Geomorphology 85 (2007) 143 – 154

www.elsevier.com/locate/geomorph

Temporal and spatial trends of precipitation and river

flow in the Yangtze River Basin, 1961–2000
Tong Jiang a,⁎, Buda Su a , Heike Hartmann b
a
Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, China
b
Department of Geography, Justus Liebig University Giessen, Germany
Received 1 September 2004; received in revised form 15 March 2005; accepted 29 March 2006
Available online 1 September 2006

Abstract

The suspected impact of climate warming on precipitation distribution is examined in the Yangtze River Basin. Daily
precipitation data for 147 meteorological stations from 1961–2000 and monthly discharge data for three stations in the basin have
been analyzed for temporal and spatial trends. The methods used include the Mann–Kendall test and simple regression analysis.
The results show (1) a significant positive trend in summer precipitation at many stations especially for June and July, with the
summer precipitation maxima in the middle and lower Yangtze River basin in the 1990s; (2) a positive trend in rainstorm frequency
that is the main contributor to increased summer precipitation in the basin; and (3) a significant positive trend in flood discharges in
the middle and lower basin related to the spatial patterns and temporal trends of both precipitation and individual rainstorms in the
last 40 years. The rainstorms have aggravated floods in the middle and lower Yangtze River Basin in recent decades. The observed
trends in precipitation and rainstorms are possibly caused by variations of atmospheric circulation (weakened summer monsoon)
under climate warming.
© 2006 Elsevier B.V. All rights reserved.

Keywords: Temporal and spatial trends; Precipitation; River flow; Yangtze

1. Introduction sited in the middle to high latitudes of the northern

hemisphere in the last half of the 20th century (IPCC,
The spatial and temporal redistribution of precipita- 2001b), these included stations from the USA (Karl et al.,
tion, which is the direct source of flood and drought 1996; Karl and Knight, 1998; Groisman et al., 1999),
hazards, is one of the major impacts of global warming Europe (Brunetti et al., 2000; Osborn and Jones, 2000),
(IPCC, 2001a). As the water-holding capacity of the and East Asia (Iwashima and Yamamoto, 1993; Zhai et
atmosphere increases with temperature, warmer tem- al., 1999), etc. Impacts of precipitation change on
peratures suggest a higher potential of intense precipi- hydrological processes have been examined by modeling
tation (Kundzewicz, 2005). An increase in the frequency and observational analyses. These impacts vary across
of heavy precipitation events was reported by stations regions because of spatial differences in precipitation
change (Kwadijk and Rotmans, 1995; Dvorak et al.,
1997; Vogel et al., 1997; Arnell, 1998; Xu, 2000; Burn
⁎ Corresponding author. and Hag Elnur, 2002; Menzel and Bürger, 2002;
E-mail address: [email protected] (T. Jiang). Eckharadt and Ulbrich, 2003). At a global scale, Labat
0169-555X/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.geomorph.2006.03.015
144 T. Jiang et al. / Geomorphology 85 (2007) 143–154

Fig. 1. Location of the Yangtze River basin in China.

et al. (2004) proposed that intense evaporation over the Increased intensity in summer rainfall, particularly from
oceans has been closely associated with the variability of rainstorms covering large areas of the basin, initiate the
continental precipitation and evaporation, leading di- floods (CWRC, 2002). The growing population pressure
rectly to the change of global continental runoff. of the past decades, deforestation, lake reclamation, and
The present study focuses on the Yangtze River embankment construction on riverbanks all exacerbated
Basin. The Yangtze River is 6300 km long with a basin the flood situation (He and Jiao, 1998). Previous studies
area of 180 × 104 km2 (Fig. 1). Owing to great topo- using trend analysis revealed an increasing precipitation
graphic variability, the Yangtze River Basin can be in the winter months in the Yangtze River Basin between
divided into three physiographic regions (Chen et al., 1990–1999 (IPCC, 2001b) plus a significantly increas-
2001a): the Tibetan plateau, Mid-basin mountains, and ing trend in summer precipitation over the middle and
the eastern low plain (Fig. 2). lower basin in the second half of the last century (Zhai
The Yangtze is frequently flooded (Hu and Luo, et al., 1999; Becker et al., 2003; Gong and Ho, 2003;
1992), a pattern that has intensified since the 1990s (Yin Gemmer et al., 2004; Su et al., 2004). Evapotranspiration
and Li, 2001). As the basin is densely populated and has decreased significantly in the Yangtze River Basin
highly industrialized, the effects tend to be devastating. over the past 30 years, due to decreasing wind speed and

