EN中文

Journal of Geographical Sciences >

2018 , Vol. 28 >Issue 8: 1039 - 1058

DOI: https://doi.org/10.1007/s11442-018-1540-7

研究论文

Characteristics of dry-wet abrupt alternation events in the middle and lower reaches of the Yangtze River Basin and the relationship with ENSO

  • SHAN Lijie , 1 ,
  • ZHANG Liping , 1, 2, * ,
  • SONG Jiyun 3, 4 ,
  • ZHANG Yanjun 1 ,
  • SHE Dunxian 1 ,
  • XIA Jun 1
Expand
  • 1. State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan 430072, China
  • 2. School of Tourism Culture and Geographical Science, Huanggang Normal University, Huanggang 438000, Hubei, China
  • 3. Department of Applied Mathematics and Theoretical Physics, Centre for Mathematical Sciences, University of Cambridge, Wilberforce Road, Cambridge, CB3 0WA, UK
  • 4. Department of Architecture, University of Cambridge, 1-5 Scroope Terrace, Cambridge, CB2 1PX, UK
*Corresponding author: Zhang Liping, PhD and Professor, specialized in hydrology and water resources research. E-mail:

Author: Shan Lijie, PhD Candidate, specialized in extreme hydrologic events. E-mail:

Received date: 2017-12-06

  Accepted date: 2018-02-27

  Online published: 2018-08-10

Supported by

National Key Research and Development Program in China, No.2017YFA0603704;National Natural Science Foundation of China, No.51339004

Copyright

Journal of Geographical Sciences, All Rights Reserved
Fold

Abstract

During recent decades, more frequent flood-drought alternations have been seen in China as a result of global climate change and intensive human activities, which have significant implications on water and food security. To better identify the characteristics of flood-drought alternations, we proposed a modified dry-wet abrupt alternation index (DWAAI) and applied the new method in the middle and lower reaches of the Yangtze River Basin (YRB-ML) to analyze the long-term spatio-temporal characteristics of dry-wet abrupt alternation (DWAA) events based on the daily precipitation observations at 75 rainfall stations in summer from 1960 to 2015. We found that the DWAA events have been spreading in the study area with higher frequency and intensity since 1960. In particular, the DWAA events mainly occurred in May and June in the northwest of the YRB-ML, including Hanjiang River Basin, the middle reaches of the YRB, north of Dongting Lake and northwest of Poyang Lake. In addition, we also analyzed the impact of El Niño Southern Oscillation (ENSO) on DWAA events in the YRB-ML. The results showed that around 41.04% of DWAA events occurred during the declining stages of La Niño or within the subsequent 8 months after La Niño, which implies that La Niño events could be predictive signals of DWAA events. Besides, significant negative correlations have been found between the modified DWAAI values of all the rainfall stations and the sea surface temperature anomalies in the Nino3.4 region within the 6 months prior to the DWAA events, particularly for the Poyang Lake watershed and the middle reaches of the YRB. This study has significant implications on the flood and drought control and water resources management in the YRB-ML under the challenge of future climate change.

Key words： dry-wet abrupt alternation; the middle and lower reaches of the Yangtze River Basin; spatio-temporal characteristics; La Niño

Cite this article

SHAN Lijie , ZHANG Liping , SONG Jiyun , ZHANG Yanjun , SHE Dunxian , XIA Jun . Characteristics of dry-wet abrupt alternation events in the middle and lower reaches of the Yangtze River Basin and the relationship with ENSO[J]. Journal of Geographical Sciences, 2018 , 28(8) : 1039 -1058 . DOI: 10.1007/s11442-018-1540-7

1 Introduction

Global climate change and excessive human activities have induced more intense and frequent summer drought and flood anomalies throughout the world during past decades (Dai et al., 1998; Frich et al., 2002; May, 2004; Djebou et al., 2014; Li et al., 2015; Song et al., 2016). In particular, the alternations between wet spells and dry spells over a short period of time, namely dry-wet abrupt alternation (DWAA) events have frequently occurred in China since the 1990s, especially in the middle and lower reaches of the Yangtze River Basin (YRB-ML) as well as in the south and southwest of China (Wang et al., 2009; Feng et al., 2012; She et al., 2013; He and Lu, 2014; He et al., 2016). As a new feature and trend of summer drought/flood anomalies, the DWAA events have significant impact on the water and food security in China, for example, the grain yields of those croplands with less tolerance to flood and drought will be largely decreased (Dickin and Wright, 2008; Yu et al., 2012; Akhtar and Nazir, 2013). In order to solve the environmental problems induced by drought/flood anomalies, numerous researches have been done to identify the long-term characteristics of precipitation and to analyze the relationship between precipitation anomalies and large-scale ocean-atmospheric features (Wu et al., 2006a; Tang et al., 2007; Sun et al., 2012; Turner and Annamalai, 2012; Agnese et al., 2013; Gitau et al., 2013; Luo et al., 2013; Yang et al., 2013; Langousis and Kaleris, 2014; Huang, 2015). Specifically, for the YRB-ML region, the summer drought and flood events are found to have an increasing trend and are usually accompanied by high Southern Hemisphere annual mode, anomalously intraseasonal oscillation of West Pacific subtropical high, cross-equatorial wind anomalies, cyclones over East Asia, Asia Polar Vortex, Asia Meridional Circulation, as well as sea surface temperature anomalies due to El Niño (Wu et al., 2006b; Yang et al., 2013; Ji and Shan, 2015). As a result of the large-scale climate impact, the increasing frequency and intensity of extreme flood/drought events could lead to more DWAA events (Tang et al., 2007).
To better identify and quantify DWAA events, numerous quantification methods have been proposed. Wu et al. (2006b) defined a long-cycle drought-flood abrupt alternation index (LDFAI) by considering the differences between May-June and July-August precipitation and analyzed the correlation between the DWAA events in the YRB-ML and large-scale atmospheric circulation anomalies. Zhang et al. (2008) analyzed the characteristics of DWAA events of a city in the YRB-ML by calculating the percentages of anomalous ten-day precipitation events. The limitations of these methods include coarse time scale (seasonal or ten-day), fixed drought-flood alternation time point and lack of consideration of drought-flood alternation duration, which could lead to inaccurate identification of DWAA events. To overcome these limitations, in this paper we modified the LDFAI proposed by Wu et al. (2006b) and proposed a new daily-scale dry-wet abrupt alternation index (DWAAI) by considering prior and posterior dry/wet conditions as well as drought-flood alternation duration to better study the spatio-temporal characteristics of DWAA events in the YRB-ML.
In addition, we also studied the impact of large-scale climatic dynamics, specifically the El Niño Southern Oscillation (ENSO) on DWAA events in the YRB-ML. The ENSO is one of the strongest interannual variability signals in the coupled global ocean-atmosphere system and has significant impact on sea surface temperature (SST). Previous researchers mainly focused on the impact of ENSO on summer precipitation patterns and found that warmer winter SST of equatorial eastern Pacific Ocean, warmer summer SST of equatorial Indian Ocean, SST anomalies of East Australian Current and South China Sea as well as West Pacific warm pool can lead to abnormal increase of summer precipitation in the YRB-ML (Luo et al., 1985; Sun and Ma, 2003; Gong and He, 2006; Hartmann et al., 2008; Liu et al., 2008; Li et al., 2009; Li, 2013; Dong, 2016). Recent years have seen a few researchers’ attempts on analyzing the impact of ENSO on DWAA events (Li et al., 2014; Ma et al., 2014; Wang et al., 2014). For example, Feng et al. (2012) found that a DWAA event in the YRB-ML in the early summer of 2011 mainly owed to the La Niño event from July 2010 to April 2011 and the corresponding anomalously cold SST in the Indian Ocean. Li et al. (2013) found that the persistent dry condition before a DWAA event in spring 2011 are induced by the eastward deflection of cold stream and subtropical high in the Northwestern Pacific Ocean as a result of the La Niño event in January-May, 2011, while the subsequent wet conditions after the DWAA event owed to the enhanced decay of La Niño in June by the increased sensible heat flux over the Tibetan Plateau. However, these studies are limited to a single DWAA event without a comprehensive statistical analysis on long-term historical DWAA events.
In this study, we will analyze the spatio-temporal characteristics of long-term historical DWAA events in the YRB-ML during the summer monsoon period (May-August) from 1960 to 2015 and assess the impact of ENSO on DWAA events on a long-term basis. The precipitation data in the study area (YRB-ML) and the methodology of new daily-scale DWAAI method are described in section 2, followed by the validation of the new DWAAI method in section 3. The spatio-temporal characteristics of DWAA events, the spatial distribution of Pacific SST anomalies and the statistics of El Niño/La Niño prior to DWAA events as well as the correlation between DWAAI and SST anomalies in the Nino3.4 region are discussed in section 4 and 5. The results of this study could give some important clues for the prediction of DWAA events in the YRB-ML in the future and provide significant guidance for drought and flood control and management in the YRB-ML.

2 Data and methods

2.1 Data

The middle and lower reaches of the Yangtze River Basin (YRB-ML) (106°54′-124°25′E, 24°30′-35°45′N) is located in the East Asian monsoon region with a drainage area of 800,000 km2. Strongly influenced by the monsoon features, the YRB-ML has significant nonuniform seasonal precipitation distribution patterns, as more than 50% of the annual precipitation comes from the monsoon season (May-August) every year. The climatic conditions in this area can easily lead to DWAA events, which means the time period with a precedent dry spell due to rainfall shortage in spring and a subsequent wet spell induced by summer monsoons and enhanced warm-cold air mixing in summer. Therefore, the YRB-ML has suffered some of the most severe DWAA disasters in China (Shen et al., 2012). However, the spatial distribution of DWAA events in the YRB-ML has substantial variations because of the wide geographic scope (Shan et al., 2015). If the characteristics of DWAA events in the YRB-ML are analyzed based on average precipitation of the whole basin, the spatial heterogenous signals of the DWAA events may be smoothed out. To overcome this limitation, the YRB-ML was divided into six sub-basins as the secondary partitions in this article, including the Hanjiang River watershed, the middle reaches of the Yangtze River, Dongting Lake watershed, Poyang Lake watershed, the lower reaches of the Yangtze River and the delta plains (Cheng et al., 2012). The location of the YRB-ML is shown in Figure 1. The daily precipitation dataset of 75 representative meteorological stations from 1960-2015 were obtained from the National Meteorological Information Center of China. Moreover, corresponding monthly SST data were generated from the National Oceanic and Atmospheric Administration (NOAA) - Extended Reconstructed Sea Surface Temperature (ERSST) dataset with 2°×2° latitude-longitude resolution.
Figure 1 The location of the middle and lower reaches of the Yangtze River Basin in China. The black dots denote the meteorological stations. Boundaries of sub-basins (black lines) and first-order streams in the basin (in blue) are shown.

