Data C
Resume:
SCIENCE CHINA
Earth Sciences
RESEARCH PAPER October 2011 Vol.54 No.10: 1576 1585
doi: 10.1007/s11430-011-4232-z
Long-term precipitation change by hourly data in Haihe River
Basin during 1961 2004
YIN ShuiQing1,2,3*, GAO Ge3, LI WeiJing3, CHEN DeLiang4 & HAO LiSheng5
1
State Key Laboratory of Earth Surface Processes and Resource Ecology, Beijing Normal University, Beijing 100875, China;
2
School of Geography, Beijing Normal University, Beijing 100875, China;
3
Beijing Climate Center, China Meteorological Administration, Beijing 100081, China;
4
Department of Earth Sciences, University of Gothenburg, Gothenburg 40530, Sweden;
5
Hebei Provincial Climate Center, Shijiazhuang 050021, China
Received August 6, 2010; accepted December 17, 2010; published online July 23, 2011
Hourly summer precipitation data recorded at 21 stations during 1961 2004 in the Haihe River Basin in North China were an-
alyzed. The results show that the precipitation frequency and amount and the morning peak mainly relating to long-duration
rainfall events decreased during this period, whereas the normalized afternoon peak mainly relating to short-duration events
increased, which may suggest that the proportion of short-duration rainfall has increased as the total summer rainfall has de-
creased. For short-duration events, the mean intensity and peak intensity increased at most stations and the time to peak inten-
sity decreased, which may be attributable to the higher thermal contrast between the warmer lower surface and cooler upper
level. In the case of long-duration events, the total amount was significantly correlated with the East Asian summer monsoon
index for the period 1961 2001 (correlation coefficient of 0.63). Although the total amount of rainfall in long-duration events
decreased in the basin, the mean intensity and peak intensity, as well as the extreme hourly precipitation, increased in the
western basin and decreased in the eastern basin.
precipitation, hourly data, diurnal variations, East Asian summer monsoon, Haihe River Basin
Citation: Yin S Q, Gao G, Li W J, et al. Long-term precipitation change by hourly data in Haihe River Basin during 1961 2004. Sci China Earth Sci, 2011, 54:
1576 1585, doi: 10.1007/s11430-011-4232-z
The increasing global surface temperature has tended to well as extremes, has significantly increased in northwest-
induce a more active hydrological cycle [1]. For many ern China including the Xinjiang region, the middle and
mid-latitude regions, it has been found that the mean total lower reaches of the Yangtze River, and the southeastern
precipitation and heavy precipitation both increased during coast. In northern China and the Sichuan Basin, there has
1951 2003, and the rate of increase for the heavy precipita- been a decreasing trend for annual total precipitation and
tion is higher than that for the mean total precipitation [2]. extremes [3 5], whereas the ratio of extreme precipitation
Even where the mean total precipitation has decreased, to the total in North China is reported to have increased [6].
heavy precipitation has had an increasing trend [1]. Yao et al. [7] investigated the regional features of the
Over the last five decades, there has been a slightly de- amounts of total and light-moderate precipitation over the
creasing trend in total precipitation across China, but with entire region of tropical-subtropical Asia during 1978 2002
distinctive regional differences. Annual precipitation, as and revealed a sandwich-like pattern of trends that was pos-
itive over southeastern and northwestern China and negative
over central China and southwestern and northeastern Asia.
*Corresponding author (email: ***********@***.***.**)
Science China Press and Springer-Verlag Berlin Heidelberg 2011 earth.scichina.com www.springerlink.com
1577
Yin S Q, et al. Sci China Earth Sci October (2011) Vol.54 No.10
Furthermore, both the frequency and amount of light rain- trend of extreme hourly rainfall and rainfall event charac-
fall have decreased with high spatial coherency in eastern teristics during a recent 40-year period and thus diagnose
China according to observational evidence and model sim- climate change signals on an hourly timescale, rather than
ulations [8, 9]. Increased surface air temperature in cloud the daily scale considered in previous research. A second
formation processes [10] and increased aerosol concentra- purpose is to analyze diurnal variations of precipitation by
tions [9] are believed to be responsible for the decreasing classifying all stations into three groups and dividing whole
trend of light rainfall. observation series into two periods, to better understand the
The Haihe River Basin provides water resources for ag- causes of trends in event characteristics. The research may
riculture and industry in North China. The annual runoff be helpful in clarifying precipitation processes and optimiz-
decreased because of decreasing annual precipitation over ing the use of water resources in the Haihe River Basin.
the past 50 years [11], implying that the water supply has
been decreasing. The water demand has been increasing
1 Data and methods
rapidly with development of the economy and population
expansion in many big cities; for example, the mega-cities
The Haihe River Basin is located at 112 120 E and
of Beijing and Tianjin. The conflict between water supply
35 43 N. The east-west extension of Yanshan Mountain is
and water demand has increasingly intensified in this area.
to the north and the south-north extension of the Taihang
Furthermore, water shortages have been projected to in-
Mountain is to the west (Figure 1). The total basin area is
crease by 27% 29% by 2030 as a result of climate change
about 31.8 104 km2, of which hills and plateau occupy
and socio-economic development [12].
