were highly rare (Vicente-Serrano and López-Moreno, 2005; Van Loon and Laaha, 2015; Barker et al, 2016), which is of important significance to reveal drought propagation process and mechanism, thus being helpful for establishing drought early warning system. Commonly, meteorological drought develops and ends relatively quickly, whilst hydrological drought is the result of meteorological drought. The two kinds of droughts can reflect the different stages of drought development to a certain degree. In general, the occurrence of hydrological drought is later than meteorological drought, and the corresponding propagation time depends on local landscape condition (Pandey and Ramasastri, 2001). Therefore, investigation of the propagation time from meteorological to hydrological drought and its potential influence factors has important significance for establishing an effective monitoring and warning system of hydrological drought based on meteorological drought. This system is highly helpful for the mitigation of droughts and is expected to largely decrease drought-caused damages. Therefore, investigation of the propagation from meteorological drought to hydrological drought and its potential influence factors has importantly practical meanings.

Although SPI has been widely adopted to investigate drought in various regions, it is broadly accepted that SPI with different time scales can impact the assessment of drought conditions in different regions due to the fact that the response of different usable water sources to precipitation deficit can be extremely different. The fundamental advantage of SPI is that it can be computed for a variety of time scales, which allows SPI to monitor short-term drought such as agricultural drought and long-term drought such as hydrological drought (Mishra and Singh, 2010). For instance, Szalai et al

(2000) stated that agricultural drought was characterized best by SPI at a scale of 2–3 months. Hence, the appropriate time scale of SPI can be used to reflect the propagation time from meteorological to hydrological drought. Recently, the Budyko hypothesis has been widely utilized to investigate basin-scale water and energy balances (Yang et al, 2008; Roderick and Farquhar, 2011; Wang and Hejazi, 2011b; Yang and Yang, 2011; Xu et al, 2014; Yang et al, 2014), and the Fu’s equation is one of the formulations of the Budyko curve (Yang et al, 2007; Li et al, 2013). The parameter w of the Fu’s equation modifies the partitioning of precipitation between evaporation and runoff. The Budyko hypothesis is an effective tool for investigating the interactions among climate, hydrological cycle, and vegetation (Roderick and Farquhar, 2011; Yang and Yang, 2011). Additionally, the parameter w is an empirical parameter controlling the shape of the Budyko curve and reflecting the impacts of other factors like land surface characteristics, which has strong effects on the propagation time from meteorological to hydrological drought. Some studies have found that the parameter w of the Budyko framework is associated with land surface characteristics such as soil types, vegetation cover, climate seasonality, as well as topography (Milly, 1993, 1994; Zhang et al, 2001, 2004; Yang et al, 2007, 2009; Shao et al, 2012; Williams et al, 2012). Hence, the parameter x can be regarded as an integrated parameter to reveal the impacts of underlying surface characteristics on the propagation time from meteorological to hydrological drought. Therefore, the correlations between the propagation time and the parameter w of the Fu’s equation were explored in this study aimed at revealing possible effects of the parameter w on the propagation time.

Arctic Oscillation (AO) is closely associated with the climate of middle and high latitudes regions (Hudgins and Huang, 1996).

Many studies demonstrated that meteorological, agricultural, and hydrological droughts are closely linked with climate indices such as the El Niño Southern Oscillation (ENSO) and Atlantic Oscillation (AO) (Talaee et al, 2014). Evaporation is a complex parameter controlling mass and energy exchange in atmosphere and terrestrial ecosystems, playing a critical role in the mass and heat fluxes in the global atmospheric system. Hence, it can be adopted to monitor the variations of moisture and energy shifting from the ground to the atmosphere. Therefore, the large-scale at