Fig. 2. Elevation of the Yangtze River Basin and the location of the observatory stations.
 T. Jiang et al. / Geomorphology 85 (2007) 143–154 145

radiation (Chen et al., 2005a). The increase in summer Monthly stream discharge data (CWRC, 2001) are
rainfall and the concurrent decrease in actual surface from three major hydrological stations on the Yangtze:
evaporation result in rising surface runoff and flood Pingshan in the upper basin (record length 1961–1987,
probability (Hu, 2005). Models for the sub-basins of the annual value until 2000), Yichang in the middle
middle and lower Yangtze show larger variability for the basin (1961–2000), and Datong in the lower basin
trend in runoff (Chen and Gao, 2005; Chen et al., 2005b). (1961–2000). Pingshan, Yichang and Datong cover
Subtropical monsoon climate prevails over most of upstream areas of about 48.5 × 104, 100.6 × 104 and
the Yangtze River Basin (Chen et al., 2001b). The 170.6 × 104 km2 respectively (Fig. 2). The annual runoff
spatial and temporal variabilities of precipitation in the at these three stations is about 1490 × 10 8 m 3
Yangtze River Basin are closely related to summer 4400 × 108 m3 and 8900 × 108 m3 in sequence.
monsoon activities that transport a huge amount of Homogeneity of precipitation and hydrological data
atmospheric moisture from the East and South China were calculated using a von Neumann ratio (N), cumu-
Sea to the basin. In a normal year, the summer monsoon lative deviations (Q / n− 0.5 and R / n− 0.5), and Bayesian
starts to influence the Yangtze River Basin in April procedures (U and A) (Buishand, 1982; Maniak, 1997).
retreating in October, generating floods over this period The dataset of 147 meteorological stations and three
(Chen et al., 2001b). Summer rainfall from June to hydrological stations tested by these three methods are
August, accounts for 45% of the annual total of about homogeneous at a significance level of 95%.
1100 mm (CWRC, 2002). Trend analysis methods used involve non-parametric
The present analytical study on the spatial and tem- Mann–Kendall test (Kendall, 1938, 1975) and linear
poral trends of precipitation and rainstorms is based on regression. The former, a rank-based procedure, was used
the available meteorological and hydrological data. to detect non-linear trends of monthly precipitation and
Correlations of long-term trends with river flow varia- runoff at Pingshan, Yichang, and Datong at various con-
tions at the three physiographic regions have been fidence levels. Linear trend analysis indicated the ten-
carried out. These help in further understanding the dency rate (slope) using least squares at the 95%
frequency, intensity, and distribution of floods in the confidence level. The Pearson correlation coefficient
Yangtze River Basin and may contribute to a better flood (two-tailed) was calculated to measure the relationship
management strategy in the future. between precipitation and runoff.
Inverse Distance Weighted (IDW) interpolation
2. Data and methodology method was used to interpolate the calculated trend
at pixels and to obtain a spatial distribution of the
The meteorological data used in this study, consisting precipitation trend. This method was chosen as individ-
of daily and monthly precipitation measurements for ual meteorological and stream gauging stations may
147 stations distributed fairly evenly in the Yangtze represent a finite area in a watershed (Gemmer et al.,
River Basin (Fig. 2), were collected from the National 2004).
Climatic Centre of the China Meteorological Adminis-
tration (CMA) (http://cdc.cma.gov.cn/index.jsp). The 3. Trends in observed precipitation
length of the data used in this study runs from 1 January
1961 to 31 December 2000 (Table 1). Daily precipita- 3.1. Monthly precipitation trends
tion at these stations is classified by intensity as
light (0.1–10 mm), moderate (10–25 mm), heavy Our results indicate remarkable differences among
(25–50 mm) and rainstorm (N 50 mm). the stations with negative and positive precipitation
trends for different months (Fig. 3). The highest number
Table 1
of stations with significant trends occur in January, with
National meteorological gauging stations used in this study 91 stations showing positive precipitation trends at 90%
confidence level, 76 at 95% confidence level, and 42 at
Yangtze River Upper Middle Lower
Basin region region region 99% confidence level. In March, June and July, 45
stations show significant positive trends at 90%
Longitude 90.6°–104.3° E 104.3°–111° E 111°–121.8° E
Mean altitude 3720 1560 400 confidence level, more than the number of stations
(above mean with negative trends. In contrast, in April, September and
sea level) (m) December more than 35 stations indicate negative trends
The number of 28 41 78 at the 90% confidence level. No obvious positive or
stations
negative trends occur for other months (Fig. 3).
146 T. Jiang et al. / Geomorphology 85 (2007) 143–154