2.2 Methodology

To describe the DWAA phenomena quantitatively and qualitatively, a long-cycle drought- flood abrupt alternation index (LDFAI) was defined by Wu et al. (2006b) as follows:
$LDFAI=({{R}_{78}}-{{R}_{56}})\cdot (|{{R}_{56}}|+|{{R}_{78}}|)\cdot {{1.8}^{-|{{R}_{56}}+{{R}_{78}}|}}$ (1)
where R56 and R78 refer to the normalized May-June and July-August precipitation, respectively. (R78-R56) represents the DWAA intensity term, (|R56|+|R78|) represents the dry-wet intensity, and ${{1.8}^{-|{{R}_{56}}+{{R}_{78}}|}}$ is the weighting coefficient, which enhances the weights of DWAA events and reduces the weights of persistent dry-wet phenomena.
The long-cycle droughts-floods abrupt alternation index (LDFAI) is the first index used to identify DWAA events, and can be easily calculated without artificial selection. Therefore, the definition of the LDFAI establishes the foundation for analyzing the characteristics and physical mechanisms of DWAA events. However, there are some limitations about this index. Firstly, the magnitude of the LDFAI was defined within 4 months, including 2 months for the dry spell and 2 months for the wet spell. Therefore, the turning point from the dry spell to the wet spell was fixed at the end of June or the beginning of July. However, DWAA events may occur at any time, and if the event occurred during the period from May to June, the antecedent precipitation could be considered as normal due to the average effect of the precipitation in the dry spell and wet spell, which would inevitably lead to the omission of some DWAA events. Secondly, the LDFAI only reflected the difference between mean precipitation in the two stages without explicitly considering the alternation duration from a dry spell to a wet spell, therefore cannot differentiate the two DWAA events from identical mean precipitation values during dry-wet spells but with different temporal distributions of precipitation in each spell.
To overcome these limitations of the LDFAI method, we modified the original LDFAI and proposed a new DWAAI as follows:
$DWAAI=\left[ K+\left( SP{{A}_{post}}-SP{{A}_{pre}} \right)\cdot (|SP{{A}_{post}}|+|SP{{A}_{pre}}|) \right]\cdot {{\alpha }^{-|SP{{A}_{pre}}+SP{{A}_{post}}|}}$ (2)
$K=\sum\limits_{i=1}^{n}{\left( \frac{SAP{{I}_{i}}-SAP{{I}_{0}}}{i} \right)}$ (3)
where SPApre and SPApost refer to the standardized precipitation anomalies (SPAs) of the pre-phase and post-phase of an event, respectively; SAPIi and SAPI0 represent the standardized antecedent precipitation index anomalies of the ith day in the post-phase and the last day in the pre-phase, respectively. In particular, standardized antecedent precipitation index anomalies are calculated by standardizing the antecedent precipitation index (McQuigg, 1954). n is the number of days prior to a precipitation event. A DWAA event in this paper is regarded as only an event that rapidly alternates from dry to wet, i.e., with a precedent dry spell and a subsequent wet spell. As the response times of dry-wet events are different, the durations of dry and wet spells should be considered independently. Lu (2009) noted that the antecedent and current precipitation contributed to the extents of floods and droughts, while the contribution from the antecedent precipitation decayed exponentially with time, and the influence of precipitation would decline to 1‰ after 44 days. Therefore, a dry spell is considered to be 44 days in this study. In addition, a wet spell is defined as 10 days. In brief, the duration of a DWAA event is the time period from 44 days before the abrupt alternation date to 10 days after that date.
In Equation (2), the term $K\cdot {{\alpha }^{-|SP{{A}_{pre}}+SP{{A}_{post}}|}}$ is defined as the “urgency” degree, the term$(SP{{A}_{post}}-SP{{A}_{pre}})\cdot (|SP{{A}_{post}}|+|SP{{A}_{pre}}|)\cdot {{\alpha }^{-|SP{{A}_{pre}}+SP{{A}_{post}}|}}$ is defined as the “alter-nation” degree from dry spell to wet spell, where ${{\alpha }^{-\left| SP{{A}_{pre}}+SP{{A}_{post}} \right|}}$ is the weighting coefficient.
To better understand the physical meaning of parameter a, the “alternation” degrees of DWAA events with different SPAs before and after DWAA events were analyzed (see Table 1). We selected 8 events for comparison, including 6 DWAA events (No.1-6) with different intensities and 2 non-DWAA events, i.e. a complete drought event (No.7) and a complete flood event (No. 8). The known order of the “alternation” degree of No.1-6 is as follows: No.3> No.2> No.1, No.6> No.5> No.4, No.1-6 > No.7-8. From Table 1, we can see that when parameter a is less than 1, the “alternation” degree of No.8 will be greater than that of No.1; when parameter a is larger than 1.4, the “alternation” degree of No.6 will be smaller than that of No.5. Both cases are unreasonable based on the known conditions of the orders. Therefore, we consider the range from 1.0 to 1.4 as the reasonable domain for parameter a, in particular, we set the value of parameter a in this article as 1.3.
Table 1 The “transition” degree of events that alter from dry to wet according to different values of parameter a
No. SPA “Transition” degree
Pre-
phase		 Post-
phase		 a=0.8 a=1.0 a=1.2 a=1.3 a=1.4 a=1.6 a=1.8 a=2.0 a=2.5 a=3.0
1 -1 1 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00
2 -1 2 11.25 9.00 7.50 6.92 6.43 5.63 5.00 4.50 3.60 3.00
3 -1 3 25.00 16.00 11.11 9.47 8.16 6.25 4.94 4.00 2.56 1.78
4 -2 1 11.25 9.00 7.50 6.92 6.43 5.63 5.00 4.50 3.60 3.00
5 -2 2 16.00 16.00 16.00 16.00 16.00 16.00 16.00 16.00 16.00 16.00
6 -2 3 31.25 25.00 20.83 19.23 17.86 15.63 13.89 12.50 10.00 8.33
7 -2 -3 -15.26 -5.00 -2.01 -1.35 -0.93 -0.48 -0.26 -0.16 -0.05 -0.02
8 2 3 15.26 5.00 2.01 1.35 0.93 0.48 0.26 0.16 0.05 0.02

3 Validation of the DWAAI method

In this section, we selected Wuhan station as test case to compare the LDFAI and DWAAI from 1960-2015 and discuss whether the modified DWAAI can better identify the occurrence of DWAA events in the YRB-ML. For each year, we selected the day with maximum DWAAI between May and August as the most “urgent” day or the most severe abrupt alternation day, and selected the period from 44 days before the “urgent” day and 10 days after the “urgent” day as the most severe DWAA event. The topmost 10 severe DWAA events in 10 different years with the maximum DWAAI or LDFAI are listed in Table 2, along with the distribution of SPA in the dry/wet periods shown in Table 3. Among the identified 10 severe years with the 10 highest LDFAI and DWAAI values, only two years (i.e. 1994 and 1998) coincide with each other. However, other DWAA events like the severe DWAA event in June 2011 can only be identified by DWAAI method. This is mainly because the DWAAI method is more flexible in the selection of dry-wet alternation point while the LDFAI method has fixed alternation point between period 1 (May-June) and period 2 (July-August). Specifically, the alternation time points identified by DWAAI method occasionally fall in the fixed time period between May-June and July-August of LDFAI method for the years 1994 and 1998, while fall out of the fixed time period of LDFAI method for other years.
Table 2 Years and comparison of normalized precipitation values for the 10 highest LDFAI values
Year LDFAI May-June July-August Year LDFAI May-June July-August
1963 5.06 -2.31 1.00 2010 1.74 -0.56 0.91
1969 3.77 -0.75 2.43 2006 1.45 -1.11 0.38
1998 2.77 -0.57 1.83 2003 1.26 -0.54 0.61
1994 2.33 -1.88 0.45 1962 1.18 0.00 1.91
1997 1.85 -1.86 0.30 1991 0.93 0.40 1.99
Table 3 Years and comparison of SPA values for the 10 highest DWAAI values
Year Abrupt
alternation date		 Pre-phase
SPA		 Post-phase
SPA		 “Urgency”
degree		 “Alternation”
degree		 DWAAI
2000 May 24 -1.93 2.21 6.69 15.99 22.68
1988 May 6 -1.79 3.19 4.99 17.19 22.18
2008 May 3 -1.50 3.61 5.75 15.02 20.77
2011 June 10 -1.66 4.96 1.74 18.41 20.15
1961 June 7 -1.29 3.49 5.33 12.82 18.15
2007 May 24 -1.64 2.88 2.60 14.70 17.30
1984 June 7 -1.49 2.69 2.97 12.73 15.70
1982 June 19 -0.76 3.76 4.74 9.29 14.03
1998 July 17 -0.78 5.81 1.77 11.60 13.37
1994 July 12 -1.26 2.22 3.90 9.39 13.29
From Table 3, we can see that the pre-phase SPA values are less than -1 for 8 high-DWAAI years, among which 5 years have SPA values less than -0.5, which indicates median and severe drought conditions (dry spells). The post-phase SPA values are greater than 2 for all the 10 years, among which 5 years have SPA values larger than 3, which indicates median and severe flood conditions (wet spells). It can also be seen from Table 3 that the high DWAAI values corresponding to high absolute SPA values capture the occurrence of DWAA events with high accuracy. To better clarify the physical meaning of DWAAI, the precipitation processes of events with high DWAAI values are shown in Figure 2, from which significant differences can be seen during and between dry spells and wet spells. For example, by comparing the precipitation process in Figures 2a and Figure 2c, the precipitation difference between wet spell and dry spell in 2011 was 54.1% higher than that in 1984, indicating that the DWAA event in 2011 was more severe than that in 1984. The same conclusion can be made by comparing the DWAAI vales of the two years since the DWAAI in 2011 (20.15) is higher than the DWAAI in 1984 (15.70). This proves that the DWAAI can reasonably represent the intensity of DWAA event, i.e., the “alternation” degree from a dry spell to a wet spell. In addition, by comparing the precipitation process in Figures 2b and 2c, we can see that the transition durations between dry spell and wet spell are similar for both year 1961 and year 1984, meaning similar “alternation” degrees of the DWAA events. However, since the first 5-day average of precipitation amount during the wet spell in 1961 is 35.66 mm greater than that in 1984, the DWAA event in 1961 is considered as more urgent and serious. This phenomenon is also reflected in the DWAAI values since the DWAAI value in 1961 (18.15) is greater than the value in 1984 (15.70).
Figure 2 Precipitation and DWAAI sequences of DWAA events in the high DWAAI years (i.e., (a) 2011, (b) 1961, (c) 1984, (d) 1982 and (e) 1998). Red lines and blue dashed lines represent the precipitation during periods of the pre-phase and post-phase, respectively. Black dotted lines represent the calculated DWAAI values during the whole event.
On the other hand, the “alternation” degree of the DWAA event in 1982 (Figure 2d) is considered low since the difference between the daily mean precipitation of the dry spell (3.72 mm) and the precipitation at the beginning of the wet spell (26.2 mm) is small. Besides, the “urgency” degree of the DWAA event in 1998 (Figure 2e) is considered low since the precipitation amount in the initial stage of wet spell is low with only 72.3 mm at the first day and 0 mm for the following 3 days. The low “alternation” degree and low “urgency” degree of the DWAA event in 1982 and 1998 respectively can also be reflected by the low values of DWAAI of both events (14.03 for the year 1982 and 13.37 for the year 1998).
All the above examples about DWAA events show that the DWAAI is an effective index to represent both the “alternation” degree and the “urgency” degree of DWAA events. Specifically, the events with DWAAI values over 15 are defined as DWAA events, and the larger the DWAAI, the more serious the DWAA event. Events with the DWAAI values between 20 and 23 are defined as moderate DWAA events, while those under 20 and over 23 are defined as mild and severe DWAA events, respectively. Moderate and severe DWAA events with DWAAI larger than 20 are considered as high-intensity DWAA events.
To validate the effectiveness of the modified DWAAI method, we identified all the DWAA events for six representative rainfall stations in six sub-basins of YRB-ML based on both the LDFAI values and DWAAI values (Table 4). It was found that only a few events occurred at the end of June or early July were selected by both indices. On the other hand, 76.47% DWAA events that occurred in May and June can only be identified by DWAAI method. In particular, for a known severe DWAA event recorded in most of the YRB-ML during spring to early summer in 2011, the DWAAI method identified DWAA events at 28 stations within the basin, which were concentrated in the southeast of Hubei Province, central-north Hunan Province and central-north Jiangxi Province, which are identical to the actual conditions. However, during this 2011 DWAA event, only 5 stations located in the northwest of the Hanjiang River basin and the delta plains were identified by the LDFAI, which does not conform to the real case. In addition, many DWAA events identified by the LDFAI are not realistic. Two typical examples of wrong identification by the LDFAI include the 1963 event at Tianmen station and the 1993 event at Shimen station. For the Tianmen station, the average precipitation values of dry and wet spells were 2.23 mm and 6.30 mm respectively on the selected “alternation” date, i.e., July 30, 1963 event. For the Shimen station, the average precipitation values of the dry and wet spells were 4.90 mm and 21.66 mm, respectively for the July 30 of 1993 event. However, in reality, both events cannot be regarded as DWAA events since the difference of mean precipitation between dry spell and wet spell is relatively small with less precipitation in the wet spell and more precipitation in the dry spell. From the above test cases, we can see that the DWAAI method can identify the DWAA events in the YRB-ML region with more accuracy in the occurrence time as well as “alternation” and “urgency” degrees than the LDFAI method.
Table 4 Comparison of DWAA events selected by the LDFAI and DWAAI
Stations Sub-basins DWAA events selected by LDFAI DWAA events (abrupt alternation date) selected by DWAAI
Tianmen Hanjiang River watershed sub-basin 1963, 1968 1961 (June 8), 1968 (July 14),
1969 (June 9), 1982 (May 12).
1997 (June 6), 2000 (May 24)
Jingzhou The middle reaches of the Yangtze River 1968, 1991, 2007 1968 (July 13), 1981 (May 27),
1991 (July 1), 2000 (May 24),
2011 (June 13)
Shimen Dongting Lake watershed sub-basin 1991, 1993, 2007, 2008, 2014 1982 (May 26), 1986(June 4), 1988 (May 5), 1991 (July 1), 1996 (May 31), 2006 (May 5), 2011 (June 10)
Zhangshu Poyang Lake watershed sub-basin _ 1982 (June 14), 1985 (June 4), 1988 (May 9), 2011 (June 3)
Huangshan The lower reaches of the Yangtze River 1965, 1987, 1997, 2009 1992 (August 26), 1994 (June 8), 1996 (June 3), 2000 (May 25), 2008 (June 8), 2009 (July 24), 2011 (June 4)
Nantong Delta plains 1965, 1980, 1982, 1987, 1997, 2003, 2006, 2007, 2010, 2014 1971 (May 17), 1974 (May 4), 1981 (June 24), 2003 (June 29), 2010 (July 3)

