Table of Contents Author Guidelines Submit a Manuscript
Corrigendum

A corrigendum for this article has been published. To view the corrigendum, please click here.

Advances in Meteorology
Volume 2017, Article ID 3197435, 15 pages
https://doi.org/10.1155/2017/3197435
Research Article

Spatiotemporal Variability and Trends of Extreme Precipitation in the Huaihe River Basin, a Climatic Transitional Zone in East China

1School of Urban and Environmental Science, Huaiyin Normal University, Huai’an, Jiangsu 223300, China
2School of Computer Science and Technology, Huaiyin Normal University, Huai’an, Jiangsu 223300, China

Correspondence should be addressed to Zhengwe Ye; moc.361@wzyfael

Received 4 October 2016; Accepted 28 December 2016; Published 30 January 2017

Academic Editor: Francesco Viola

Copyright © 2017 Zhengwe Ye and Zonghua Li. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Precipitation data from 30 stations in the Huaihe River basin (HRB), a climatic transitional zone in east China, were used to investigate the spatiotemporal variability and trends of extreme precipitation on multitimescales for the period 1961–2010. Results indicated that (1) the spatial pattern of the annual precipitation, rainy days, extreme precipitation, and maximum daily precipitations shows a clear transitional change from the south (high) to the north (low) in the HR; it confirmed the conclusion that the HRB is located in the transitional zone of the 800 mm precipitation contour in China, where the 800 mm precipitation contour is considered as the geographical boundary of the south and the north. (2) Higher value of the extreme precipitation intensity mainly occurs in the middle of the east and the central part of the basin; it reveals a relatively distinct west-east spatial disparity, and this is not in line with the spatial pattern of the extreme precipitation total, the sum of the precipitation in 95th precipitation days. (3) Annual precipitation of 22 stations exhibits increasing trend, and these 22 stations are located from the central to the northern part. There is no significant trend detected for the seasonal precipitation. The summer precipitation exhibits a larger change range; this might cause the variation of the flood and drought in the HBR. However, the increasing trend in winter precipitation may be beneficial to the relief of winter agricultural drought. Rainy days in 12 stations, mostly located in and around the central northeastern part, experienced significant decreasing trend. Extreme precipitation days and precipitation intensity have increasing trends, but no station with significant change trend is detected for the maximum precipitation of the basin. (4) The spatiotemporal variability in the HRB is mainly caused by the geographic differences and is largely influenced by the interdecadal variations of East Asian Summer Monsoon in eastern China. The output of this paper could provide references for the practical decision making and integrated basin management of Huaihe River basin.

1. Introduction

It is believed that the projected global climate changes have the potential to accelerate the global hydrological cycle, triggering changes in redistribution of precipitation and runoff which may further enhance uneven spatial and temporal distribution of the changing patterns of extreme weather events in both time and space (e.g., [1, 2]), such as precipitation, floods, and droughts. The frequency of great floods increased substantially during the twentieth century [3], and destructive floods observed in the 1990s have led to record-high material damage [4].

More and more concerns over extreme hydroclimatic events have been raised; for example, Burke et al. [5] discussed the extreme precipitation threshold selection based on climatological criteria in Europe. The total precipitation amounts exceeding 30 mm/day which occur in 4 or more provinces of the country are considered to be risky for floods in Bulgaria [6]. In the Great Plains of Eastern Nebraska, the hydrology and hydrometeorology of extreme floods are discussed [7]. Nyeko-Ogiramoi et al. [8] found that hydrometeorological extremes in the Lake Victoria basin are experiencing positive linear trends. In Africa, the heterogeneity of hydrometeorological extremes has been analyzed [9]. The maximum length of dry spell and the maximum 5-day rainfall are used to analyze the variability of rainfall extremes in the Philippines [10]. However, less research is focused on a climatic transitional zone, which could be influenced by global warming easily.

There are numerous studies indicating the changing features and underlying causes of precipitation extremes in China. Bordi et al. [11] suggested that the northern part of eastern China is experiencing dry conditions more frequently from the 1970s onwards. Liang et al. [12] revealed that observations from more than 78 meteorological stations are characterized by decreasing trends in both annual and summer precipitation in northeast China from 1961 to 2008. Zhai et al. [13] indicated that annual total precipitation has significantly decreased over southern northeast China, north China, and over the Sichuan basin but significantly increased in western China, the Yangtze River valley, and the southeastern coast in 1951–2000.

In river basins, Fischer et al. [14] showed that rainfall intensity has increased along the coastline and in the far west of the Zhujiang river basin from 1961 to 2007. Jiang et al. [15, 16] found that the significant precipitation rise detected in June, July, and August tends to aggravate the flood hazard between 1961 and 2000 in the Yangtze River. Zhang et al. [17] also found that number of rainy days with daily precipitation exceeding 95th and 99th percentiles and related precipitation intensities are in increasing tendency in summer during 1960–2005 in the Yangtze River basin. Liu et al. [18] confirmed that decreasing trends detected in most parts of the Yellow River basin are apparent during 1961–2006.

The Huaihe River basin (HRB), located in the middle eastern China, a flood-prone basin in China in which most floods are caused by the storms rains and the extreme precipitation on a basin scale, is strongly influenced by the northward/southward propagation of the East Asian Summer Monsoon (EASM) [19].

Liang et al. [20] suggest that most areas along the main stream of the HRB have a high probability of large precipitation under the condition that the rainy day has a dry preceding day. Xia et al. [21] pointed out that positive trend of annual maximum precipitation is detected at most of used stations of the HRB and that annual maximum precipitation series can be better fitted by GEV model. Svensson [22] found that, in the upper reaches of the HRB, heavy rainfall exceeding 50 mm/day occurs from late February to late November, and there is a higher concentration between late June and late August. Svensson [23] also pointed out that the connection between the temporal variation in rainfall intensity and the temporal succession of spatial patterns over three-day periods would be more useful in the construction of design rainfalls for the upper reaches of the HRB.