Fig. 3. Number of stations with observed precipitation trends at the 90%, 95% and 99% confidence levels.
 T. Jiang et al. / Geomorphology 85 (2007) 143–154 147

Results of the Mann–Kendall test for monthly 3.3. Rainstorm trends

precipitation over the whole basin are shown in Fig.
4a. In addition to the inter-annual variability, the Mann– Linear regression analysis for both summer rainstorm
Kendall test also demonstrates that the regional precip- and precipitation of 1960–2000 suggest increasing
itation has spatially different trends. In the upper Yangtze trends in precipitation with time (Fig. 6). With the slopes
River Basin, 4 months (January, February, March and of the regression line being 1.32 mm a− 1 and 2.43 mm
July) show significant positive trends, and no significant a− 1, respectively (at 95% confidence level), the summer
negative trend is detected. In the middle basin, months rainstorm total is about 54% for summer precipitation.
showing significant negative trends (April, May, Sep- Linear regression tests on summer rainstorm frequen-
tember, October and December) exceed those with cy (Fig. 7a) and intensity (Fig. 7b) further suggest that an
positive trends (June and July). In the lower basin, increasing monotonic trend for the rainstorm frequency,
4 months (January, March, June and July) show signi- but not for rainstorm intensity. Positive trends of
ficant positive trends, but only 1 month (April) has a rainstorm frequency anomaly between 1991–2000 and
negative trend (Fig. 4b). Notably, positive trends of 1961–1990 (Fig. 8) are found in most of the lower
precipitation in June and July prevail in the middle and Yangtze River Basin. Negative trends occur mainly in
lower basins. the northwestern part of the upper-middle and northern
upper basin. The increase in rainstorm frequency is
3.2. Interpolated summer precipitation trends within 5 days/decade on the average, with a much higher
rate of increase of 10 days/decade in the northeastern
The interpolated precipitation demonstrates that basin.
positive trends in June dominate the eastern center
of the middle Yangtze River Basin and the northern 4. Long-term runoff trends
lower basin (Fig. 5a). Positive trends are also present
in July for the middle and the lower Yangtze, though At Yichang, the monthly runoff tested by Mann–
the location shifts from the northern part of the basin Kendall shows positive trends for February and July,
to the south. Positive trends also occur in the upper and negative trends for May, October and November
drainage basin around Pingshan (Fig. 5b). In August, (Fig. 9a). Positive trends prevail at Datong in January,
the area with positive trend is much smaller than in February, March, July, and August, and a negative trend
June and July, and the location of the precipitation is found only for 1 month (May) (Fig. 9b). Both stations
shifts to the southeastern lower Yangtze River Basin show significant positive trends in July.
(Fig. 5c). Summer runoff variation during last 40 years is plotted
Summer precipitation anomalies between 1991–2000 in Fig. 10, which shows that no appreciable linear trend is
and 1961–1990 indicate a large area with highest found for Yichang (Fig. 10a), but the positive trend is
positive anomaly (N 100 mm) in the southeastern detected at Datong, where the average rate of increment
part of the middle and lower Yangtze River Basin. of the summer runoff is 16.5 × 108 m3a 1 (Fig. 10b).
High values (50–100 mm) appear to the southwest of The annual maximum discharge at Yichang does not
Pingshan in the upper region of the Yangtze River. show a monotonic trend for the period 1961–2000. The
Negative anomalies occur mostly at the northern part of maximum value recorded in the last 40 years was
the upper-middle basin of the Yangtze (Fig. 5d). 70,000 m3 s− 1 in 1981 (Fig. 11a). In contrast, the