Boldface and underlined font indicate that the events that occurred in the same year were identified by both indices

4 Spatio-temporal characteristics of DWAA events in the YRB-ML

4.1 Temporal evolution characteristics of DWAA events

Based on the daily precipitation data at 75 rainfall stations from 1960 to 2015 in the YRB-ML, the characteristics of the temporal distribution of DWAA events during summer were analyzed by calculating the DWAAI. The statistics of the stations where different levels of events occurred in each year of the basin are shown in Figure 3. It can be seen that DWAA events occurred throughout the years in the YRB-ML with significant interannual differences between different rainfall stations, especially after the year 1986 (Figure 3a). Generally, the DWAA events are increasing in the YRB-ML with enhanced frequency and intensity since 1986, especially for moderate and severe DWAA events. The 3 most severe years after 1986 were 1988, 2000 and 2011 when there were relatively more stations with DWAA events. In addition, the mean DWAAI values of these 3 years were greater than 19, and the stations with high-intensity events accounted for 43.48%, 55.17% and 44.83% of the total stations with DWAA events in 1988, 2000 and 2011, respectively, which indicates that these years had higher incidences of high-intensity DWAA events.
Figure 3 Statistics of stations where summer DWAA events occurred in the YRB-ML. White, reticular and black rectangles represent the number of stations with severe, moderate and mild DWAA events, respectively
In particular, during the summer monsoon season (May-August), most of the DWAA events with enhanced frequency and intensity occurred in May and June, which accounted for 31.29% and 37.41% of the total events, respectively (Figures 3b and 3c). The number of stations with DWAA events in May and June exhibited a comparatively large interannual difference, especially during 1988-2015. On the other hand, there were fewer stations with DWAA events in July and August than in May and June. Moreover, the DWAA events in July (Figure 3d) features non-significant interannaual change of station numbers and decreased intensity throughout the years, while a significant increasing trend of enhanced DWAA events has been found in August (Figure 3e).

4.2 Spatial distribution characteristics of DWAA events

The spatial distribution and interdecadal variability of the frequency of DWAA events in the YRB-ML are shown in Figure 4. DWAA events tend to occur more frequently in the Hanjiang River Basin, the middle reaches of the Yangtze River, the north region of the Dongting Lake watershed and the northwest region of the Poyang Lake watershed (Figure 4a). Among these events, the high-intensity DWAA events are concentrated in the central region of the Hanjiang River watershed sub-basin and the middle reaches of the Yangtze River with high frequency of once per six years, while the low-intensity DWAA events usually occurred in the southern part of the Dongting Lake watershed sub-basin, the southern part of the Poyang Lake watershed sub-basin and the delta plains with low frequency of once per 12 years.
Figure 4 Spatial distributions of the interdecadal frequencies of summer DWAA events in the YRB-ML. All of the stations with DWAA events are shaded, and stations with high-intensity DWAA events are marked by green dots.
Figures 4b-4g show the spatial distribution of summer DWAA events in the 1960s, 1970s, 1980s, 1990s, 2000s and 2010s respectively. Based on the comparison of these figures, the main characteristics of each decade can be summarized as follows. In the 1960s and the 1970s, there are less DWAA events with low intensities. Specifically, DWAA events mainly occurred in the middle reaches of the Yangtze River and northern region of the basin in the 1960s and occurred in the Hanjiang River watershed sub-basin and the southeast of the Dongting Lake watershed sub-basin in the 1970s. In the 1980s, DWAA events occurred throughout the northwest of the YRB-ML with higher frequencies and intensities, especially for the middle reaches of the Hanjiang River Basin, the northwest of Dongting Lake watershed and the middle reaches of the Yangtze River. In the 1990s, the DWAA events have lower frequencies and intensities with shifted coverage mainly in the middle reaches of the Hanjiang River Basin, the west of the Dongting Lake watershed, the northeast of Poyang Lake watershed and the lower reaches of the Yangtze River. In the 2000s, DWAA events occurred throughout the YRB-ML with significantly enhanced intensities. In addition, high-intensity DWAA events occurred in most areas in the first six years of the 2010s. In general, from 1960 to the present, DWAA events have been spreading out in the whole YRB-ML with increasing trends of frequencies and intensities.

5 Impact of ENSO on DWAA events in the YRB-ML

5.1 Distribution of Pacific SST anomalies before DWAA events

To investigate the impact of ENSO on DWAA events in the YRB-ML, we analyzed the abnormalities of Pacific SST in the precedent year before each DWAA event. The spatial distributions of Pacific SST abnormalities are shown in Figures 5-7. The results show that 6 months before a DWAA event, the SST of the West Pacific starts to become abnormally high while the SST of the eastern equatorial Pacific (with the anomaly center point located in the range of 5°N-5°S, and 150°W-130°W or more specifically in the Nino3.4 region) starts to be abnormally low and then gradually becomes higher after the coldest time (i.e., two months before the DWAA event). In the meanwhile, the center point of eastern equatorial Pacific SST anomaly center will be shifted eastward from the region lying between 5°N-5°S and 150°W-130°W to that of 5°N-5°S and 110°W. It is noteworthy that the higher the intensity of the event, the more severe the cold SST anomaly in the eastern equatorial Pacific will be. Specifically, the SST anomalies during the period from 4 months before the occurrence of a severe event to the occurrent month of the event are continuously less than -0.4°C, which coincides with a La Niño event.
Figure 5 Spatial distributions of Pacific SST anomalies before mild DWAA events. The dotted regions indicate values that are significant at the 0.1 level
Figure 6 Spatial distributions of Pacific SST anomalies before moderate DWAA events. The dotted regions indicate values that are significant at the 0.1 level
Figure 7 Spatial distributions of Pacific SST anomalies before severe DWAA events. The dotted regions indicate values that are significant at the 0.1 level
In addition, we also analyzed the SST distributions in the South China Sea, since the South China Sea is the closest part of the West Pacific to China and is within the upstream of monsoon currents that will directly affect the precipitation patterns in the YRB-ML. We found that the SSTs of the South China Sea are abnormally high with a continuous high anomaly of 0.4°C during the period from 4 to 6 months before the occurrence of DWAA events, especially for high-intensity DWAA events. The anomalously high SST in the South China Sea could also be contributed to the occurrence of La Niño based on Tan et al. (1995). Thus, we assume that there could be some correlation between SST anomalies led by La Niño and the occurrence of DWAA events in the YRB-ML and will test our assumptions with more details in the following context.
To better understand the distributions of SST anomalies prior to DWAA events in the YRB-ML, we have done statistical analysis for the SST in the Nino3.4 area within one year before the occurrence of all DWAA events (Figure 5). Based on the national standard Identification Standard for El Niño/La Niño Events, an El Niño event is defined as a phenomenon in the equatorial Pacific Ocean with a 3-monthly moving average of SST anomaly greater than or equal to 0.5°C for at least 5 consecutive months; while a La Niño event is defined as a phenomenon in the equatorial Pacific Ocean with a 3-monthly moving average of SST anomaly less than or equal to 0.5°C for at least 5 consecutive months. We found that more than 52% of the DWAA events occurred after the phenomena of persistently low SST in the Nino3.4 region (such as La Niño events or a year of negative SST anomalies), and 22.22% of the DWAA events occurred with the phenomena of consistently high SST anomalies in the Nino3.4 region. It indicates that the DWAA events in the YRB-ML have more tendency to occur when the SST in the Nino3.4 region is anomalously lower for a continuous period of time, which is consistent with the conclusions from Figures 5-7.
Table 5 Abnormal conditions of SST in the Nino3.4 region within 1 year before DWAA events
SST abnormal conditions Occurrence frequency Percentage (%)
La Niño 181 41.04
SST anomalies persistently less than 0 (La Niño has not formed yet) 52 11.79
El Niño 62 14.06
SST anomalies persistently larger than 0 (El Niño has not formed yet) 36 8.16
Other 110 24.95

5.2 Statistics of El Niño/La Niño events that occurred before DWAA events

To analyze the impact of La Niño events on DWAA events, we selected 3 severe DWAA events in 3 La Niño years, including the years 1988, 2000 and 2011 (see Figure 3a) and did statistical analysis based on the concurrent La Niño- DWAA events. The statistics of mild, moderate and severe DWAA events that occurred during the declining stages of La Niño events or within the first 8 months after the end of La Niño events are presented in Table 6. The corresponding spatial distributions of Pacific SST anomalies in the Nino3.4 region before mild, moderate and severe DWAA events are shown in Figures 5-7, respectively. It can be seen from Table 6 that 181 DWAA events occurred during the declining stages of La Niño or within the first 8 months after La Niño ended, accounting for approximately 41.04% of all events. Specifically, 120 of those events occurred 4 to 6 months after the month with the coldest SST. In addition, there are more high-intensity DWAA events occurred within the first 8 months after La Niño, and with the lowest SST in the 4th to 6th month before the DWAA event. The higher the intensity of the event, the larger the proportions of events that occurred within the first 8 months after La Niño ended and occurred 4 to 6 months after the month with the coldest SST. Moreover, we found that there are 62 events that were preceded by El Niño events within 1 year, including 54 mild DWAA events, 6 moderate DWAA events and 2 severe DWAA events, among which the proportion of severe events is small. These results indicate that La Niño events could be predictive signals for the occurrence of DWAA events, and the higher the intensity of the DWAA events, the higher the probability that these events will occur after La Niño.
Table 6 Statistics of DWAA events that occurred during the declining stages of La Niño events or within the first 8 months after La Niño events ended
Intensity of events Occurrence frequency Months with the coldest SST in the Nino3.4 region before DWAA events occurred Percentage (%)
1st month 2nd month 3rd month 4th month 5th month 6th month 7th month 8th month and above Total number of times
Mild 330 4 4 1 19 32 20 18 18 116 35.15
Moderate 74 2 1 0 16 14 2 2 6 43 58.11
Severe 37 1 0 0 6 8 3 1 3 22 59.46
Total 441 7 5 1 41 54 25 21 27 181 41.04