However, as a climatic transitional zone, the HRB is a sensitive area for the climate change; few studies have been raised on the multiaspects of the precipitation change in the HRB, especially the spatiotemporal changes of the extreme precipitation and the changing reason in a view of the climatic transitional zone. Thus, a better understanding of the spatial and temporal change of precipitation will be helpful for practical basin flood mitigation. Furthermore, the understanding of spatial and temporal pattern of precipitation is still insufficient for the HRB. Therefore, the objective of this paper is to detect the spatial and temporal trend and variability of annual precipitation, seasonal precipitation, and extreme precipitation in the Huaihe River basin.

2. Study Area

The Huaihe River basin, located in the eastern China, lies between 112°E and 121°E in longitude and between 31°N and 36°N in latitude. The 1,000-kilometer Huaihe River, with a total drainage area of 270,000 km2, originates in central China’s Henan province and runs through Anhui, Shandong, and Jiangsu province (Figure 1) [22, 2427]. Climatologically, the HRB is a region greatly influenced by the Asia Monsoon. Mean annual precipitation in the HRB ranges from 600 to 1400 mm with a long-term mean of 883 mm. As the 800 mm precipitation contour is believed as a very important line, it indicates the geographical boundary of south and the north in China. Thus, the HRB locates in the transitional belt of the south and the north [24, 2830], and it is easier and more sensitive to the impact of the climate change. Anomalies associated with the Meiyu (Plum Rain) flooding season often cause basin wide floods. The HRB is the most flood-prone basin in China, it has a long history of flooding over many centuries [2426]. In 2003, heavy floods along the HRB claimed at least 16 lives and caused direct economic losses of 2.2 billion US dollars and an inundated area of  ha [31].

Figure 1: Topographical sketch map of the Huaihe River basin with 30 weather stations.

3. Data and Methodology

For this study, climatic data from the China Meteorological Administration (CMA) were gathered from 30 meteorological stations located throughout the HRB. The time series are 50 years long (from 1961 to 2010). Figure 1 provides the locations of these meteorological stations. Monthly data were derived from daily data, and the annual data were derived from the monthly data. The 50-year period investigated was considered long enough to ascertain reliable climatic conclusions for which to reveal the true state of temporal precipitation changes that have occurred in the HRB.

To provide insight into the variability of extreme precipitation, the year was divided into four seasons: winter (December, January, and February: DJF), spring (March, April, and May: MAM), summer (June, July, and August: JJA), and autumn (September, October, and November: SON).

The indicators and trends were spatially interpolated using the Spline method in the Geographical Information System ArcGIS. Spline interpolation is widely used in the field of climate change [32]. Spline interpolation is a form of interpolation using a special type of piecewise polynomial called a spline.

In order to analyze and describe the temporal and spatial distribution of precipitation changes, 11 indicators were created. Most indicators were defined on fixed terms or thresholds predetermined by CMA or international research standards; similar indicators have been used in various studies (e.g., Liu, 2008; [33]).

The peaks over threshold indicators (P95 in this paper) were calculated from daily precipitation records for the time period 1961 to 2010. The approach using percentile values can give insights to the intensity of local changes of climate pattern and thus a better comparability between single stations [33]. The precipitation metrics are defined in Table 1.

Table 1: Definition of precipitation indices.
3.1. Mann-Kendall Test

The trends of the various indicators have been calculated by applying the nonparametric Mann-Kendall test to the data of the weather stations. The Mann-Kendall (MK) test is one of the most important statistical methods commonly used for detecting a trend in hydroclimatic time series. In this paper, the MK test is used to detect the trends of the indicators defined below. The MK test is a simple test and Mann-Kendall test is a rank-based test for a monotonic trend and is therefore resistant to the effects of outliers [3437]. The monotonic trend can be assessed by the test statistic . The can be calculated as follows:where and are the annual values in years and , respectively, and , if ; , if ; and , if . The variance of can be computed using the following equation:where is the number of ties of extent . For , the test is conducted using a normal distribution approximation and the standardized test statistic is computed as follows:

A positive indicates an upward trend while a negative indicates a downward trend [38]. Two-sided test at a level of significance of 10% are carried out in this study. The null hypothesis will be rejected if , where is the point on the normal distribution that has a probability of exceedance of 0.05, that is, 1.96.

3.2. Theil-Sen’s Slope Estimator Method

In this paper, the magnitude of a trend is estimated using the slope calculated by Theil-Sen’s estimator, which is a robust estimate that has been widely used in identifying the slope of the trend line in hydroclimatic time series. The Theil-Sen’s Slope estimator is robust against nonnormal distribution, missing data, and extreme outliers [39, 40]. Theil-Sen’s Slope estimator method for slope estimation requires a time series of data and calculates the slope as a change in measurement per change in time.

If a linear trend is present in a time series, then the true slope (change per unit time) can be estimated by using a simple nonparametric procedure developed by Sen [39]. The slope estimates of pairs of data are first computed by , for , where and are data values at times and , respectively. The median of these values of is Theil-Sen’s slope estimator of slope. If is odd, then the slope estimator is computed by , and if is even, then Theil-Sen’s slope estimator is computed by . Finally, is tested by a two-sided test at the % confidence interval and the true slope may be obtained by the nonparametric test [41].

4. Results and Discussion

4.1. Variation of Annual Precipitation

Annual precipitation (PA) change in the HRB indicates that the annual precipitation varies greatly throughout the study period with a maximum value of 1240 mm in 2003 and the minimum 598 mm in 1978 (Figure 2(a)); the slope of the annual precipitation is 5.56 mm per decade.

Figure 2: Temporal annual precipitation (a) and rainy days (b) for the Huaihe River basin.

The mean annual precipitation of the analyzed 50 years is 896 mm for the entire basin; this is in accordance with the proved conclusion that the Huaihe River is located in the transitional zone for the 800 mm precipitation contour in China [25]. As for the decadal change of PA, the decade of 2001–2010 witnessed the highest precipitation amount at 927 mm (Table 2); the big flood of the years 2003, 2005, and 2007 could be explained by this highest precipitation decade. However, the decade of 1981–1990 met the lowest annual precipitation in the HRB.