Fig. 4. Monthly area-average precipitation trends in the Yangtze Basin for 1961–2000. (a) trends for the whole basin, (b) trends of the different parts,
dashed lines denote thresholds at the 90%, 95% and 99% confidence levels.
148 T. Jiang et al. / Geomorphology 85 (2007) 143–154

Fig. 5. (a–c) Interpolated Mann–Kendall trends of precipitation in June, July, August for 1961–2000. (+) positive trends, (−) negative trends, (d)
Summer precipitation anomaly for the periods 1991–2000 and 1961–1990.
 T. Jiang et al. / Geomorphology 85 (2007) 143–154 149

Fig. 6. Linear trends of summer rainstorm (a) and summer precipitation (b) in the Yangtze River basin for 1961–2000.

maximum discharge at Datong has been rising since the Pingshan and in June at Datong. However, at Yichang,
1960s, and the highest value was recorded in the late monthly runoff occurs concurrently with area-averaged
1990s (84,500 m3 s− 1 in 1999). In comparison with the precipitation with the highest values in July (Fig. 12b).
1961–1990 period, the average increment in maximum
discharge at Datong in the 1990s is about 12,500 m3 s− 1 5.2. Rainstorms and flood discharge
(Fig. 11b).
Consecutive rainstorms (accumulated 3-day precipi-
5. Correlation between precipitation and runoff tation N 150 mm) and water levels of the Yangtze at
Yichang and Datong stations from 1 June to 10
5.1. Area-averaged precipitation and runoff September 1998 are shown in Fig. 13. A major flood
occurred in the summer of 1998. In June, rainstorms that
At Pingshan, correlation between monthly area- continued from 12 to 21in the lower Yangtze River Basin
averaged precipitation and runoff (Fig. 12a) indicates (Fig. 13b) and sporadic rainstorms in the middle basin
that monthly runoff from April to October is positively from 14 to the end of the month (Fig. 13a) raised the river
associated with monthly precipitation at the 95% stage. No flood peaks (above warning water level)
confidence level, but negatively between November during that time were reported in the main channel of the
and March except for December. At Yichang, Pearson Yangtze River. After June, however, there were eight
correlation coefficients pass the 90% confidence level flood peaks passed Yichang between July and August.
for all months, and even pass the 99% confidence level With no rainstorm after August, the flood peaks declined
for 10 months. At Datong, monthly runoff is positively in September (Fig. 13a). At Datong, the water level
associated with monthly precipitation at the 95% remained higher than the warning water level (14.5 m)
confidence level, except for September–November. from late June to early September (Fig. 13b).
Regarding the inter-annual distribution, monthly
runoff lags behind area-average precipitation at Ping- 6. Discussion
shan and Datong for 1 month. Highest runoff occurs in
August at Pingshan and in July at Datong, but the In this study, precipitation and runoff for the period
highest area-average precipitation is detected in July at 1961–2000 have been analyzed to determine temporal

Fig. 7. Linear trends of summer rainstorm frequency (a) and summer rainstorm intensity (b) in the Yangtze River basin for 1961–2000.
150 T. Jiang et al. / Geomorphology 85 (2007) 143–154

Fig. 8. Rainstorm frequency anomalies for the periods 1991–2000 and 1961–1990.