5.3 Correlation between DWAAI and SST anomaly values before DWAA events

To better quantify the impact of SST anomaly on DWAA events, we calculated the correlation coefficients between the DWAAI values in different sub-basins and the SST anomalies in the Nino3.4 region before DWAA events. Temporally, there is a general significantly negative correlation at the 0.05 significance level between the DWAAI values of all stations in the YRB-ML and the SST anomalies during the first six months before DWAA events with the strongest negative correlation in the 3 months prior to DWAA events. Spatially, the strongest negative correlation between DWAAI values and SST anomalies occurred in the Poyang Lake watershed since the correlations from the 2nd month to the 6th month before the DWAA events occurred are all significant at the 0.01 level with the largest absolute correlation coefficient of 0.39 in the 3rd month before the DWAA event. The second strongest negative correlations between the DWAAI values and SST anomalies occurred in the middle reaches of the Yangtze River for the first 3 months before DWAA events with the largest absolute correlation coefficient of 0.44 in the first month before the DWAA event. On the other hand, the correlation coefficients in the lower reaches of the Yangtze River and the delta plains range from 0.26 to -0.19, which are not statistically significant at the 0.05 level. In addition, the correlations in the Hanjiang River Basin and Dongting Lake watershed are the weakest with absolute correlation coefficients less than 0.1. Based on the above results, we can see that there are significant negative correlations between the DWAAI values and the SST anomalies in the Nino3.4 region in the precedent 3 months of the DWAA events for the Poyang Lake watershed and the middle reaches of the Yangtze River.
It is also noteworthy that different types of El Niño/La Niño events have different developing features of SST anomalies and could result in completely different summer precipitation patterns in China for the following year (Yuan et al., 2012). The East Pacific type of El Niño/La Niño index (IEP) and Central Pacific type of El Niño/La Niño index (ICP) were also calculated in this paper according to the Identification Standard for El Niño/La Niño Events. The DWAAI~IEP and DWAAI~ICP correlations for the 6 months before an event occurring in different sub-basins are shown in Table 8. It can be found that the DWAAIs have weak correlations with pre-phase IEP, but strong correlations with pre-phase ICP. More specifically, the DWAAI values of all stations in the YRB-ML have significantly negative correlation with ICP at the 0.05 significance level during the first five months before DWAA events with the largest correlation coefficients at the 2nd month before the DWAA event. In addition, by comparing Table 8 with Table 7, we can see that although the general trends of correlation between the DWAAI and different large-scale climatic symbols (SST anomaly, IEP or ICP) are similar for each sub-basin, the significance of correlation between DWAAI and climatic symbols are different. For example, the correlations between the DWAAI and the SST anomalies in the lower reaches of the Yangtze River and the delta plains are not significant at the 0.05 level, as shown in Table 7. However, the correlation between the DWAAI values and ICP in the second month before the DWAA events are significant at the 0.05 level with correlation coefficients of -0.30 and -0.43 respectively for the lower reaches of YRB and the delta plains. The results give us some important hints that we can use more climatic symbols including but not limited to SST anomaly, IEP and ICP to better investigate the impact of ENSO or global climate change on DWAA events.
Table 7 Correlations between DWAAI values and SST anomalies in the Nino3.4 region before DWAA events
Sub-basins Occur. frequency Months before DWAA events occurred
1st month 2nd month 3rd month 4th month 5th month 6th month
Hanjiang River watershed 103 0.06 0.09 0.03 0.00 0.00 0.00
Dongting Lake watershed 144 -0.06 -0.04 -0.06 -0.05 -0.07 -0.06
Poyang Lake watershed 80 -0.23 -0.31 -0.39 -0.38 -0.37 -0.34
Middle reaches of the Yangtze River 47 -0.44 -0.42 -0.36 -0.23 -0.20 -0.18
Lower reaches of the Yangtze River 47 -0.21 -0.20 -0.24 -0.26 -0.22 -0.20
Delta plains 20 -0.19 -0.20 -0.26 -0.24 -0.22 -0.25
Total basin 441 -0.12 -0.12 -0.15 -0.14 -0.14 -0.12

Boldface and underline font indicate the correlations that are significant at the 0.05 level.

Table 8 Correlations between DWAAI values and IEP/ICP before DWAA events
Sub-basins Occur. frequency Indices Months before DWAA events occurred
1st month 2nd
month		 3rd
month		 4th month 5th month 6th month
Hanjiang River
watershed		 103 IEP 0.13 0.12 0.03 0.04 0.05 0.05
ICP 0.02 0.02 0.03 -0.02 -0.04 -0.04
Dongting Lake
watershed		 144 IEP 0.03 0.10 0.04 0.09 0.02 0.01
ICP -0.11 -0.13 -0.08 -0.15 -0.16 -0.14
Poyang Lake
watershed		 80 IEP -0.05 -0.05 -0.12 -0.26 -0.33 -0.28
ICP -0.29 -0.41 -0.39 -0.40 -0.22 -0.20
The middle reaches of the Yangtze River 47 IEP -0.19 -0.25 -0.39 -0.26 -0.23 -0.27
ICP -0.24 -0.30 -0.25 -0.11 -0.08 -0.00
The lower reaches of the Yangtze River 47 IEP -0.01 -0.06 -0.13 -0.32 -0.30 -0.25
ICP -0.25 -0.23 -0.19 -0.06 -0.07 -0.06
Delta plains 20 IEP 0.23 0.20 -0.03 -0.25 -0.23 -0.29
ICP -0.37 -0.43 -0.36 -0.12 -0.14 -0.13
Total basin 441 IEP 0.01 0.02 -0.05 -0.07 -0.10 -0.09
ICP -0.15 -0.18 -0.14 -0.12 -0.10 -0.09

Boldface and underline font indicate the correlations that are significant at the 0.05 level

6 Conclusions

In this study, we modified a seasonal-scale LDFAI method and proposed a new daily-scale DWAAI method and applied the new method to analyze the spatio-temporal characteristics of DWAA events in summer periods (May-August) from 1960 to 2015 over the YRB-ML region in China. The new DWAAI method is found to be more accurate and effective than the original LDFAI method in identifying the summer DWAA events in the YRB-ML, since it considers finer time scale, extra prior and posterior dry/wet conditions of DWAA events as well as drought-flood alternation duration. Based on the new DWAAI method, we found that the frequency and intensity of DWAA events are more significantly increased in May, June, less significantly increased in August and slightly decreased in July throughout the years. Besides, we found that although the occurrence frequency of DWAA events is increasing throughout the YRB-ML, the spatial distribution of DWAA events in the YRB-ML is uneven. Specifically, it is found that DWAA events mainly occur in the northwestern part of the YRB-ML, including Hanjiang River Basin, the middle reaches of the YRB, north of Dongting Lake watershed and northwest of Poyang Lake watershed.
In addition, we also analyzed the impact of large-scale climatic dynamics (i.e., ENSO) on the DWAA events in the YRB-ML by establishing the correlation between DWAAI and SST in the Nino3.4 region prior to DWAA events. It is found that the continuously low SST in the Nino3.4 region has profound impact on the occurrence of DWAA events in the YRB-ML. In particular, 41.04% of DWAA events occurred during declining stages of La Niño, within the subsequent 8 months after a La Niño, or in the subsequent 4th-6th months after the coldest month in the Nino3.4 region, implying that La Niño is a key predictive signal of DWAA events in terms of occurrence time. On the other hand, significant negative correlations were found between the DWAAI at all stations and the SST anomalies in the Nino3.4 region prior to DWAA events, especially for the Poyang Lake watershed and the middle reaches of the Yangtze River.
In order to predict the occurrence time and intensity of DWAA events more accurately in the future, we can try to establish the correlations between DWAA events with more climatic symbols (including not only SST anomaly but also other climatic indices like IEP and ICP) as well as use land cover change characteristics under climate change and rapid urbanization in the fast-developing China (Song and Wang, 2016). In addition, we need to have a better understanding on the physical mechanisms of atmospheric dynamics at local, regional and global scale as well as land-atmospheric interactions (Song and Wang, 2015) before, during and after DWAA events. And then DWAA events can be simulated and predicted eventually by using climate models, which can be able to capture the genesis, intensification of depressions and their track (Stowasser et al., 2009; Turner and Annamalai, 2012). These mechanisms and predictability need further investigations and will be researched in other papers.

The authors have declared that no competing interests exist.

References

Publishing order | Descend order by publishing year | Descend order by cited within
1
Agnese C, Baiamonte G, Cammalleri C, 2013. Modelling the occurrence of rainy days under a typical Mediterranean climate.Advances in Water Resources, 64: 62-76.The statistical inference of the alternation of wet and dry periods in daily rainfall records can be achieved through the modelling of inter-arrival time-series, IT, defined as the succession of times elapsed from a rainy day and the one immediately preceding it. In this paper, under the hypothesis that ITs are independent and identically distributed random variables, a modelling framework based on a generalisation of the commonly adopted Bernoulli process is introduced. Within this framework, the capability of three discrete distributions, belonging to the Hurwitz–Lerch-Zeta family, to reproduce the main statistical features of IT time-series was tested. These distributions 61 namely Lerch-series (Lerch), polylogarithmic-series (Polylog) and logarithmic-series (Log) 61 were selected thanks to their capability to describe some peculiar properties usually observed in IT series derived from daily rainfall records: very high standard deviation and skewness, relatively high frequency associated to the unitary IT, monotonically decreasing frequencies with a slow decay. Both Polylog and Log distributions are special cases of the 3-parameter Lerch distribution with a decreasing number of free parameters (2- and 1-parameter, respectively). The analysis, performed on 55 raingauges located in Sicily (Italy) under a typical Mediterranean climate, suggests that a reliable statistical representation of IT can be attained with the 3-parameter Lerch distribution. Despite the marked seasonality of rainfall in the study area, a simple subdivision of the year into two 6-month periods, roughly corresponding to the dry “semester” (D-sem) and the wet “semester” (W-sem), allows a satisfactory reproduction of IT, as well as of wet spells (WS) and dry spells (DS), separately. It was also noticed that the 2-parameter Polylog distribution could be successfully used to reconstruct the DS frequency distribution only. This result suggests that the additional parameter of the Lerch distribution is required by the inclusion of WS into the analysis. Finally, considering that Polylog outperforms the commonly adopted Log, a noteworthy step forward in DS modelling can be achieved by using Polylog distribution rather than Log one.

DOI

2
Akhtar I, Nazir N, 2013. Effect of waterlogging and drought stress in plants.International Journal of Water Resources and Environmental Sciences, 2(2): 34-40.

3
Cheng Z, Ding Xi J, Xu Met al., 2012. Climate characters of typical droughts-floods abrupt alternation events in the middle-lower reaches of the Yangtze River.Resources and Environment in the Yangtze Basin, 21(Z2): 115-120. (in Chinese)Using standardized precipitation index and historical disaster data,droughts-floods abrupt alternation events in second-level hydrology partition of middle-lower reaches of the Yangtze River were selected,their spatio-temporal distribution characters and the differences in frequency and intensity of partitions were also analyzed.The results showed that from 1960 to 2011 typical droughts-floods abrupt alternation events happened,the frequency was about 10 years,in which the intensity index and rainfall intensity in lower main reaches were stronger.There were obvious adjustments when droughts-floods abrupt alternation events happened.There are notable differences in south and east Asian high latitude areas of 200 hPa height field and subtropical east Asian areas in 850 hPa wind field.

4
Dai A G, Trenberth K E, Karl T R, 1998. Global variations in droughts and wet spells: 1900-1995.Geophysical Research Letters, 25(17): 3367-3370.The Palmer Drought Severity Index (PDSI) was calculated globally using gridded monthly air temperature and precipitation. From 1900 to 1995, there are large multi-year to decadal variations in the percentage areas in severe drought (PDSI < -3.0) and severe moisture surplus (PDSI > +3.0) over many land areas while secular trends are small. Since the late 1970s, however, there have been some increases in the combined percentage areas in severe drought and severe moisture surplus, resulting from increases in either the drought area (e.g., over the Sahel, eastern Asia and southern Africa) or both the drought and wet areas (e.g., over the U.S. and Europe). Although the high percentages of the dry and wet areas in the recent decades are not unprecedented during this century (except the Sahel), the recent changes are closely relate to the shift in El Ni o-Southern Oscillation (ENSO) towards more warm events since the late 1970s and coincide with record high global mean temperatures. Moreover, for any given value of ENSO indices, the PDSI anomalies tend to be larger than would be expected from previous records. These changes are qualitatively consistent with those expected from increased green-house gases in the atmosphere.