Table 2: Decadal change of different indicators concerning the precipitation in the HRB.

Monthly area-averaged precipitation over 100 mm occurs in all months from June to August which is the rainy season (flood season) in the HRB, and the summer precipitation (467 mm) has the most portion for the annual precipitation, which accounts the 52.1 percent of the whole year. Also, the precipitation in July contributes the largest portion of annual precipitation in HRB, and the spring precipitation is 185 mm; the autumn precipitation is 174 mm and the winter precipitation is 70 mm. The visualization of the monthly precipitation from 1961 to 2010 also confirmed the temporal monthly precipitation distribution pattern of the basin (Figure 3); it exhibits a tendency that precipitation in July tends to be much higher from 2003 to 2010.

Figure 3: The visualization of temporal monthly precipitation distribution pattern of the HRB.

Figure 4(a) illustrates the spatial distribution of the station based mean annual precipitation (MPA) in the HRB; it can be seen that MPA differs across the entire basin which shows a clear transition from the southern part (high) to the northern part (low). On average from 1961 to 2010, the MPA is above 1050 mm along the southeastern coastline region and the MPA below 650 mm is in the northwestern part; the maximum annual precipitation amounting to 1350 mm occurs mostly in the mountainous southwestern part. The distribution of MPA could also be explained by the fact that the Huaihe River is located in climatic transitional zone [13, 21].

Figure 4: Spatial pattern of mean annual precipitation (a) and its slope and trend (b); rainy days (c) and its slope and trend (d).

Following the result of Theil-Sen slope estimator and Mann-Kendall monotonic trends test applied to the precipitation time-series, it can be perceived that most parts of the HRB show positive slopes with the largest occurring in the central-western plain area (Figure 4(b)); however, negative slopes can be found in the region along the coastline where the largest negative slope is in the mountainous northeastern part.

No statistically significant trend in PA can be observed for the 30 stations tested, but it is apparent that most stations (22 stations) located from the central to the northern part exhibit increasing trend implying that the most stations with less annual total precipitation shows an upward trend though not significantly. This trend change might be the result of the northward propagation of the Meiyu belt influenced by the interdecadal variations of East Asian summer monsoon in eastern China [15, 16, 19].

For basin based PA in the HRB, an insignificant upward trend can be observed, its Mann-Kendall statistic is 0.22. Results of Mann-Kendall detection and Theil-Sen’s slopes for monthly precipitation from January to December are listed in Tables 3 and 4, respectively. Only 4 months (April, September, October, and November) show a downward trend while the rest of the 8 months show an upward trend; this change trend in the monthly precipitation implies that it could be drier in Autumn and the other seasons could be much wetter. The slope in JJA especially is much bigger (Table 4); it means there could be an increasing risk of flood in the HRB.

Table 3: Mann-Kendall statistic values for monthly precipitation.
Table 4: Theil-Sen’s slope for monthly precipitation.
4.2. Variation of Rainy Days in the HRB

A negative linear trend was detected in the basin based average rainy days (Figure 2(b)) where the largest rainy days (over 124) occurred in 1964 while the least rainy days (75) were in 1995. The Mann-Kendall test result for the basin based rainy days is −2.23, a significant negative trend for the entire basin from 1961 to 2010. In consideration of the aforementioned upward trend in annual precipitation for the HRB, it might be concluded that there could be an increase in precipitation intensity for the most part of the basin as there is a decreasing trend of rainy days in the HRB; Ren et al.’s recent research also shows a similar result [42].

For the decadal change of RD, the decade of 1961–1970 has the highest rainy days (99 days), and the decade of 1991–2000 met the lowest rainy days (91 days) in the HRB (Table 2).

Spatially, average rainy days (RD) distribution is presented in Figure 4(c). It can be found that RD varies strongly across the basin. The distribution of annual rainy days is in line with the distribution of the annual precipitation, which means the more precipitation the region has, the more rainy days it covers. Spatial distribution of RD also exhibits a south-north transition. Stations with the highest number of rainy days (up to 139 days/year) occur in the southwestern stations located in Mt. Dabieshan. The lowest number of rainy days (around 73 days/year) can be found at stations in the far north part of the basin.

Slope and Mann-Kendall trends test of the RD is given in Figure 4(d). It depicts that the entire area is undergoing a prevalent negative slope with the largest slope in the central-northeastern part. Decreasing trend can be seen in all analyzed stations for the rainy days throughout the studying period where 12 stations experienced a significant decreasing trend; these stations are mostly located in and around the central northeastern part with an exception in the southeastern area. This implies there might be less rainy days in the north part; however, as aforementioned that precipitation in most north part are increasing (Figure 4(b)), it means that the precipitation intensity could be heavier.

4.3. Variation of Seasonal Precipitation
4.3.1. Spatial Changes of Seasonal Precipitation

Seasonal precipitations of the basin are calculated based on the monthly precipitation data of the 30 stations in the HRB. For , its spatial distribution is demonstrated in Figure 5(a). It is clear to find that there is an obvious south-north transition in the spatial distribution of ; the maximum (above 350 mm) appears in the southern part with a decreasing gradient to the northern area where it meets minimum (about 90 mm) in the north border.

Figure 5: Spatial distribution of mean spring precipitation and (a) its slope and trend (b); summer precipitation (c) and its slope and trend (d); mean autumn precipitation (e) and its slope and trend (f); and mean winter precipitation (g) and its slope and trend (h).

As main precipitation is concentrated during summer, the flood (rainy) season, , shows a relatively more complicated spatial disparity in general (Figure 5(c)). Two main parts with much higher precipitation amount can be seen; one is in the eastern area which is in the lower reaches of the HRB close to the Yellow sea which is the main influencing area of the Meiyu Belt, and the other is in the southwestern mountainous part; the upper reaches of the HRB where there might be more mountainous effect resulting in more precipitation. In contrary, the central and south-northwestern area experienced less precipitation in summer. Basin based ranges from 673 mm to 308 mm in the HRB; is 467 mm on average in the analyzed 50 years which amounts to above 52% of the average annual precipitation.