trends, spatial variability, and correlation between these the January precipitation has little influence on the river
two factors as well as between rainstorms and flood flow; its precipitation accounts for less than 3% of the
discharges. The Mann–Kendall test for the temporal annual total (CWRC, 2002).
distribution of precipitation trends reveals significant Locations with significant summer (June–August))
positive trends at the 90%, 95% and 99% confidence precipitation trends illustrate a shifting of a large amount
levels in January, March, June and July. Significant of precipitation from north to south and also towards
negative trends occur in April, September and December the southeastern parts of the middle and lower basin
(Fig. 3). These results are in good agreement with (Fig. 5a–c).This could happen from a change in
precipitation trends for the entire basin (Fig. 4a). atmospheric circulation since the 1980s (Qian et al.,
Averaged over the basin a positive trend is discernible 2003). The average summer precipitation for the period
in the summer months (significantly in June and July) and 1961–2000 in the Yangtze River Basin was 498.6 mm.
a significant negative trend in April and September. May The decadal precipitation anomalies from the average
is a deviant month when only 20 stations in the basin had were − 3.04% (1960s),− 5.25% (1970s),− 0.9% (1980s),
significant negative precipitation trends (Fig. 3). The and 9.2% (1990s). Summer precipitation was most
basin-average precipitation still shows a statistically sig- abundant in the 1990s. This has been reported in earlier
nificant negative trend (Fig. 4a), as the majority of the studies (CWRC, 2002; Gong and Ho, 2003). Anomaly
stations do not display a significant positive trend. of the summer precipitation for the periods 1991–2000
Variability in the spatial precipitation trends is evi- and 1961–1990 is on the average about 61 mm/decade
dent in the Yangtze River Basin (Fig. 4b). The same for the whole basin (Fig. 5d). The precipitation
positive trend of precipitation in July prevails across the increment over such a great area influenced the summer
entire basin and the in the middle and lower basin in runoff in the 1990s.
June (Fig. 4b). Any increase in summer precipitation The average summer precipitation from 147 meteo-
would aggravate the flood situation. On the other hand, rological stations in the basin is composed as 16% light

Fig. 9. Mann–Kendall analyses of the monthly runoff at Yichang station (a) and Datong station (b), dashed lines in the figure denote the 90%, 95%
and 99% confidence levels.
 T. Jiang et al. / Geomorphology 85 (2007) 143–154 151

Fig. 10. Summer runoff and its linear trends for 1961–2000 at Yichang station (a) and Datong station (b).

rain, 27% moderate rain, 27% heavy rain and 30% precipitation (Fig. 12a) as the winter precipitation is
rainstorm. For the period 1961–2000, the tendency rates snow that does not contribute immediately to the river
(slopes of linear trend) of these rain types are 0.06 mm flow. In the lower basin, time lags occur between
a− 1, 0.34 mm a− 1, 0.71 mm a− 1 and 1.3 mm a− 1 in the precipitation and runoff (Fig. 12b). Runoff regulation by
same sequence. Fig. 6 indicates that 54% of the summer human activity, e.g. reservoirs, that may affect the lag
precipitation increment in the Yangtze comes from has been discussed in earlier studies (Yin and Li, 2001;
rainstorms. As a significant increase in rainstorm Wang, 2005; Hartmann et al., 2005).
intensity has not been detected (Fig. 7b), the greater Annual maximum discharge usually occurs in the
rainstorm total is probably due to an increased frequency period from July to October in the middle basin and
(Fig. 7a). An increase in rainstorm frequency is most from May to September in the lower parts. About 60%
obvious in the southeastern part of the lower basin, and 70% of the annual maximum are recorded in July
where the anomaly for the periods 1991–2000 and for the middle and lower basins respectively. The annual
1961–1990 is up to 10 days/decade (Fig. 8). The maximum discharge of the lower river has been rising
frequent occurrence of rainstorms should lead to a rise in since 1960 (Fig. 11b). Fig. 13 illustrating the 1998 flood
flood probability. event shows that high discharges from the upper basin
The discharge of the Yangtze River comes mainly and the tributaries contributed to the higher water level
from rainfall. Monthly runoff trends for the period in July, which was raised by further increments from
1961–2000 at Yichang and Datong (Fig. 9) parallel the intense precipitation occurring afterwards. Thus, rainfall
corresponding area-average precipitation (Fig. 4). In in late June and early July over the Yangtze River Basin
summer, an increasing trend in discharge occurs at is of prime importance for flood forecast and control.
Datong (Fig. 10b) but not at Yichang (Fig. 10a). The advance and retreat of the monsoon determine to
Monthly area-average precipitations show a close a large degree the timing of the rainy season and the
relationship with the monthly discharge for most of amount of precipitation that falls across east China,
the months (Fig. 12a). In the upper basin, however, including the Yangtze River basin (Becker et al., 2003;
monthly discharge from November to March (excluding Gemmer et al., 2004). Substantial spatial and temporal
December, which receives the least precipitation in a variations in rainfall follow the changes in the duration
year) is negatively associated with the monthly and intensity of the monsoon. The stronger the summer