DOI

5
Dickin E, Wright D, 2008. The effects of winter waterlogging and summer drought on the growth and yield of winter wheat (Triticum aestivum L.).European Journal of Agronomy, 28(3): 234-244.Winter waterlogging and summer drought may become more prevalent as a result of climate change. Their effects on the growth and yield of winter wheat were investigated. Wheat was grown in lysimeters in an unheated glasshouse, over two seasons. Seed rate was included as an additional factor in the first season, and cultivar in the second. Root growth was investigated in both seasons using mini-rhizotrons. Waterlogging for 44 days at 93 days after sowing in 2002, and 58 days at 64 days after sowing in 2003, decreased grain yield by 20% and 24%, respectively. Drought during grain filling further decreased yields but there was no evidence that winter waterlogged plants were more susceptible to damage from drought the following summer, the effects of the two stresses being additive. Waterlogging decreased the total length, but not the final depth of the root system. Plots with a lower plant density demonstrated a smaller decrease in yield due to waterlogging. There was a significant positive linear relationship between the number of shoots per plant and nodal root axes per plant. There appeared to be a difference between cultivars in root system architecture, and in their response to waterlogging, but these differences were not reflected in grain yield.

DOI

6
Djebou D C S, Singh V P, Frauenfeld O W, 2014. Analysis of watershed topography effects on summer precipitation variability in the southwestern United States.Journal of Hydrology, 511: 838-849.With climate change, precipitation variability is projected to increase. The present study investigates the potential interactions between watershed characteristics and precipitation variability. The watershed is considered as a functional unit that may impact seasonal precipitation. The study uses historical precipitation data from 370 meteorological stations over the last five decades, and digital elevation data from regional watersheds in the southwestern United States. This domain is part of the North American Monsoon region, and the summer period (June–July–August, JJA) was considered. Based on an initial analysis for 1895–2011, the JJA precipitation accounts, on average, for 22–43% of the total annual precipitation, with higher percentages in the arid part of the region. The unique contribution of this research is that entropy theory is used to address precipitation variability in time and space. An entropy-based disorder index was computed for each station’s precipitation record. The JJA total precipitation and number of precipitation events were considered in the analysis. The precipitation variability potentially induced by watershed topography was investigated using spatial regionalization combining principal component and cluster analysis. It was found that the disorder in precipitation total and number of events tended to be higher in arid regions. The spatial pattern showed that the entropy-based variability in precipitation amount and number of events gradually increased from east to west in the southwestern United States. Regarding the watershed topography influence on summer precipitation patterns, hilly relief has a stabilizing effect on seasonal precipitation variability in time and space. The results show the necessity to include watershed topography in global and regional climate model parameterizations.

DOI

7
Dong Z L, 2016. Effects of the preceding SST and water vapor transport path on the summer precipitation over the Yangtze River Valley [D]. Hefei: University of Science and Technology of China. (in Chinese)

8
Feng G L, Yang H W, Zhang S Xet al., 2012. A preliminary research on the reason of a sharp turn from drought to flood in the middle and lower reaches of the Yangtze River in late spring and early summer of 2011.Chinese Journal of Atmospheric Sciences, 36(5): 1009-1026. (in Chinese)Using some kinds of meteorological factors data provided by NCEP/NCAR NOAA and National Climate Center(NCC),and NOAA-Hysplit model to preliminarily analyze characteristics of precipitation and mechanism of a sharp turn from drought to flood in the middle and lower reaches of the Yangtze River in the early summer of 2011,the authors establish a synoptic concept model and conclude that:1) The most serious sharp turn from drought to flood occurred in the middle and lower reaches of the Yangtze River,especially in the region of(27oN-32oN,110oE-120oE),in the 1st pentad in June(around 3 June);2) analysis of sea surface temperature(SST) anomaly and the singular value decomposition(SVD) between SST and precipitation shows that the variance contribution of this sharp turn is smaller and it is an anomalous event,whose early signals exist in the equatorial middle an eastern Pacific and the equatorial Indian Ocean;3) analysis of vorticity low frequency oscillation and 10-30 d stable component indicates that dynamic mechanism across the Pacific Ocean has a great influence on precipitation in the middle and lower reaches of the Yangtze River in June,and the stable circulation in middle and high latitude areas provides a large scale circulation background field;4) before the sharp turn,there existed a stronger La Ni in the equatorial middle an eastern Pacific,and the equatorial Indian Ocean was controlled by negative SST anomaly on a large scale,influenced by the both,the western Pacific subtropical high was abnormally eastward,the Walker circulation was stronger,and the Hadley circulation over the Indian Ocean was weaker.All of those leaded to the lack of water vapor transport from southwest and southeast,thus a serious and sustained drought occurred in the middle and lower reaches of the Yangtze River.As the SST anomaly in both the oceans weakened,the western Pacific subtropical high suddenly extended westward and became stable(in the 1st pentad in June),the Walker circulation became weak,and the Hadley circulation over the Indian Ocean became stronger.Matching with the stable circulation on a large scale in middle and high latitude areas,it resulted in continuous converging of cold and warm air and vast precipitation over the middle and lower reaches of the Yangtze River.Finally,the sharp turn happened.

DOI

9
Frich P, Alexander L V, Della-Marta Pet al., 2002. Observed coherent changes in climatic extremes during the second half of the twentieth century.Climate Research, 19(3): 193-212.Abstract A new global dataset of derived indicators has been compiled to clarify whether frequency and/or severity of climatic extremes changed during the second half of the 20th century, This period provides the best spatial coverage of homogenous daily series, which can be used for calculating the proportion of global land area exhibiting a significant change in extreme or severe weather. The authors chose 10 indicators of extreme climatic events, defined from a larger selection, that could be applied to a large variety of climates. It was assumed that data producers were more inclined to release derived data in the form of annual indicator time series than releasing their original daily observations. The indicators are based on daily maximum and minimum temperature series, as well as daily totals of precipitation, and represent changes in all seasons of the year. Only time series which had 40 yr or more of almost complete records were used, A total of about 3000 indicator time series were extracted from national climate archives and collated into the unique dataset described here. Global maps showing significant changes from one multi-decadal period to another during the interval from 1946 to 1999 were produced. Coherent spatial patterns of statistically significant changes emerge, particularly an increase in warm summer nights, a decrease in the number of frost days and a decrease in intra-annual extreme temperature range. All but one of the temperature-based indicators show a significant change. Indicators based on daily precipitation data show more mixed patterns of change but significant increases have been seen in the extreme amount derived from wet spells and number of heavy rainfall events. We can conclude that a significant proportion of the global land area was increasingly affected by a significant change in climatic extremes during the second half of the 20th century. These clear signs of change are very robust; however, large areas are still not represented, especially Africa and South America.

DOI

10
Gitau W, Ogallo L, Camberlin Pet al., 2013. Spatial coherence and potential predictability assessment of intraseasonal statistics of wet and dry spells over Equatorial Eastern Africa.International Journal of Climatology, 33(12): 2690-2705.The aim of this study was to derive components of the intraseasonal rainfall variations from the daily rainfall in the Equatorial Eastern Africa region and assess their spatial coherence, a pointer to their potential predictability. Daily rainfall observations from 36 stations distributed over Equatorial Eastern Africa and extending from 1962 to 2000 were used. The March to May and October to December periods commonly referred to as the long and short rainfall seasons respectively were considered.Seasonal and intraseasonal statistics at the local (station) level were first defined. The stations were also grouped into near-homogeneous (sub-regional) zones based on daily rainfall. Similarly, seasonal and intraseasonal statistics were then derived at sub-regional level using three different approaches. Inter-station correlation coefficients of the intraseasonal statistics at local levels were finally computed and plotted as box-plots.For the two rainfall seasons, the two statistics showing the highest spatial coherence were the seasonal rainfall totals and the number of the wet days at sub-regional level. The local variance explained for these two variables, as an average over all the sub-regions, was more than 40%. At the bottom of the hierarchy were the mean rainfall intensity and frequency of dry spells of 5 days or more which showed the least coherence, with the local variance explained being less than 10% in each season. For each of the intraseasonal components of daily rainfall considered, the short rainfall season statistics were more coherent compared to the long rainfall season. Lag-correlations with key indices depicting sea-surface temperatures in the Pacific and Indian Oceans showed that the hierarchy between the rainfall statistics in the strength of the teleconnections reflected that of spatial coherence.

DOI

11
Gong Z S, He M, 2006. Relationship between summer rainfall in Yangtze River Valley and SSTA of various seasons.Meteorological Monthly, 32(1): 56-61. (in Chinese)

12
Hartmann H, Becker S, King L, 2008. Predicting summer rainfall in the Yangtze River basin with neural networks.International Journal of Climatology, 28(7): 925-936.The neural network algorithms proved to be capable of explaining most of the rainfall variability in the Yangtze River basin. For five out of six regions, our predictions explain at least 77% of the total variance of the measured rainfall. Copyright 08 2007 Royal Meteorological Society

DOI

13
He H, Liao X P, Lu Het al., 2016. Features of long-cycle drought-flood abrupt alternation in South China during summer in 1961-2014.Acta Geographica Sinica, 71(1): 130-141. (in Chinese)In this study, summer long-cycle drought-flood abrupt alternation index(LDFAI) is calculated based on the monthly precipitation data from May to August obtained from 110 weather stations in South China from 1961 to 2014. LDFAI, which compares the precipitation amounts under different disaster conditions, can reflect the features of drought- flood abrupt alternation in South China in summer. The spatial-temporal change of summer LDFAI in South China have been studied by rotated empirical orthogonal function analysis, trend coefficient estimation, linear trend analysis, t-test, and Mann-Kendall test. Results show the following:(1)the overall change trend of the average LDFAI in South China was not remarkable, whereas the LDFAI intensity exhibited significant periodic changes, including two strong periods and a weak period.(2) The summer LDFAI can be divided into five main spatial regions. The analysis of the data from the representative stations in different spatial regions showed that the summer LDFAI values in Region 1(North Guangdong and Northeast Guangxi) and Region 2(West Guangdong and Southeast Guangxi) demonstrated downward trends. The decline in Region 1 was significant, and an abrupt decline occurred in 1988. The summer LDFAI values in Region 3(East Guangdong) and Region 5(Hainan Island) showed significant upward trends.An abrupt rise occurred in Region 3 in 1980. The LDFAI in Region 4(West Guangxi) exhibited periodic change features.(3) As regards the interdecadal change features of the LDFAI for the study period, the LDFAI values in Region 1 and Region 2 declined, whereas the amount of precipitation in May and June increased over the years. In the 1990 s and 2000 s, The LDFAI values in Region 3 and Region 4 were high, and the precipitation amount in July and August were higher than that in other months. In Region 5, the LDFAI increased, and the precipitation in July and August increased over the years as well.

DOI

14
He H, Lu H, 2014. Characteristics of the sharp turn from drought to flood over Guangxi in summer of 2013.Tropical Geography, 34(6): 767-775. (in Chinese)By using the global reanalysis data provided by NCEP/NCAR, and the daily precipitation data at 88 national meteorological observational stations of Guangxi provided by Guangxi Meteorological Information Center, the phenomenon and its circulation background of the sharp turn from drought to flood in summer in 2013 over Guangxi were briefly analyzed. The conclusions are as follows: 1) In summer of 2013, the total precipitation over Guangxi was almost normal, but a sharp turn from drought to flood occurred. The mean precipitation from the 31 st to the 44 th pentad was 36% less than normal, while that from the 45 th to the 48 th pentad was 1.4 times more, and the kickpoint was at the 45 th pentad(the 3rd pentad in August). 2) The mean circulation from the 31 st to the 44 th pentad showed the typical characteristics of drought, such as in middle and high latitude areas of Eurasian, the western Pacific subtropical high was more intense and controlled Guangxi, the moisture flux divergence over south China was positive anomaly. And those from the 45 th to the 48 th pentad showed the obvious characteristics of flood over the lower troposphere, Guangxi was located south of the western Pacific subtropical high, and controlled by ITCZ, the wind fields over south China were cyclonic anomaly, the moisture flux divergence was negative anomaly, and convection enhanced. 3) The atmospheric circulation changed at the 45 th pentad. The western Pacific Subtropical high along 110 E jumped to north from 25 N to 30-35 N abruptly. The ITCZ moved northward with the Subtropical high, and remained stable along the line of Burma-south China-the Philippines. Guangxi was influenced by typhoons, and the weather rapidly turned from drought to flood.