Similar spatial precipitation distribution can be observed in autumn where the area of high precipitation enlarges to the most part of the southwestern dominated by the mountainous area (Figure 5(e)). However, precipitation amount in autumn is lower (higher) than that of in spring (winter).

Similarly, precipitation in winter shows the same spatial pattern as that in spring, but the precipitation amount in winter is less (Figure 5(g)). The largest (least) precipitation in winter is above 150 mm (only 17 mm).

In general, it can be summarized that spatial distribution of the slopes of seasonal precipitation varies in different season; the slopes of precipitation in spring, autumn, and winter show a relatively smaller change magnitude while the summer precipitation exhibits a larger change range. It is the larger change range in summer precipitation that causes the variation of the flood and drought in the HBR. Furthermore, the spatial pattern of , , and is quite similar, while pattern is more complicated.

4.3.2. Trend Changes of Seasonal Precipitation

An observed slope at −2.993 mm/10a in spring precipitation of the entire basin is shown in Figure 6(a) indicating a negative small linear trend. Spring precipitation shows the highest amounts (above 349 mm) in 1998 and the lowest amounts (below 44 mm) in 2001. Negative slope of precipitation can be found in almost the entire basin in spring with an exception in the northeastern mountainous area where positive slope of spring precipitation occurred (Figure 5(b)). No significant trend can be detected for the spring precipitation during the period from 1961 to 2010. Totally, 12 (18) stations are found to be nonsignificant decreasing (increasing) trend, stations with negative trend occurred mostly in the east and the west part, while stations with positive trend occurred mostly in the middle and the north of the basin.

Figure 6: Temporal changes of spring precipitation (a), summer precipitation (b), autumn precipitation (c), and winter precipitation in the Huaihe River basin.

Figure 6(b) shows the positive linear trend of the summer precipitation of the basin; a larger slope at 12.61 mm/10a can be seen. The highest amounts of summer precipitation (above 731 mm) occurred in 2003 and the lowest amounts (below 253 mm) in 1966. However, slope of summer precipitation shows an east-middle-west spatial variation which is not in line with that of the spring precipitation (Figure 5(d)); larger negative slope can be seen along the coastline area, and the larger positive slope gradually increases from the middle to western parts of the basin where the largest positive slope can be found in the central of the northwestern part and at the corner of the southeastern part of the Mt. Dabieshan.

The basin based slope of autumn precipitation is −10.195 mm/10a (Figure 6(c)); a decreasing trend is indicated. The highest amounts of autumn precipitation (above 308 mm) occurred in 1984 and the lowest amounts (below 45 mm) in 1998. The slope of autumn precipitation is contrary to the spatial pattern of the slope in summer precipitation (Figure 5(f)); negative slope is apparent in the whole basin in which lower negative slope is found in east and the west areas while higher negative slope is found in the middle and the central areas.

The basin averaged slope change for winter precipitation is 3.077 mm/10a in winter in the HRB (Figure 6(d)). Winter precipitation shows the highest amounts (above 120 mm) in 1968 and the lowest amounts (below 15 mm) in 1967. On the contrary, slope of winter precipitation exhibits a positive value in the whole basin in comparison to the slope of autumn precipitation (Figure 5(h)), but there are exceptions with positive slope in the far northeastern corner and the area in the west Mt. Tongbaishan.

And for the decadal change of seasonal precipitation, both the summer precipitation and the winter precipitation get the highest precipitation amount (542 mm and 69 mm, resp.) in the decade of 2001–2010 (Table 2). However, the decade of 2001–2010 met the lowest autumn precipitation in the HRB, while the spring precipitation shows a higher amount in the decade of 1961–1970. The decade of 1981–1990 seems to be a lower seasonal precipitation for spring precipitation, summer precipitation, and winter precipitation, where the autumn precipitation is an exception which has a higher precipitation amount on the contrary.

Temporally, results of Mann-Kendall detection for the spring, summer, autumn, and winter precipitations in the entire basin of the HRB are 0.03, 1.34, −1.77, and 1.42, respectively; no significant trend can be found. In consideration of the aforementioned fact that most precipitation occurs in summer, the upward trend in summer might increase the flood risk in the HRB. However, increasing trend in winter precipitation may be beneficial to the relief of winter agricultural drought in the western and the northern area of the basin in which most droughts appeared.

Spatial pattern of trend in summer precipitation is in nice accordance with its slope distribution; declining trend can be seen in the coastline area, and the rest part show an increasing trend in which 2 of 30 stations are significantly increased; they are in the central western area and the far southwestern mountainous area. Most part of the basin show a downward trend in autumn precipitation, in which 2 of 30 stations, in the southwestern mountainous area, show statistically significant decreasing trend at the 5% confidence level. However, most stations (28 of 30) show an upward trend in winter where 2 stations show significant increasing trend in the south part. It implies that the entire HRB has experienced more precipitation in winter season.

4.4. Variation of Extreme Precipitation
4.4.1. Changes of Extreme Precipitation Days

For extreme precipitation days (DP95) analysis, station based extreme precipitation threshold of P95 (TP95) distribution versus annual precipitation pattern is shown in Figure 7; it can be seen that the P95 threshold differs greatly; the threshold gradually increases from as low as 33 mm/day in the northwest to more than 46 mm/day in the northeast; that is, the spatial pattern of the P95 threshold shows a transitional spatial pattern from the southern part (low) to the most of northern part (high); this spatial disparity of the TP95 has a converse spatial pattern compared to the spatial variations of the annual precipitation (Figure 7); it is apparent that the more precipitation the area has, the lower threshold the area might have; this could be explained by the spatial distribution of rainy days and the annual precipitation of the basin. More rainy days and more precipitation in the southern might result in a lower P95 threshold, or vice versa.