about 60% and the rest is plain. The Haihe River Basin in-
Most of the above research has been based on daily pre-
cludes the Haihe River, Luanhe River and Tuhaimajia River,
cipitation data because it is easily available. However, the
and provides a vital water resource to North China. Hourly
time scales relevant to many hydrological processes, such as
precipitation data recorded at 21 stations during 1961 2004
infiltration, surface runoff, erosion, and urban sewer sys-
were collected from the Hebei, Henan, Shanxi and Beijing
tems and cropping systems, are short; thus, fine-resolution
Meteorological Bureaus (Figure 1). The hourly precipitation
rainfall information is important [13]. At the same time,
data were obtained by scanning the precipitation autograph-
many newly developed mechanism models, such as SWIM
ic record paper from a siphon self-recording rain gauge and
[14], WEPP [15], FLOCR [16] and WAVE [17], require as
converting the precipitation curve to a precipitation amount
inputs rainfall information that is more detailed than daily
per minute, which was then summed for a specified clock
precipitation. Daily precipitation extremes tend to increase
interval [22]. Most precipitation falls in summer (June, July,
at a rate of 7% per degree of warming according to the
Clausius-Clapeyron relation in global climate models,
whereas hourly observations in the Netherlands showed that
one-hour precipitation extremes increased at a rate of 14%
per degree of warming when the daily mean temperature
exceeded 12 C [18], which implies that the short-duration
precipitation extremes reflected by hourly precipitation data
may increase at a faster rate than those reflected by daily
precipitation. It is well known that these short-duration ex-
treme rainfall events are responsible for local flooding, soil
erosion and water damage. Traditional flood risk assessment
and infrastructure design based on the hypothesis of sta-
tionary rainfall series may need to be adapted considering
climate trends [19].
There is evidence of precipitation diurnal variations un-
dergoing long-term changes in other parts of the world. This
could be due to climate change induced by the enhanced
greenhouse effect, the intensification of urban heat islands,
the alteration of wet and dry periods, or low-frequency vari-
ations in the El Ni o/La Ni a-Southern Oscillation [20].
Hourly precipitation records from one station in Beijing
indicate that long-duration rainfall events decreased in
amount, whereas short-duration events increased in amount,
in a recent period of 40 years [21], which may suggest that
the proportion of higher-intensity rainfall has been increas- Figure 1 Meteorological stations having hourly precipitation data in this
ing. One purpose of the present research is to survey the study.
1578 Yin S Q, et al. Sci China Earth Sci October (2011) Vol.54 No.10
August; JJA) in the Haihe River Basin; thus, only summer rainfall frequency and amount for short-duration events to
precipitation was analyzed in this study. those for long-duration events were high in the northwestern
When the total precipitation amount in an hour was no mountainous area, which suggests that short-duration events
less than 0.1 mm, this hour was termed a precipitation hour. were more common than long-duration ones there (data not
An individual rainfall event was defined as a period of rain- shown). Figure 2 shows that the diurnal variations in after-
fall with at least two preceding and two succeeding noon peaks of summer precipitation related mainly to short-
non-precipitation hours [21]. The hourly data were then duration events and the diurnal variations in morning peaks
used to compute rainfall event characteristics including the related mainly to long-duration events, which is consistent
amount, duration, mean intensity, peak intensity and the with the results of Li et al. [21]. The stations located in the
time to peak intensity. An event amount (mm) was defined northwestern mountainous area of the Haihe River Basin
as the total rainfall received in the duration (h) of the event. recorded more precipitation in the afternoon and those near
The event mean intensity (mm h 1) was the event amount the eastern coast recorded more precipitation in the morning.
divided by the event duration. Event peak intensity (mm h 1) This is the probable reason for short-duration events being
more common than long-duration events in the northwest-
was the maximum hourly intensity during an event. The
ern mountainous area.
event total amount can be no less than 0.1 mm and the event
duration can be no less than 60 min. The time to peak inten-
sity (h) was the kth hour when the hourly precipitation was 2.2 Long trend of extreme hourly precipitation and
greatest during an event. The normalized time to peak in- event characteristics
tensity was defined as the time to peak intensity divided by
The amount of summer rainfall decreased in the Haihe Riv-
the event duration, which reflects the timing of the peak
er Basin in a recent 40-year period, mainly because of
during an event. Long-duration and short-duration events
weakening southerly monsoon winds in North China [24].
had different long-term trends; thus, rainfall events were
The analysis of hourly precipitation data also shows that
classified according to their durations. Long-duration events
precipitation hours and the amount of total summer precipi-
are longer than 6 h and short-duration events are shorter
tation decreased during the 40 years with high spatial co-
than 6 h [21].
herency (Figure 3(a), (b)). Zunhua has the most prominent
Diurnal variations in precipitation were analyzed using
decreasing trend for precipitation hours, with a magnitude
hourly precipitation data and the same methods employed
of 14.5 h (10 a) 1. The summer precipitation amounts in
by Yin et al. [23]. To compare the characteristics of diurnal
variations for two different periods, the normalized precipi-
tation occurrence frequency, which was defined as the orig-
inal occurrence frequency divided by the 24-h mean, was
used.