15
Huang R, 2015. Research on evolution and countermeasures of droughts-floods abrupt alternation events in Huaihe River Basin [D]. Beijing: China Institute of Water Resources & Hydropower Research. (in Chinese)

16
Ji Z H, Shan H Y, 2015. Threshold diagnosis and hazard dangerousness evaluation for the disaster of drought-flood abrupt alternation in the middle and lower reaches of the Yangtze River.Resources and Environment in the Yangtze Basin, 24(10): 1793-1798. (in Chinese)Based on the related impact factors of the precipitation anomaly in summer in the Middle and Lower Reaches of the Yangtze River,15 atmospheric circulation indicators as the independent variables and the hazard dangerousness level of drought-flood abrupt alternation index as the dependent variable were selected to construct the nonlinear and non-parametric classification and regression tree(CART)model for the hazard evaluation.The time series of data is from 1954 to 2010.Results showed that Arctic Oscillation Indices in summer(AOI_SU)and spring(AOI_SP),Asia Meridional Circulation Index in spring(AMCI_SP),and Asia Polar Vortex Area Index in spring(APVAI_SP)were the four main impact factors,which were proved to be suitable for the hazard evaluation of disaster about drought- flood abrupt alternation through the CART model.The type of the disaster(flood to drought or drought to flood)and the predicted dangerousness level under different diagnosis conditions can also be obtained from the model.The diagnosis thresholds and dangerousness level as follows:1)When AOI_SU more than 1.11,the disaster of the quick turn from flood to drought is more apt to happen(dangerousness level is 1).2)When the AOI_SU is less than 1.11 and the AOI_SP is more than-1.11,the disaster of the abrupt alternation from drought to flood will occur(dangerousness level is 6).3)When the AOI_SU is less than1.11,the AOI_SP less than-1.11,and the AMCI_SP more than 61,the same situation will happen as the one before this,and the predicted level is 5.5.4)When the AOI_SU is less than 1.11,the AOI_SP less than-1.11,and the AMCI_SP less than 57,the abrupt alternation from flood to drought will happen,and the predicted level is 1.29.5)When the AOI_SU less than 1.11,the AOI_SP less than-1.11,the AMCI_SP between 57 and 61,and the APVAI_SP less than 172,the abrupt alternation from flood to drought will happen,and the predicted level is 2.1.The corresponding indicators from 2011 to 2013were selected to verify the final model through the comparison between the predicted values and actual levels,and the model was proved to be reliable for the close values.The CART proposed in this study provides a new method that can predict the hazard dangerousness level from the disaster of drought-flood abrupt alternation in the middle and lower reaches of the Yangtze River.

17
Langousis A, Kaleris V, 2014. Statistical framework to simulate daily rainfall series conditional on upper-air predictor variables.Water Resources Research, 50(5): 3907-3932.propose a statistical framework to generate synthetic rainfall time series at daily resolution, conditional on predictor variables indicative of the atmospheric circulation at the mesoscale. We do so by first introducing a dimensionless measure to assess the relative influence of upper-air variables at different pressure levels on ground-level rainfall statistics, and then simulating rainfall occurrence and amount by proper conditioning on the selected atmospheric predictors. The proposed scheme for conditional rainfall simulation operates at a daily time step (avoiding discrete approaches for identification of weather states), can incorporate any possible number and combination of predictor variables, while it is capable of reproducing rainfall seasonality directly from the variation of upper-air variables, without any type of seasonal analysis or modeling. The suggested downscaling approach is tested using atmospheric data from the ERA-Interim archive and daily rainfall measurements from western Greece. The model is found to accurately reproduce several statistics of actual rainfall time series, at both annual and seasonal levels, including wet day fractions, the alternation of wet and dry intervals, the distributions of dry and wet spell lengths, the distribution of rainfall intensities in wet days, short-range dependencies present in historical rainfall records, the distribution of yearly rainfall maxima, dependencies of rainfall statistics on the observation scale, and long-term climatic features present in historical rainfall records. The suggested approach is expected to serve as a useful tool for stochastic rainfall simulation conditional on climate model outputs at a regional level, where climate change impacts and risks are assessed.

DOI

18
Li C H, Li T, Lin A L, 2015. Relationship between summer rainfall anomalies and sub-seasonal oscillations in South China.Climate Dynamics, 44(1/2): 423-439.Sub-seasonal variability of summer (May–October) rainfall over South China exhibits two dominant timescales, one with a quasi-biweekly (QBW) period (10–2002days) and the other with an intraseasonal oscillation (ISO) period (20–6002days). A significant positive correlation (at a 9902% confidence level) was found between the summer precipitation anomalies and the intensity of the QBW and ISO modes. By examining the composite structure and evolution characteristics, we note that the QBW and ISO modes are characterized by a northwest-southeast oriented wave train pattern with a pronounced baroclinic vertical structure, moving northwestward. A marked feature is the phase leading of low-level moisture relative to convection. For the QBW mode, such a phase leading feature appears in both the strong and weak composites. However, for the ISO mode, this feature is only clearly seen in the strong composite. The high positive correlation between the summer precipitation and the sub-seasonal variability suggests that the summer mean state may exert a large-scale control on the sub-seasonal modes. It is found that when South China is anomalously wet, large-scale atmospheric conditions in the key QBW/ISO activity region are characterized by deeper moist layer, more convectively unstable stratification, and greater ascending motion. Such environmental conditions favor the growth of the QBW and ISO perturbations.

DOI

19
Li M, 2013. The causes of abruptly drought-flooding turn in the mid-lower reaches of Yangtze River during spring to summer in 2011 [D]. Beijing: Chinese Academy of Meteorological Sciences. (in Chinese)

20
Li X, Yuan D M, Yin Z Cet al., 2014. Preliminary analysis of sudden turn of drought and flood in the middle and lower reaches of the Yangtze River during 2011.Climatic and Environmental Research, 19(1): 41-50. (in Chinese)Using the sudden turn of drought and flood(STDF) in the middle and lower reaches of the Yangtze River during 2011, as an example, the relation between the STDF and circulation and sea surface temperature(SST) is studied. The reasons for the STDF are the following.(1) The atmospheric circulation quickly adjusted from strong winter monsoon to the station, called two troughs and one ridge.(2) The location and strength of the western Pacific subtropical high also changed quickly. The convergence of water vapor transport was obviously weaker in the middle and lower reaches of the Yangtze River from January to May. In June, the water vapor transport and budget changed and there was an obviously stronger center of water vapor convergence in the middle and lower reaches of the Yangtze River.(3) Because of the cyclonic anomalous circulation over the Tibetan Plateau, the convection is active and frequently moves to the middle and lower reaches of the Yangtze River. This convection strengthened the Mei-yu front and triggered rainfall five times.(4) The persistent La Ni顝竌 event and its variations created favorable background conditions by influencing the Walker circulation, the western Pacific subtropical high, and so on.

DOI

21
Li Y, Wang Y F, Dong M, 2009. Simulation of the Effects of the Preceding SST Anomalies over the Tropical Eastern Pacific on Precipitation to the South of the Yangtze River in June.Acta Meteorologica Sinica, 23(6): 691-700.

22
Liu G, Zhang Q Y, Sun S Q, 2008. The relationship between circulation and SST anomaly east of Australia and the summer rainfall in the middle and lower reaches of the Yangtze River.Chinese Journal of Atmospheric Sciences, 32(2): 231-241. (in Chinese)

23
Lu E, 2009. Determining the start, duration, and strength of flood and drought with daily precipitation: Rationale.Geophysical Research Letters, 36: L12707.1] Determinations of the start, duration, and strength of flood and drought are important in the operational monitoring and to decision-makers, but cannot be made through the commonly-used Standardized Precipitation Index. Physical considerations are provided in this note for measuring the daily flood and drought extent with precipitation. The rationale is that, because of the memory of the soil-land surface system, the flood and drought extent of a day is contributed by the precipitation of both the day and the earlier days, while the contribution of the earlier day precipitation is decayed due to the demands of the water balance, including the runoff, evapotranspiration, and percolation. The proposed index has the same unit and magnitude as precipitation, and is easy to calculate. It can be used to evaluate the interannual variabilities and long-term changes of the start, duration, and strength of flood and drought.

DOI

24
Luo S H, Jin Z H, Chen L T, 1985. The analysis of correlation between sea surface temperature in the Indian South-China Sea and precipitation in the middle and lower reaches of the Yangtze River.Chinese Journal of Atmospheric Sciences, 9(3): 314-320. (in Chinese)

25
Luo W, Zhang X, Deng Z Met al., 2013. Variation of the total runoff into Poyang Lake and drought-flood abrupt alternation during the past 50 years.Journal of Basic Science and Engineering, 21(5): 845-856. (in Chinese)In recent years,the low water level in Poyang Lake appeared earlier and lasted longer than before,and it continued to break the recorded historic lowest value,which caused the disadvantageous influence on the health of the wetland ecological system in Poyang lake. The total runoff into Poyang Lake is closely related with the water level change and drought-flood alternation,so the study on the variation and periodic law of the total runoff into Poyang Lake plays an important role in the protection of ecological system in Poyang Lake basin. Based on the daily time series of the total runoff into Poyang Lake from 1959 to 2009,the variation trends of the total runoff into Poyang Lake was analyzed by the linear regression analysis,the methods of the Mann-Kendall and the moving average. The results show that the annually total runoff into Poyang Lake has a non-significant upward trend. The wavelet analysis shows that there is the first main 19-yearly period in the annually total runoff time series. And the seasonally total runoff time series,divided into four seasonal time intervals,Jan to Mar,Apr to Jun,Jul to Sep and Oct to Dec,have several main time scales of periods with 31,19,17 and 14-yearly main periods. It is concluded that the 19-yearly main period of the annually total runoff time series is determined by the combination of the main periods of the four seasonally total runoff time series. Based on the analysis of the variation of the total runoff into Poyang Lake,the droughtflood abrupt alternation was analyzed by using the Long-cycle Drought-Flood Abrupt Alternation Index( LDFAI). The results show that the alternating cycle from drought to flood and from flood to drought exists between Apr to May and Jun to Jul in the long term. And into 21st century,the interval from drought to flood and from flood to drought is becoming shorter and shorter,which means the oscillation of the alternation of drought-flood is intensified.

DOI

26
Ma P H, Yang Y J, Liu T J, 2014. Cause analysis for the sharp turn from drought to flood in the middle and lower reaches of the Yangtze River during 2011.Meteorology and Disaster Reduction Research, 37(3): 1-6. (in Chinese)Some kinds of climate monitoring and diagnosis data provided by National Climate Center(NCC)were used to preliminarily analyze the reasons for the sharp turn from drought to flood(STDF)in the middle and lower Yangtze River during 2011. The results showed that firstly this STDF occurred under a special large-scale circulation background,the polar vortex was stronger in May 2011,while both the south Asia high and subtropical high were weaker. In June the polar vortex became weaker with smaller area,meanwhile the south Asia high strengthened and moved northwestward,the subtropical high strengthened and extended westward. Secondly,abnormal monsoon activity was an important factor to causing this STDF,the southwest monsoon was weaker in early May 2011 and interrupted for a time,then suddenly intensified in early June and extended northward to the middle and lower of Yangtze River. Thirdly,three main moisture transportation channels to China were noticeably weaker at the early stage and strengthened in early June,which caused abundant warm and humid air to gather and converge with cold air in the middle and lower of Yangtze River,and provided moisture conditions for the occurring of STDF. Fourthly,the La Ni event beginning in July 2010 and ending in April 2011 was probably an important external forcing condition for this STDF.