Figure 7: Spatial distribution of extreme precipitation threshold of P95 versus annual precipitation.

However, a converse spatial disparity of the station based average number of DP95 can be concluded in comparison to the spatial variation of the TP95 (Figures 7 and 8(a)) and that more days of extreme precipitation (up to 7.2) occur in the southern part and the lowest number (below 3.8) can be detected in the northern part; this also shows a south-north transition and the average number of DP95 for the whole basin in the studying period is 5.1 days with a change magnitude from 2.45 to 8 days. The slope of number of P95 days exhibits a very small change magnitude in the whole HRB (Figures 8(b) and 9(a)), and the decadal change of the DP95 seems to be quite stable, averaged 5 days per decade in the HRB.

Figure 8: Spatial distribution of extreme precipitation days (a) and its slope and trend (b); extreme precipitation total (c) and its slope and trend (d).
Figure 9: Temporal changes of extreme precipitation days (a) and extreme precipitation total (b) in the HRB.

Spatially, the coastline area shows a negative slope and the rest area a positive slope. Significant trends of DP95 can only be found at 2 stations. Only 1 (1) station with significant positive (negative) trends is located in the central northwestern (northeastern corners) of the basin. For the rest of the stations, 18 (10) stations with upward trend are mostly scattered in the northern and western part (southern and eastern part). However, the Mann-Kendall test statistic for the basin averaged DP95 in the entire basin is 0.66, a positive trend. The slope of DP95 is 0.073 d/10a presented by Figure 9(a) where the maximum days of DP95 are 8 in 2003 and the minimum days of DP95 are 2.45 in 1978.

In summary, stations with higher DP95 in southeastern (southwestern) part are found to have negative (positive) change magnitudes and trends, while stations with low DP95 in both northeastern and northwestern part have positive change magnitudes and positive trend. This finding is in line with the observations made by Xia et al. [21]. This could result in an aggravation of the spatial divergence in both annual extreme precipitation and annual total precipitation. It might be concluded that there is an increasing extreme precipitation events in the northern and western area while the southeastern part as well as the coastal area might be less extreme precipitation events in the near future.

4.4.2. Changes of Extreme Precipitation Total

Spatial distribution of extreme precipitation total (PT95) is shown in Figure 8(c), the highest value (370–433 mm/year) can be found in the southwestern and the southeastern area of the basin, and lower values are located in the northern part of the basin. To some extent, it can be safely concluded that PT95 decreases from south to north generally; in other words, this south-north transitional spatial distribution of PT95 is quite similar to the spatial distribution of annual precipitation in the HRB.

For the change trend of the PT95 (Figure 8(d)), stations with decreasing trends are mainly located in the eastern part of the HRB, while the most stations with increasing trends are located in the middle to the west part of the HRB. Only 1 (1) station shows significant decreasing (increasing) trend that occurred in the east (west), in which it has the detected lowest (highest) slope, and the rest of the stations show no significant decreasing/increasing change trend; an east-west spatial pattern of slope of the PT95 can be safely concluded.

A slope at 6.448 mm/10a for the basin averaged PT95 is shown in Figure 9(b) a slight increasing trend generally for the whole basin. However, the decadal change of the PT95 has a relatively larger variation. The lowest amount of PT95 is in the decade of 1981–1990 and the maximum decadal of PT95 is in the decade of 2001–2010.

4.4.3. Changes of the Extreme Precipitation Intensity

Extreme precipitation intensity, the mean of the annual total precipitation in 95th precipitation days (PI95), has a changing magnitude at 54~76 mm; higher PI95 mainly falls in the middle east and the central part of the HRB, which reveals a relatively distinct west-east disparity Figure 10(a). This spatial distribution of PI95 is not in line with the spatial distribution of the PT95, it is because more (less) extreme precipitation days (DP95) occurred in the southern (northern) part, which might result in less (more) PI95 especially in the southwestern (the northeastern) areas of the basin.

Figure 10: Spatial distribution of the mean of PI95 (a) and its slope and trend (b); maximum daily precipitation (c) and its slope and trend (d).

The results of the trend test for extreme precipitation intensity (PI95) show that only one station is detected to have significant increasing which is in the southwest corner of the basin (Figure 10(b)). The number of nonsignificant negative trends stations is 11 which occurred mostly in the north part, while the rest of the 18 stations, nearly evenly spread in the basin, show nonsignificant positive trend. The slope of the PI95 shows the similar spatial distribution where the highest positive slope located in the central part.

For basin averaged PI95, it has a slope of 0.713 mm/10a (Figure 11), and the mean of basin averaged PI95 is 62.9 mm. The temporal changes of PI95 also shows that the maximum PI95 is 77.1 mm/day in 2000 and the minimum PI95 is 52.2 mm/day in 2001, both in the beginning of the 21st century. The observed increase of precipitation intensity can be explained by a decreasing number of rainy days and stable amounts of total precipitation on annual basis.

Figure 11: Temporal changes of PI95 in the Huaihe River basin.

For decadal change of the PI95, the first three decades of the analyzed period are quite stable with a PI95 at 62 mm/day, and the remaining two decades experienced a relatively higher PI95 in which the decade of 1991–2000 is the highest.

4.5. Variation of Maximum Daily Precipitation in the HRB

Station based maximum daily precipitation of a year (PM) varies strongly from as low as 24 mm/day to as high as 421 mm/day in the analyzed period (Figure 12). The majority value of PM is between 50 and 200 mm/day, and the highest value of PM (421 mm/day) occurred in station Zhumadian in 1982.

Figure 12: Temporal changes of maximum daily precipitation for each station in HRB. Stations numbers 1 to 30 stand for the stations Bengbu, Baofeng, Bozhou, Dangshan, Dongtai, Fuyang, Ganyu, Gaoyou, Gushi, Heze, Huaian, Huoshan, Juxian, Kaifeng, Linyi, Luan, Rizhao, Shangqiu, Sheyang, Shouxian, Suzhou, Xihua, Xinyang, Xuyi, Xuzhou, Xuchang, Yanzhou, Yiyuan, Zhengzhou, and Zhumadian, respectively.