Fig. 11. Annual maximum discharges and their linear trends for 1961–2000 at Yichang (a) and Datong (b).
152 T. Jiang et al. / Geomorphology 85 (2007) 143–154

Fig. 12. Correlation of monthly areal precipitation and runoff between 1961 and 2000 at the three hydrologic stations, dashed lines in the figures
denote the 90%, 95% and 99% confidence levels.

monsoon, the more northerly travels the moisture. In 7. Conclusions

fact, the summer precipitation pattern shows a southward
shift in June–August (Fig. 5a–c) in the Yangtze River The precipitation trend for the summer season is an
Basin over the years. Significant positive trends tend to important factor in the Yangtze River Basin, and its rate
migrate from the north to the southeast and a shifting of of increase is significant for the lower Yangtze River
the rain belt is happening in same direction in summer Basin. Anomalous summer precipitation between 1991–
(Qian et al., 2003). Qian et al. (2003) suggest that the 2000 and 1961–1990 was 61 mm/decade on the average
variability of summer precipitation in the Yangtze River for the whole basin, but up to 86 mm for the lower
Basin in the recent decades can be explained by a section. Rainstorm is the main form of summer
weakened thermal contrast between the tropical ocean precipitation in the Yangtze River Basin explaining
and the inland Asia due to global warming. This has 54% of the increase in summer rainfall. Rainstorm events
weakened the summer monsoon since the 1970s occurred for 10 more days in 1990–2000 than in 1961–
preventing it from migrating north and keeping it 1990 in the southeastern Yangtze River Basin. This
stationary over the southern Yangtze Basin for a longer could result from a changed atmospheric circulation
time (Qian et al., 2003). In the 1990s, it translated into (weakened summer monsoon) in the recent decades
frequent floods, and the mean decadal index of summer under global warming. Over the last 40 years, higher
monsoon minimized when compared to the index value summer precipitation and rainstorms increased summer
for the other decades (Cao et al., 2005). runoff and floods in the lower Yangtze River Basin. In
 T. Jiang et al. / Geomorphology 85 (2007) 143–154 153

Fig. 13. Rainstorms and flood peaks of 1998 flood at Yichang and Datong, the dashed line denotes the warning water level.

low-lying territories, the lower Yangtze is extremely and Dr. Chongyu Xu. Special thanks are also extended to
vulnerable to floods. An even higher probability of flood Prof. Dr. Chen Zhongyuan for reviewing and revising the
events can be expected in the lower region of the Yangtze manuscript.
River if the observed trends continue.
References
Acknowledgements
Arnell, N.W., 1998. Climate change and water resources in Britain.
This paper is financially supported by the key project Climate Change 39, 83–110.
of the Chinese Academy of Sciences (No. KZCX3-SW- Becker, S., Gemmer, M., Jiang, T., 2003. Observed and interpolated
331), National Natural Science Foundation of China (No. precipitation trends and variability in the Yangtze basin area.
Journal of Lake Sciences Supplement 15, 123–129.
40371112), and Scientific Initial Funds (No. 240025) Brunetti, M., Buffoni, L., Maugeri, M., Nanni, T., 2000. Precipitation
from Nanjing Institute of Geography and Limnology, intensity trends in northern Italy. International Journal of
CAS. We would like to thank the National Climate Center Climatology 21, 455–466.
(NCC) in Beijing and Changjiang Water Resources Buishand, T.A., 1982. Some methods for testing the homogeneity of
Commission (CWRC), Ministry of Water Resources, in rainfall records. Journal of Hydrology 58, 11–27.
Burn, D.H., Hag Elnur, M.A., 2002. Detection of hydrologic trends
Wuhan, for providing valuable climate and hydrological and variability. Journal of Hydrology 255, 107–122.
datasets. We have benefited from useful the discussions Cao, H.X., Liu, H.T., He, H.Zh., Cai, X.H., Shi, N., 2005. Climate
with Dr. Stefan Becker, Dr. Marco Gemmer, Dr. Qi Hu, change in the East-Asian monsoon—introduction of a new method
154 T. Jiang et al. / Geomorphology 85 (2007) 143–154