27
May W, 2004. Variability and extremes of daily rainfall during the Indian summer monsoon in the period 1901-1989.Global and Planetary Change, 44(1-4): 83-105.In this study, the variability and extremes of daily rainfall events during the Indian summer monsoon are investigated on the basis of a long observational record for the period 1901 1989, originating from the India Meteorological Department. The data are daily averages for 51 blocks, extending 2.5 in the meridional and zonal directions, except for the blocks on the Indian west coast with a width of 1 . Each block contains about seven stations. In addition to some basic properties such as the seasonal mean rainfall, frequency of wet days, and rainfall intensity on wet days, the variability of daily rainfall is described via gamma distribution. Extreme daily rainfall events, on the other hand, are investigated by quantiles and suitable theoretical extreme value distributions. The spatial distribution of the rainfall intensity on wet days during the Indian summer monsoon is very similar to the spatial distribution of the overall monsoon rainfall, with maxima on the west coast of the Indian peninsula and in Assam, and a small amount of precipitation in northwest India and in the southeastern part of the Indian peninsula. The gamma distribution is suitable for characterizing the time series of daily rainfall during the Indian summer monsoon by two parameters (i.e., the scale and the shape parameter). The spatial distribution of the scale parameter has a number of similarities with the spatial distribution of the rainfall intensity but shows an additional maximum in the westernmost part of India. A substantial amount of precipitation in association with heavy rainfall events is not only found on the west coast of the Indian peninsula and in Assam, where the maxima of the rainfall intensity are located, but also in the westernmost part of India and on the north coast of the Bay of Bengal. It is found that the generalized Pareto (GPA) distribution is generally better suited for describing the behaviour of heavy rainfall events during the Indian summer monsoon than the generalized extreme value distribution. This is a consequence of the substantial temporal variability of daily rainfall during the Indian summer monsoon, both on subseasonal and interannual time scales. A method is developed to objectively determine the locally varying thresholds that the daily data have to exceed in order to enter the fits to the GPA distribution. For this purpose, the GPA distributions for a set of locally varying thresholds are compared with a suitable quantile, an alternative description of heavy rainfall events.

DOI

28
McQuigg J, 1954. A simple index of drought conditions.Weatherwise, 7(3): 64-67.

DOI

29
QX/T 370-2017. Identification Standard for El Niño/La Niño Events.

30
Shan L J, Zhang L P, Chen X Cet al., 2015. Spatio-temporal evolution characteristics of drought-flood abrupt alternation in the middle and lower reaches of the Yangtze River Basin.Resources and Environment in the Yangtze Basin, 24(12): 2100-2107. (in Chinese)Based on the daily precipitation data from 75 rainfall gauging stations covering 1960-2012 in the middle and lower reaches of the Yangtze River Basin,the trend variation and temporal and spatial distribution characteristics of drought-flood abrupt alternation was analyzed by using long-cycle droughtflood abrupt alternation index(LDFAI)and short-cycle drought-flood abrupt alternation index(SDFAI).The results are as follows:1)Flood to drought abrupt alternation events are the main performance of the long-cycle drought-flood abrupt alternation(LDAF),as well as the short-cycle drought-flood abrupt alternation(SDFA);2)The north shore of the middle and lower reaches of the Yangtze River tend to occur drought to flood events,while the south shore tend to occur flood to drought events;3)Intensive short-cycle drought to flood abrupt alternation events during June to July in 1998 and 2011occurred in the north shore of the Yangtze River,and flood to drought events occurred in the south shore.However,intensity distribution of drought-flood abrupt alternation events during periods of May to June and July to August was just the opposite to that during June to July;4)In general,both the occurrence of long and short-cycle flood to drought abrupt alternation events shows decreasing trends,while the drought to flood abrupt alternation events are on the rise.It is worth noting that the trend during July to August is somewhat special.Drought to flood abrupt alternation events are increasing in the Xiangjiang River Basin,and increasing significantly in the Dongting Lake region,which may provide a preferable reference for the work of the Yangtze River flood and drought management.

31
She D X, Xia J, Song J Yet al., 2013. Spatio-temporal variation and statistical characteristics of extreme dry spell in Yellow River Basin, China.Theoretical and Applied Climatology, 112(1/2): 201-213.Drought is one of the most detrimental natural hazards in Yellow River Basin (YRB). In this research, spatio-temporal variation and statistical characteristic of drought in YRB is studied by using dry spell. Two extreme series, including annual maximum series (AMS) and partial duration series (PDS), are used and simulated with generalized extreme value (GEV), generalized Pareto (GP), and Pearson type III (PE3) distributions. The results show that the northern part is drier than the southern part of YRB. Besides, the maximum dry spell usually starts in October, November, and December. According to the trend analysis, mean maximum length of dry spell (MxDS) shows a negative trend in most stations. From the L-moments and Kolmogorov-Smirnov test method, it can be found that GEV model can better fit AMS while GP and PE3 can better fit PDS. Moreover, the quantiles from optimal model of AMS and PDS depict a similar distribution with values increases from south to north. The spatial distribution of scale and location parameters of GEV model for AMS shows a south-to-north gradient, while the distribution of shape parameter is a little irregularity. Furthermore, based on the linear correlation analysis, there is an evident linear relation between location and scale parameters with mean and standard variation of MxDS, respectively.

DOI

32
Shen B Z, Zhang S X, Yang H Wet al., 2012. Analysis of characteristics of a sharp turn from drought to flood in the middle and lower reaches of the Yangtze River in spring and summer in 2011.Acta Physica Sinica, 61(10): 109202. (in Chinese)We use global reanalysis data probided by NCEP/NCAR,precipitation data at 740 observational stations of China provided by the National Climate Center of the China Meteorological Administration,and grid data of precipitation in 2011 provided by National Meteorological Information Center to analyze the phenomenon of a sharp turn from drought to flood in the middle and lower reacher of Yangtze River in early June 2011,and the characteristics of its circulation background and briefly conclude as follows:1) the precipitation in the middle and lower reaches of the Yangtze River was less and its change rate was smaller than that of corresponding climatological normals from January to May in 2011,both surged suddenly in June,leading to the appearence of a sharp turn from drought to flood in June,and the kickpoint was at the 31st pentad(the 1st pentad in June);2) around the sharp turn,both flood water vapor flux and the space-time evolution characteristics of the first and the second modes of EOF analysis represented the transform of water vapor transport from a weaker state to a stronger one;3) before and after the turn,atmospheric circulation fields were significantly different.Before the sharp turn,winter monsoon in northern hemisphere was strong,and summer monsoon in southern hemisphere was weak,leading to the delay of monsoon tranform,stronger East Asian Trough,which went against warm-moist air blowing to the north.All of that eventually led to less rainfall in south China and occurance of this sharp turn.In early June,the period of turining,the circulation was adjusted quickly,which presented that the western Pacific subtropical high extended to west and jump to north abruptly, East Asian Trough kept strong and was maintained in the west,and blocking high located in the Okhotsk Sea weakened.Thus,cold and warm air converged in the middle and lower reaches of the Yangtze River and contributed to the occurance and continuation of precipitation.It is the main reason of the sharp turn from drought to flood in the middle and lower reaches of the Yangtze River.

DOI

33
Song J Y, Wang Z H, 2015. Interfacing the urban land-atmosphere system through coupled urban canopy and atmospheric models.Boundary Layer Meteorology, 154(3): 427-448.We couple a single column model (SCM) to a cutting-edge single-layer urban canopy model (SLUCM) with realistic representation of urban hydrological processes. The land-surface transport of energy and moisture parametrized by the SLUCM provides lower boundary conditions to the overlying atmosphere. The coupled SLUCM-SCM model is tested against field measurements of sensible and latent heat fluxes in the surface layer, as well as vertical profiles of temperature and humidity in the mixed layer under convective conditions. The model is then used to simulate urban land-atmosphere interactions by changing urban geometry, surface albedo, vegetation fraction and aerodynamic roughness. Results show that changes of landscape characteristics have a significant impact on the growth of the boundary layer as well as on the distributions of temperature and humidity in the mixed layer. Overall, the proposed numerical framework provides a useful stand-alone modelling tool, with which the impact of urban land-surface conditions on the local hydrometeorology can be assessed via land-atmosphere interactions.

DOI

34
Song J Y, Wang Z H, 2016. Evaluating the impact of built environment characteristics on urban boundary layer dynamics using an advanced stochastic approach.Atmospheric Chemistry and Physics, 16(10): 6285-6301.

DOI

35
Song J Y, Xia J, Zhang L Pet al., 2016. Streamflow prediction in ungauged basins by regressive regionalization: A case study in Huai River Basin, China.Hydrology Research, 47(5): 1053-1068.

DOI

36
Stowasser M, Annamalai H, Hafner J, 2009. Response of the South Asian summer monsoon to global warming: Mean and synoptic systems.Journal of Climate, 22(4): 1014-1036.Recent diagnostics with the Geophysical Fluid Dynamics Laboratory Climate Model, version 2.1 (GFDL CM2.1), coupled model's twentieth-century simulations reveal that this particular model demonstrates skill in capturing the mean and variability associated with the South Asian summer monsoon precipitation. Motivated by this, the authors examine the future projections of the mean monsoon and synoptic systems in this model's simulations in which quadrupling of CO2 concentrations are imposed. In a warmer climate, despite a weakened cross-equatorial flow, the time-mean precipitation over peninsular parts of India increases by about 10%09“15%. This paradox is interpreted as follows: the increased precipitation over the equatorial western Pacific forces an anomalous descending circulation over the eastern equatorial Indian Ocean, the two regions being connected by an overturning mass circulation. The spatially well-organized anomalous precipitation over the eastern equatorial Indian Ocean forces twin anticyclones as a Rossby wave response in the lower troposphere. The southern component of the anticyclone opposes and weakens the climatological cross-equatorial monsoon flow. The patch of easterly anomalies centered in the southern Arabian Sea is expected to deepen the thermocline north of the equator. Both these factors limit the coastal upwelling along Somalia, resulting in local sea surface warming and eventually leading to a local maximum in evaporation over the southern Arabian Sea. It is shown that changes in SST are predominantly responsible for the increase in evaporation over the southern Arabian Sea. The diagnostics suggest that in addition to the increased CO2-induced rise in temperature, evaporation, and atmospheric moisture, local circulation changes in the monsoon region further increase SST, evaporation, and atmospheric moisture, leading to increased rainfall over peninsular parts of India. This result implies that accurate observation of SST and surface fluxes over the Indian Ocean is of urgent need to understand and monitor the response of the monsoon in a warming climate. To understand the regional features of the rainfall changes, the International Pacific Research Center (IPRC) Regional Climate Model (RegCM), with three different resolution settings (0.500° 01— 0.500°, 0.7500° 01— 0.7500°, and 1.000° 01— 1.000°), was integrated for 20 yr, with lateral and lower boundary conditions taken from the GFDL model. The RegCM solutions confirm the major results obtained from the GFDL model but also capture the orographic nature of monsoon precipitation and regional circulation changes more realistically. The hypothesis that in a warmer climate, an increase in troposphere moisture content favors more intense monsoon depressions is tested. The GFDL model does not reveal any changes, but solutions from the RegCM suggest a statistically significant increase in the number of storms that have wind speeds of 1509“20 m s-1 or greater, depending on the resolution employed. Based on these regional model solutions a possible implication is that in a CO2-richer climate an increase in the number of flood days over central India can be expected. The model results obtained here, though plausible, need to be taken with caution since even in this 0904best0909 model systematic errors still exist in simulating some aspects of the tropical and monsoon climates.

DOI

37
Sun P, Liu C L, Zhang Q, 2012. Spatio-temporal variations of drought-flood abrupt alternation during main flood season in East River Basin.Pearl River, 33(5): 29-34. (in Chinese)Based on the monthly precipitation data from 32 rainfall gauging stations covering 1956-2009 in the East River Basin.The immediate flood-drought shift,e.g.shift from drought to flood and vice versa,was analyzed for flooding season using long-cycle drought-flood abrupt alternation index(LDFAI in short) and short-cycle drought-flood abrupt alternation index(SDFAI for short).The major results are: ①the high LDFAI and SDFAI can well describe the drought-to-flood shifts;while the low LDFAI and SDFAI can well describe the flood-to-drought shifts.SDFAI has better performance when compared to LDFAI with respect to description of flood-drought shifts;②LDFAI shows the highest probability of floods and SDFAI the highest probability of droughts.Moreover,the higher probability of floods can be detected during April to May when compared to other time intervals;③flood–to-drought shifts are decreasing in the middle and upper East River Basin from prior-flood season to posterior-flood season,from June to July.While drought–to-flood shifts are increasing in the middle and upper East River basin,while flood-to-drought shifts are increasing in the lower East River basin during periods of April to May,July to August and August to September.Moreover,the drought-flood shifts are subject to higher frequency in the upper East River basin when compared to those in the middle and the lower East River basin,which may imply much for the basin-scale water resources management across the East River basin in the changing environment.