For each of the 30 stations, mean PM also shows a strong spatial variation with a changing magnitude of around 30 mm/day. Stations with the highest value of mean PM (up to 112 mm/day) occurred in the south-western mountainous areas and the middle east lowlands of the basin. The lowest value of mean PM (around 83 mm/day) can be found at stations along the northern part and across the central-western areas of the basin. Thus, to some extent, a disparity with higher value in the east (lowlands) and west (mountainous areas) part and the lower value in the central and north part of the basin (inland) can be identified (Figure 10(c)). This is not in line with the annual precipitation which shows the higher amounts across the southern part and the lower amounts in the north areas (see above). However, it is quite similar to the spatial distribution of PT95 and the distribution pattern of the precipitation in summer; this is because main precipitation is concentrated during summer and higher values of PM occurred mainly in summer months.

The slope distribution of PM also shows a spatial disparity in the HRB (Figure 10(d)). Negative slope can be seen in the eastern parts and the far northwestern corner which are mountainous areas and the area around station Gushi, an area that is close to the foot of the Mt. Dabieshan, but positive slope prevails in the rest parts of the basin where higher positive slopes occur in along the south border and in the central part.

No station with significant change trend is detected for the PM of basin; totally 8 (22) stations show a decreasing (increasing) trend, where stations with decreasing trend mostly occurred in the eastern part while stations with increasing trend located mostly in the middle and the western areas in which 3 stations with decreasing trend scattered in the western part. It is interesting that most stations with higher value of mean PM in the eastern areas have negative trends.

5. Conclusion

The Mann-Kendall trend test and Theil-Sen’s slope estimator were used to investigate the spatiotemporal trends and variability of precipitation data from 30 stations in the Huaihe River on the annual, seasonal timescales, and the extreme precipitation for the period 1961–2010. Some interesting conclusions can be drawn as follows.

Mean annual precipitation is 896 mm in the HRB, in which the summer season contributes a large portion at 52.1%, and a clear south (high) to north (low) transition of annual precipitation can be found. Trend test in annual precipitation shows that most stations (22 stations) located from the central to the northern part exhibit increasing trend implying that the most stations with less annual total precipitation show an upward trend though not significantly. The basin based annual precipitation has an insignificant upward trend. However, the slopes in summer months are much bigger, which means there could be an increasing risk of flood in summer in the HRB.

Spatial distribution of rainy days also exhibits a south-north transition pattern, and the entire area is undergoing a prevalent negative slope. 12 stations, mostly located in and around the central northeastern part, experienced significant decreasing trend implying that there might be less rainy days and that the precipitation intensity could be heavier in consideration of the increasing trend of the annual precipitation in most north part.

Spatial pattern of precipitation in spring, autumn, and winter is quite similar, while summer precipitation pattern is more complicated. At the same time, the slopes of precipitation in spring, autumn, and winter show a relatively smaller change magnitude while the summer precipitation exhibits a larger change range which might cause much more uncertainty and variation of the flood and drought in the HBR. For the seasonal precipitation, no observed significant trend can be found, but the upward trend in summer might increase the flood risk in the HRB in consideration of the fact that most precipitation occurs in summer. However, increasing trend in winter precipitation may be beneficial to the relief of winter agricultural drought in the western and the northern area of the basin in which most droughts appeared.

Interestingly, we find that the more precipitation an area has, the lower extreme precipitation threshold the area might have. The reason of this spatial disparity is that there are more rainy days in the southern part of the HRB; it might result in a lower extreme precipitation threshold and vice versa. The extreme precipitation also shows a south-north transitional spatial pattern, which is quite similar to the spatial pattern of annual precipitation aforementioned. The extreme precipitation days have an increasing trend in the northern and western area, while it decreased in the southeastern part as well as the coastal area.

As for the change of extreme precipitation intensity, higher extreme precipitation intensity mainly fall in the middle east and the central part of the basin; it reveals a relatively distinct west-east disparity. This is not in line with the spatial distribution of the extreme precipitation total, possibly because that more (less) extreme precipitation days occurred in the southern (northern) part. This may lead to less (more) extreme precipitation intensity especially in the southwestern (the northeastern) areas of the basin. Hence, the observed increase of precipitation intensity can be explained by a decreasing number of rainy days.

Station based maximum daily precipitation of a year varies strongly from as low as 24 mm/day to as high as 421 mm/day. Mean maximum precipitation also shows a strong spatial variation with a changing magnitude of around 30 mm/day. No station with significant change trend is detected for the maximum precipitation of basin; 8 (22) stations show a decreasing (increasing) trend, where stations with decreasing trend mostly occurred in the eastern part while stations with increasing trend located mostly in the middle and the western areas.

Spatial disparity of maximum daily precipitation is quite similar to the spatial distribution of extreme precipitation total and the distribution pattern of the precipitation in summer; this is because main precipitation is concentrated during summer and higher values of maximum precipitation occurred mainly in summer months.

In general, the findings in this paper show that the spatiotemporal variability in the HRB is mainly caused by the geographic differences and the transition phase of the east Asia summer monsoon. The spatiotemporal changes of the precipitation is largely connected with the location of the HRB, a transitional zone of the south-north, and is connected with the result of the change of the Meiyu belt influenced by the interdecadal variations of East Asian summer monsoon in eastern China [19, 42]. The output of this paper could provide references for the practical decision making and integrated basin management of Huaihe River basin.

Competing Interests

The authors declare that they have no competing interests.

Acknowledgments

This study was financially supported by the National Natural Science Foundation of China (Grant no. 41471425), the Philosophy and Social Science Fund of Education Department of Jiangsu Province (Grant no. 2016SJB630122). and the Fund of Science and Technology Bureau of the Huai’an (Grant no. HAS2015005-1). The meteorological data used in this study were collected from China Meteorological Administration (CMA), which is highly appreciated.