of time series analysis. In: Jiang, T., King, L., Gemmer, M., IPCC. 2001a, Climate Change 2001: Synthesis Report. A contribution
Kundzewicz, Z.W. (Eds.), Climate Change and Yangtze Floods. of working groups I, II and III. In: Watson, R.T., and the Core
Science Press, Beijing, pp. 246–258. Writing Team (Eds.), Third Assessment Report of the Intergov-
Changjiang Water Resources Commission (CWRC), 2001. Hydrolog- ernmental Panel on Climate Change. Cambridge University Press,
ical Records on Sediment and Water, 1961–2000. (unpublished Cambridge, UK, 398 pp.
materials, in Chinese). IPCC, 2001b. Climate Change 2001: The Scientific Basis. Cambridge
Changjiang Water Resource Commission, Ministry of Water University Press, Cambridge. 785 pp.
Resources (CWRC), 2002. Floods and Droughts in the Yangtze Iwashima, T., Yamatomo, R., 1993. A statistical analysis of the
River Catchment. Water Conservancy and Water Electricity extreme events: long-term trends of heavy precipitation. Journal of
Publication House, Beijing (in Chinese, 326 pp.). Meteorological Society Japan 71, 637–640.
Chen, D.L., Gao, G., 2005. Impact of climate change on runoff from Karl, T.R., Knight, R.W., 1998. Secular trends of precipitation amount,
Hanjing and Ganjiang in theYangtze River basin. In: Jiang, T., frequency and intensity of the United States. Bulletin of American
King, L., Gemmer, M., Kundzewicz, Z.W. (Eds.), Climate Change Meteorological Society 79, 223–241.
and Yangtze Floods. Science Press, Beijing, pp. 155–167. Karl, T.R., Knight, R.W., Plummer, N., 1996. Indices of climate
Chen, Z., Yu, L.Z., Gupta, A., 2001a. The Yangtze River: an change for the United States. Bulletin of American Meteorological
introduction. Geomorphology 41, 1–248. Society 77, 279–292.
Chen, Z., Li, J.f., Shen, H.T., 2001b. Yangtze River, China, historical Kendall, M.G., 1938. A new measure of rank correlation. Biometrika
analysis of discharge variability and sediment flux. Geomorphol- 30, 81–93.
ogy 41, 77–91. Kendall, M.G., 1975. Rank Correlation Methods. Griffin, London. 202 pp.
Chen, D., Gao, G., Xu, C.Y., Guo, J., Ren, G., 2005a. Comparison of Kundzewicz, Z.W., 2005. Flood risk in the changing world—Yangtze
Thornthwaite method and Pan data with the standard Penman– floods. In: Jiang, T., King, L., Gemmer, M., Kundzewicz, Z.W.
Monteith estimates of potential evapotranspiration for China. (Eds.), Climate Change and Yangtze Floods. Science Press,
Climate Research 28, 123–132. Beijing, pp. 246–258.
Chen, X., Su, B.D., Jiang, T., Shi, Y.F., 2005b. Impact study of climate Kwadijk, J.C.J., Rotmans, J., 1995. The impact of climate change on
change on runoff in the Yuanjiang River basin. In: Jiang, T., King, the River Rhine: a scenario study. Climate Change 30, 397–425.
L., Gemmer, M., Kundzewicz, Z.W. (Eds.), Climate Change and Labat, D., Godderis, Y., Probst, J.L., Guyot, J.L., 2004. Evidence for
Yangtze Floods. Science Press, Beijing, pp. 145–154. global runoff increase related to climate warming. Advances in
Dvorak, V., Hladny, J., Kasparek, L., 1997. Climate change, hydrology Water Resources 27, 631–642.