38
Sun S Q, Ma S J, 2003. Analysis and numerical experiment on the relationship between the 1998 summer monsoon activities and SSTA in tropical regions.Chinese Journal of Atmospheric Sciences, 27(1): 36-52. (in Chinese)Using SVD analysis between OLR in summer and SST in the pre-winter and spring, the coupled modes are obtained to describe the relationship of the two fields. When SST in winter and spring presents a La Ni -like pattern, the corresponding anomalous OLR in the later summer will have a distribution with the positive areas over the Yangtze River valley and the tropical eastern Pacific (indicating a weaker convection), and a negative belt from the Indian Peninsula to the South China Sea and the western Pacific (indicating a stronger convection). However, for the El Ni pattern, it will be opposite. Taking the 1998 summer monsoon as a typical case, a numerical experiment has been done by using the T42L9 global spectral model developed by the Institute of Atmospheric Physics, Chinese Academy of Sciences (IAP/CAS) for investigating the influence of SSTA of that year on the severe flooding in the Yangtze River valley and the South China Sea Summer Monsoon (SCSM). The model has successfully simulated SCSM including its onset date and strength. Meanwhile, the features of tropical current, subtropical high in the western Pacific, Qinghai-Xizang high in the upper troposphere and the westerlies are all well simulated. More important is that the model gives a heavy rainfall around the Yangtze River valley in June and July. The percentage departure of rainfall reaches above 100%. Moreover, the situation is just opposite in the South China Sea region, where the summer rainfall is weak in 1998. All these are in good agreement to the observed results. Sensitivity experiments show that SSTA in the central and eastern Pacific plays a main role in the variation of the onset and strength of SCSM and the precipitation in the eastern China. Replacing the SST of 1998 in the tropical eastern Pacific region by the climate mean, we may find that SCSM becomes stronger than the observed, and the precipitation in the Yangtze River valley turns to be normal, hence, severe flooding would no longer appear. The relationship between SST of the western Indian Ocean and summer monsoon and rainfall in 1998 has also been investigated. It is found that SST of the western Indian Ocean plays a similar role to that of the eastern Pacific in the Asian summer monsoon activitives.

DOI

39
Tan J, Zhou F X, Hu D Xet al., 1995. The correlation between SST anomaly in the South China Sea and ENSO.Oceanologia et Limnologia Sinica, 26(4): 377-382. (in Chinese)

40
Tang M, Shao D G, Yao C L, 2007. Causes and countermeasures of sudden changing from drought to waterlogging in Huaibei region.Journal of China Institute of Water Resources and Hydropower Research, 5(1): 26-32. (in Chinese)The authors discuss the meaning and characteristics of Sudden Changing from Drought to Waterlogging(SCDW) in Huaibei region along the Huai River,which include specific occurrence time and weak overlapping in focal areas hit by drought and waterlogging.The Trend Analysis is applied to investigate the daily rainfall data,which spans more than 30 years and involves 10 representative stations in this area.The results show a growing trend in rainstorm frequency,rainfall intensity and water shortage probability over the past 10 years.It is concluded that the SCDW mainly results from two unsolved contradictions in Huaibei region.One is between its waterlogging-prone character and insufficient drainage capacity,and the other is between its drought-prone character and inadequate drought relief capacity,in addition to unfavorable climate change.In the end,nine countermeasures are put forward in terms of structural and non-structural measures.

41
Turner A G, Annamalai H, 2012. Climate change and the South Asian summer monsoon.Nature Climate Change, 2(8): 587-595.The vagaries of South Asian summer monsoon rainfall on short and long timescales impact the lives of more than one billion people. Understanding how the monsoon will change in the face of global warming is a challenge for climate science, not least because our state-of-the-art general circulation models still have difficulty simulating the regional distribution of monsoon rainfall. However, we are beginning to understand more about processes driving the monsoon, its seasonal cycle and modes of variability. This gives us the hope that we can build better models and ultimately reduce the uncertainty in our projections of future monsoon rainfall.

DOI

42
Wang F, Sun J L, Wu D X, 2014. Characteristics of drought-floods switch in the lower and middle reaches of Yangtze River in late spring and early summer of 2011.Periodical of Ocean University of China, 44(3): 10-16. (in Chinese)

43
Wang S, Tian H, Ding X Jet al., 2009. Climate characteristics of precipitation and phenomenon of drought-flood abrupt alternation during main flood season in Huaihe River Basin.Chinese Journal of Agrometeorology, 30(1): 31-34. (in Chinese)Huaihe River Basin locates in the north and south climate transition belt in China.The change rate of the inter-annual precipitation is obvious,and the flood-drought disaster occurs frequently in main flood season.Based on the precipitation data of 126 meteorological stations,the spatial-temporal distribution characteristics of the precipitation,the precipitation variations in typical dry-flood years and the phenomenon of drought-flood abrupt alternation during the main flood season in Huaihe River Basin were analyzed,by using EOF,linear-trend estimate and Mann-Kendall catastrophe test.The results indicated that the spatial distribution of the precipitation showed that there were more precipitation in the South,mountain areas and inshore areas,compared to the North,plain and inland.The drought and flood mainly occurred in the South.The inter-annual variation of precipitation was also significant,especially in the lately decade.The concentrating heavy rain mainly occurred in the first ten days of July.The phenomenon of drought-flood abrupt alternation frequently occurred.Its frequency had obviously increased since 2000.

DOI

44
Wu Z W, Li J P, He J Het al., 2006a. Occurrence of droughts and floods during the normal summer monsoon in the mid- and lower reaches of the Yangtze River.Geophysical Research Letters, 33(5): L05813.The daily precipitation data at 720 stations over China for the 1957-2000 period during summer (May-August) are used to investigate the droughts-floods coexistence (DFC) phenomenon during the normal summer monsoons. A droughts-floods coexistence index on seasonal timescale over the mid- and lower reaches of the Yangtze River (MLYRV) is defined to quantify this phenomenon and the associated ocean-atmospheric features in the strong DFC years are examined statistically. Results demonstrate that the occurrence of the strong summer DFC in the MLYRV is of an increasing trend for the period of 1957-2000. The strong summer DFC in the MLYRV is often accompanied by the anomalously subseasonal oscillation of the western Pacific subtropical high, the low-level westerly winds anomalies over the equatorial oceanic areas from the Indian Ocean to the western Pacific and the northward cross-equatorial winds anomalies near Sumatra and Somalia during summer, the strong Southern Hemisphere annual mode during the preceding November through January, high sea surface temperature in the oceanic areas from the Arabian Sea to the South China Sea, and El Ni o or the developing phase of El Ni o in the 6 preceding months. All these offer some predictive signals for the summer DFC in the MLYRV.

DOI

45
Wu Z W, Li J P, He J Het al., 2006b. Large-scale atmospheric singularities and summer long-cycle droughts-floods abrupt alternation in the middle and lower reaches of the Yangtze River.Chinese Science Bulletin, 51(14): 1717-1724. (in Chinese)The daily precipitation data at 720 sta- tions over China for the 1957―2003 period during summer (May―August) are used to investigate the summer subseasonal long-cycle droughts-floods abrupt alternation (LDFA) phenomenon and a long-cycle droughts-floods abrupt alternation index (LDFAI) in the middle and lower reaches of the Yangtze River (MLYRV) is defined to quantify this phenomenon. The large-scale atmospheric circula- tion features in the anomalous LDFA years are ex- amined statistically. Results demonstrate that the summer droughts-to-floods (DTF) in the MLYRV usually accompany with the more southward western Pacific subtropical high (WPSH), negative vorticity, strong divergence, descending movements develop- ing and the weak moisture transport in the low level, the more southward position of the South Asia high (SAH) and the westerly jets in the high level during May―June, but during July―August it is in the other way, northward shift of the WPSH, positive vorticity, strong convergence, ascending movements and strong moisture transport in the low level, and the northward shift of the SAH and the westerly jets in the high level. While for the summer floods-to-droughts (FTD) in the MLYRV it often goes with the active coldair mass from the high latitude, positive vorticity, strong convergence, ascending movement develop- ing and the strong moisture transport in the low level, and the SAH over the Tibetan Plateau in the high level, but during July―August it is often connected with the negative vorticity, strong divergence, de- scending movements developing and the weak moisture transport in the low level, the remarkable northward shift of the WPSH, the SAH extending northeastward to North China and the easterly jets prevailing in the high level over the MLYRV. In addi- tion, the summer LDFA in the MLYRV is of significant relationship with the Southern Hemisphere annual mode and the Northern Hemisphere annual mode in the preceding February, which offers some predictive signals for the summer LDFA forecasting in the MLYRV.

DOI

46
Yang S Y, Wu B Y, Zhang R Het al., 2013. Relationship between an abrupt drought-flood transition over mid-low reaches of the Yangtze River in 2011 and the intra-seasonal oscillation over mid-high latitudes of East Asia.Acta Meteorologica Sinica, 27(2): 129-143.AbstractNCEP/NCAR daily reanalysis data and Chinese daily gridded precipitation data are used to study the relationship between an aprupt drought-flood transition over the mid-low reaches of the Yangtze River in 2011 and the intraseasonal oscillation (ISO; 30 60 days) in the mid-high latitude meridional circulation of the upper troposphere over East Asia. The abrupt transition from drought to flood occurs in early June. The first two recovered fields of the complex empirical orthogonal function show that northward-propagating westerlies from low latitudes converge with southward-propagating westerlies from high latitudes over the mid-low reaches of the Yangtze River (MLRYR) in mid-late May. The timing of this convergence corresponds to the flood period in early-mid June. The ISO index is significantly and positively correlated with rainfall over the MLRYR. During the dry phase (before the transition), the upper troposphere over the MLRYR is characterized by cyclonic flow, easterly winds, and convergence. The regional circulation is dominated by a wave train with a cyclone over east of Lake Baikal, an anticyclone over northern China, and a cyclone over the MLRYR. During the wet phase, the situation is reversed. The configuration of the wave train during the dry phase favors the southward propagation of westerly wind disturbances, while the configuration of the wave train during the wet phase favors the development and maintenance of a pumping effect and sustained ascending motions over the MLRYR.

DOI

47
Yu S G, Miao Z M, Shao G Cet al., 2012. The crop-water level response model of rice under alternate drought and waterlogging.Journal of Food, Agriculture & Environment, 10(3/4): 1515-1519.

48
Yuan Y, Yang H, Li C Y, 2012. Study of El Niño events of different types and their potential impact on the following-summer precipitation in China.Acta Meteorologica Sinica, 70(3): 467-478. (in Chinese)According to the SSTA distribution during the peak phase of the El Nio event, the paper classified all the El Nio events since 1950 into Eastern Pacific type (EP), Central Pacific type (CP), and Mixed type. Different types of El Nio showed dissimilar developing features of SSTA, also with distinct responses of the Outgoing Longwave Radiation (OLR) anomalies and the anomalous Walker Cell over the tropical Indian-Pacific Ocean. In the following summer of an El Nio event, through influencing 850 hPa wind, water vapor transport, and 500 hPa western Pacific subtropical High, EP El Nio, CP El Nio, and Mixed El Nio would possibly induce Type III, Type II, and Type I of the summer rainfall pattern in China.

DOI

49
Zhang P, Wang F H, Wu Z Let al., 2008. Analysis of types of droughts-floods abrupt alternation in Huaibei Province.Express Water Resources & hydropower Information, 29(S1): 139-140, 151. (in Chinese)

Options
Outlines

/

〈 〉