References

  1. Q. Zhang, C.-Y. Xu, Z. Zhang, and Y. D. Chen, “Changes of atmospheric water vapor budget in the Pearl River basin and possible implications for hydrological cycle,” Theoretical & Applied Climatology, vol. 102, no. 1-2, pp. 185–195, 2010. View at Publisher · View at Google Scholar · View at Scopus
  2. Q. Zhang, V. P. Singh, J. Li, and X. Chen, “Analysis of the periods of maximum consecutive wet days in China,” Journal of Geophysical Research: Atmospheres, vol. 116, no. 23, pp. 2053–2056, 2011. View at Publisher · View at Google Scholar · View at Scopus
  3. P. C. D. Milly, R. T. Wetherald, K. A. Dunne, and T. L. Delworth, “Increasing risk of great floods in a changing climate,” Nature, vol. 415, no. 6871, pp. 514–517, 2002. View at Publisher · View at Google Scholar · View at Scopus
  4. Z. W. Kundzewicz and H.-J. Schellnhuber, “Floods in the IPCC TAR perspective,” Natural Hazards, vol. 31, no. 1, pp. 111–128, 2004. View at Publisher · View at Google Scholar · View at Scopus
  5. E. J. Burke, R. H. J. Perry, and S. J. Brown, “An extreme value analysis of UK drought and projections of change in the future,” Journal of Hydrology, vol. 388, no. 1-2, pp. 131–143, 2010. View at Publisher · View at Google Scholar · View at Scopus
  6. L. Bocheva, T. Marinova, P. Simeonov, and I. Gospodinov, “Variability and trends of extreme precipitation events over Bulgaria (1961–2005),” Atmospheric Research, vol. 93, no. 1–3, pp. 490–497, 2009. View at Publisher · View at Google Scholar · View at Scopus
  7. Y. Zhang, J. A. Smith, and M. L. Baeck, “The hydrology and hydrometeorology of extreme floods in the Great Plains of Eastern Nebraska,” Advances in Water Resources, vol. 24, no. 9-10, pp. 1037–1049, 2001. View at Publisher · View at Google Scholar · View at Scopus
  8. P. Nyeko-Ogiramoi, P. Willems, and G. Ngirane-Katashaya, “Trend and variability in observed hydrometeorological extremes in the Lake Victoria basin,” Journal of Hydrology, vol. 489, pp. 56–73, 2013. View at Publisher · View at Google Scholar · View at Scopus
  9. A. Tall, A. G. Patt, and S. Fritz, “Reducing vulnerability to hydro-meteorological extremes in Africa. A qualitative assessment of national climate disaster management policies: Accounting for heterogeneity,” Weather and Climate Extremes, vol. 1, pp. 4–16, 2013. View at Publisher · View at Google Scholar · View at Scopus
  10. M. Q. Villafuerte, J. Matsumoto, I. Akasaka, H. G. Takahashi, H. Kubota, and T. A. Cinco, “Long-term trends and variability of rainfall extremes in the Philippines,” Atmospheric Research, vol. 137, pp. 1–13, 2014. View at Publisher · View at Google Scholar · View at Scopus
  11. I. Bordi, K. Fraedrich, J.-M. Jiang, and A. Sutera, “Spatio-temporal variability of dry and wet periods in eastern China,” Theoretical & Applied Climatology, vol. 79, no. 1-2, pp. 81–91, 2004. View at Publisher · View at Google Scholar · View at Scopus
  12. L. Liang, L. Li, and Q. Liu, “Precipitation variability in Northeast China from 1961 to 2008,” Journal of Hydrology, vol. 404, no. 1-2, pp. 67–76, 2011. View at Publisher · View at Google Scholar · View at Scopus
  13. P. M. Zhai, X. B. Zhang, H. Wan, and X. Pan, “Trends in total precipitation and frequency of daily precipitation extremes over China,” Journal of Climate, vol. 18, no. 7, pp. 1096–1108, 2005. View at Publisher · View at Google Scholar · View at Scopus
  14. T. Fischer, M. Gemmer, L. Lüliu, and S. Buda, “Temperature and precipitation trends and dryness/wetness pattern in the Zhujiang River Basin, South China, 1961–2007,” Quaternary International, vol. 244, no. 2, pp. 138–148, 2011. View at Publisher · View at Google Scholar · View at Scopus
  15. T. Jiang, Z. W. Kundzewicz, and B. Su, “Changes in monthly precipitation and flood hazard in the Yangtze River Basin, China,” International Journal of Climatology, vol. 28, no. 11, pp. 1471–1481, 2008. View at Publisher · View at Google Scholar · View at Scopus
  16. Z. Jiang, S. Yang, J. He, J. Li, and J. Liang, “Interdecadal variations of East Asian summer monsoon northward propagation and influences on summer precipitation over East China,” Meteorology & Atmospheric Physics, vol. 100, no. 1–4, pp. 101–119, 2008. View at Publisher · View at Google Scholar · View at Scopus
  17. Q. Zhang, C.-Y. Xu, Z. Zhang, Y. D. Chen, C.-L. Liu, and H. Lin, “Spatial and temporal variability of precipitation maxima during 1960–2005 in the Yangtze River basin and possible association with large-scale circulation,” Journal of Hydrology, vol. 353, no. 3, pp. 215–227, 2008. View at Publisher · View at Google Scholar · View at Scopus
  18. Q. Liu, Z. Yang, and B. Cui, “Spatial and temporal variability of annual precipitation during 1961–2006 in Yellow River Basin, China,” Journal of Hydrology, vol. 361, no. 3-4, pp. 330–338, 2008. View at Publisher · View at Google Scholar · View at Scopus
  19. S. Dong, Y.-H. Ding, and Y.-J. Liu, “Decadal northward shift of the Meiyu belt and the possible cause,” Chinese Science Bulletin, vol. 55, no. 1, pp. 68–73, 2010. View at Google Scholar
  20. L. Liang, L. Zhao, Y. Gong, F. Tian, and Z. Wang, “Probability distribution of summer daily precipitation in the Huaihe basin of China based on Gamma distribution,” Acta Meteorologica Sinica, vol. 26, no. 1, pp. 72–84, 2012. View at Publisher · View at Google Scholar · View at Scopus