and water resources impact and adoption for selected river basins Maniak, U., 1997. Hydrologie und Wasserwirtschaft. Springer, Berlin.
in Czech Republic. Climate Change 36, 93–107. 650 pp.
Eckharadt, K., Ulbrich, U., 2003. Potential impact of climate change Menzel, L.G., Bürger, G., 2002. Climatic change scenarios and runoff
on groundwater recharge and stream flow in central Europe low response in the Mulde basin (Southern Elbe, Germany). Journal of
mountain range. Journal of Hydrology 284, 243–252. Hydrology 267, 53–64.
Gemmer, M., Becker, S., Jiang, T., 2004. Observed monthly Osborn, T.J., Jones, P.D., 2000. Air flow influences on local climates:
precipitation trends in China 1951–2002. Theoretical and Applied observed United Kingdom climate variations. Atmospheric
Climatology 77, 39–45. Science Letters 1, 62–74.
Gong, D.Y., Ho, C.H., 2003. Shift in the summer rainfall over the Qian, W., Hu, Q., Zhu, Y., Lee, D.K., 2003. Centennial-scale dry–wet
Yangtze River valley in the late 1970s. Geographical Research variation in East Asia. Climate Dynamics 21 (1), 77–89.
Letters 78, 1–4. Su, B.D., Jiang, T., Shi, Y.F., Becker, S., Gemmer, M., 2004. Observed
Groisman, P.Y., Karl, T., Easterling, D., Knight, R.W., Jamason, P., precipitation trends in the Yangtze River basin from 1951–2002.
Hennessy, K.J., Suppiah, R., Page, M.C., Wibig, J., Fortuniak, K., Journal of Geographical Science 14, 204–218.
Razuvaev, V., Douglas, A., Forland, E., Zhai, P.M., 1999. Changes Vogel, R.M., Bell, J.C., Fennessey, N.M., 1997. Stream flow and water
in the probability of the heavy precipitation: important indicators supply in northeastern United States. Journal of Hydrology 198,
of the climate change. Climate Change 42, 243–283. 42–68.
Hartmann, H., Becker, S., King, L., 2005. The three-gorge project: its Wang, J., 2005. Flood characteristics of the Yangtze River and
influence on the flood risks along the middle reaches of the countermeasures for flood control. In: Jiang, T., King, L., Gemmer,
Yangtze River and on the economy of the region surrounding the M., Kundzewicz, Z.W. (Eds.), Climate Change and Yangtze
reservoir. In: Jiang, T., King, L., Gemmer, M., Kundzewicz, Z.W. Floods. Science Press, Beijing, pp. 385–395.
(Eds.), Climate Change and Yangtze Floods. Science Press, Xu, C.Y., 2000. Modelling the effects of climate change on water
Beijing, pp. 396–404. resources in central Sweden. Water Resources Management 14,
He, X.B., Jiao, J.R., 1998. The 1998 flood and soil erosion in the 177–189.
Yangtze River. Water Policy 1, 653–658. Yin, H.F., Li, C.A., 2001. Human impact on floods and flood disaster
Hu, Q., 2005. Flood risk in the changing world—Yangtze floods. In: on the Yangtze river. Geomorphology 41, 105–109.
Jiang, T., King, L., Gemmer, M., Kundzewicz, Z.W. (Eds.), Climate Zhai, P.M., Sun, A.J., Ren, F.M., Liu, X.L., Gao, B., Zhang, Q., 1999.
Change and Yangtze Floods. Science Press, Beijing, pp. 63–72. Changes of climate extremes in China. Climate Change 42, 203–218.
Hu, M.S., Luo, C.Z., 1992. Historical Deluges of China. Chinese
Bookstore Press, Beijing. 521 pp. (In Chinese).