  21. J. Xia, D. She, Y. Zhang, and H. Du, “Spatio-temporal trend and statistical distribution of extreme precipitation events in Huaihe River Basin during 1960–2009,” Journal of Geographical Sciences, vol. 22, no. 2, pp. 195–208, 2012. View at Publisher · View at Google Scholar · View at Scopus
  22. C. Svensson, “Empirical orthogonal function analysis of daily rainfall in the upper reaches of the Huai River Basin, China,” Theoretical & Applied Climatology, vol. 62, no. 3, pp. 147–161, 1999. View at Publisher · View at Google Scholar · View at Scopus
  23. C. Svensson, “Occurrence, shape and movement of daily rainfall in the Huai Basin, China,” Hydrological Processes, vol. 13, no. 8, pp. 1197–1215, 1999. View at Publisher · View at Google Scholar · View at Scopus
  24. Z. W. Ye and Y. Xu, “Spatial trend of summer precipitation and its implications for flood risks in the Huaihe River Basin, China,” in Proceedings of the 18th International Conference on Geoinformatics, pp. 1–5, Beijing, China, June 2010. View at Publisher · View at Google Scholar · View at Scopus
  25. Y. Ning, M. Qian, and Y. T. Wang, Hydrological Handbook of Huaihe River, Science Press, Beijing, China, 2003 (Chinese).
  26. C. A. Lin, L. Wen, G. H. Lu et al., “Atmospheric-hydrological modeling of severe precipitation and floods in the Huaihe River Basin, China,” Journal of Hydrology, vol. 330, no. 1-2, pp. 249–259, 2006. View at Publisher · View at Google Scholar · View at Scopus
  27. W.-W. Hu, G.-X. Wang, W. Deng, and S.-N. Li, “The influence of dams on ecohydrological conditions in the Huaihe River basin, China,” Ecological Engineering, vol. 33, no. 3-4, pp. 233–241, 2008. View at Publisher · View at Google Scholar · View at Scopus
  28. Z. Yuan, D. H. Yan, Z. Y. Yang, J. Yin, and Y. Yuan, “Research on temporal and spatial change of 400 mm and 800 mm rainfall contours of China in 1961–2000,” Advances in Water Science, vol. 25, no. 4, pp. 494–502, 2014 (Chinese). View at Google Scholar · View at Scopus
  29. W. Zhang, S. M. Pan, L. G. Cao et al., “Changes in extreme climate events in eastern China during 1960–2013: a case study of the Huaihe River Basin,” Quaternary International, vol. 380-381, pp. 22–34, 2015. View at Publisher · View at Google Scholar · View at Scopus
  30. X. P. Li, X. M. Shi, and A. Wang, “The interdecadal spatial changes of typical rainfall contours in China,” Journal of Desert Research, vol. 36, no. 1, pp. 232–238, 2016. View at Google Scholar
  31. China Daily, 2006, http://www.chinadaily.com.cn/bizchina/2006-08/02/content_655445.htm.
  32. P. A. Hancock and M. F. Hutchinson, “Spatial interpolation of large climate data sets using bivariate thin plate smoothing splines,” Environmental Modelling & Software, vol. 21, no. 12, pp. 1684–1694, 2006. View at Publisher · View at Google Scholar · View at Scopus
  33. Z. Li, F.-L. Zheng, W.-Z. Liu, and D. C. Flanagan, “Spatial distribution and temporal trends of extreme temperature and precipitation events on the Loess Plateau of China during 1961–2007,” Quaternary International, vol. 226, no. 1-2, pp. 92–100, 2010. View at Publisher · View at Google Scholar · View at Scopus
  34. H. B. Mann, “Nonparametric tests against trend,” Econometrica, vol. 13, pp. 245–259, 1945. View at Publisher · View at Google Scholar · View at MathSciNet
  35. M. G. Kendall, Rank Correlation Methods, Charles Griffin & Company Limited, London, UK, 1975.
  36. R. M. Hirsch, J. R. Slack, and R. A. Smith, “Techniques of trend analysis for monthly water quality data,” Water Resources Research, vol. 18, no. 1, pp. 107–121, 1982. View at Publisher · View at Google Scholar · View at Scopus
  37. D. H. Burn and M. A. Hag Elnur, “Detection of hydrologic trends and variability,” Journal of Hydrology, vol. 255, no. 1–4, pp. 107–122, 2002. View at Publisher · View at Google Scholar · View at Scopus
  38. R. Modarres and V. de Paulo Rodrigues da Silva, “Rainfall trends in arid and semi-arid regions of Iran,” Journal of Arid Environments, vol. 70, no. 2, pp. 344–355, 2007. View at Publisher · View at Google Scholar · View at Scopus
  39. P. K. Sen, “Estimates of the regression coefficient based on Kendall's tau,” Journal of the American Statistical Association, vol. 63, pp. 1379–1389, 1968. View at Publisher · View at Google Scholar · View at Zentralblatt MATH · View at MathSciNet · View at Scopus
  40. H. Theil, “A rank-invariant method of linear and polynomial regression analysis,” in Henri Theil's Contributions to Economics and Econometrics: Econometric Theory and Methodology, vol. 23 of Advanced Studies in Theoretical and Applied Econometrics, pp. 386–392, Springer, Berlin, Germany, 1950. View at Publisher · View at Google Scholar
  41. E. Kahya and S. Kalayci, “Trend analysis of streamflow in Turkey,” Journal of Hydrology, vol. 289, no. 1–4, pp. 128–144, 2004. View at Publisher · View at Google Scholar · View at Scopus
  42. G. Ren, Y. Ren, Y. Zhan et al., “Spatial and temporal patterns of precipitation variability over mainland China: recent trends,” Advances in Water Science, vol. 26, no. 4, pp. 451–465, 2015 (Chinese). View at Publisher · View at Google Scholar