2 Results and discussion
2.1 General statistics of short- and long-duration events
The total summer rainfall amount averaged over the Haihe
River Basin during 1961 2004 for short-duration events
was 156.1 mm, which accounted for about 44% of the total
summer rainfall amount (Table 1). The number of events
averaged over the basin during 1961 2004 for short-dura-
tion events was 33.7, which accounted for about 82% of all
events in summer (Table 1). Referring to the spatial distri-
butions, both short-duration events and long-duration events Figure 2 Diurnal variations in occurrence frequency of the event peak
had maximum rainfall amounts on the eastern coast, where intensity for short-duration events (solid line with open circles) and
moist supplies were more sufficient. However, the ratios of long-duration events (dashed line with open squares).
Table 1 Comparison of precipitation event characteristics of short- and long-duration events averaged over the Haihe River Basin during 1961 2004
Event intensity Event peak
Event types classified by Total summer Number of events Event duration
Event amount (mm)
(mm h 1) intensity (mm h 1)
duration length amount (mm) (times) (h)
Short-duration events 156.1 33.7 4.7 2.4 1.8 3.4
Long-duration events 195.1 7.3 27.1 11.9 2.2 8.5
1579
Yin S Q, et al. Sci China Earth Sci October (2011) Vol.54 No.10
Figure 3 Trends for hourly precipitation and extreme hourly precipitation. (a) The number of hours with precipitation no less than 0.1 mm; (b) the total
amount of hourly precipitation no less than 0.1 mm; (c) the number of extreme hours with precipitation no less than 10 mm; (d) the total amount of extreme
hourly precipitation with precipitation no less than 10 mm. Solid circles represent the stations with trends significant at the a=0.1 level.
trend of 10.9 mm (10 a) 1 significant at the a=0.1 level.
Laoting and Zunhua decreased at rates of about 50 mm (10
a) 1, which were the most remarkable trends across the The average intensity and peak intensity of long-duration
whole river basin. events also had opposite trends for the east and west of the
The trends for extreme hourly rainfall (intensity no less basin decreasing in the east and increasing in the west
than 10 mm h 1) were opposite for the east and west (Figure (Figure 4(a), (c)). Previous research using daily precipita-
tion data showed that effective precipitation days have de-
3(c), (d)). The extreme hourly rainfall at most stations in the
creased relatively homogeneously in North China [4, 7, 8].
east had a decreasing trend, with the magnitude exceeding
0.5 h (10 a) 1 at seven stations (Fengning, Beijing, Zunhua, The opposite trends of extreme hourly precipitation and
intensity of long-duration events for west and east of the
Qinhuangdao, Tangshan, Laoting and Raoyang). The de-
basin may relate to topography variations within the basin,
creasing trends for Fengning, Zunhua and Laoting were
which will be discussed in detail in sections 2.3 and 2.4. At
significant at the a=0.1 level. Extreme hourly rainfall at
most stations in the northwest, the normalized time to peak
most stations in the west had an increasing trend, with that
for long-duration events had a decreasing trend, which im-
for Weixian exceeding 0.5 h (10 a) 1 and being significant
plies that the peak is a little earlier during a rainfall event.
at the a=0.1 level (Figure 3(c)). The total amount of extreme
The decreasing trend for the Zhangjiakou station was sig-
hourly rainfall had a trend similar to that for the number of
nificant at the a=0.1 level. However, there were still nine
extreme hours; i.e. decreasing in the east and increasing in
stations with increasing trends (Figure 5(a)).
the west (Figure 3(d)). Zunhua had the most remarkable
The average intensity of short-duration events, as well as
decreasing trend of 28.5 mm (10 a) 1 significant at the
the peak intensity, increased at most stations in the northern
a=0.1 level; Weixian in the western basin had an increasing
1580 Yin S Q, et al. Sci China Earth Sci October (2011) Vol.54 No.10
Figure 4 Trends for event-average intensity and peak intensity for long- and short-duration events. (a) Average intensity for long-duration events; (b)
average intensity for short-duration events; (c) peak intensity for long-duration events; (d) peak intensity for short-duration events.
Haihe River Basin (Figure 4(b), (d)). The increasing trends than morning peaks (Figure 6(a)). Land surfaces in the
of the average intensity were significant at the a=0.1 level mountainous area are easily heated by incoming solar radia-
for Zhangbei, Weixian and Zunhua, with Zunhua having the tion during the day, and this may induce convective precip-
most remarkable increasing trend (0.18 mm h 1 (10 a) 1. The itation more often in the afternoon.
Yuanping, Shijiazhuang, Beijing, Miyun, Zunhua,
increasing trends of peak intensity were significant at the a=
Baxian, Xingtai, Anyang and Xinxiang are located in val-
0.1 level for Zhangbei, Weixian, Zunhua and Tangshan, with
Tangshan having the most remarkable increase (0.40 mm h 1 leys and constitute a second group with typical nighttime
rainfall (Figure 6(c)). The results are consistent with previ-
(10 a) 1). The normalized time to peak decreased for most
ous research showing that stations in the mountainous area
stations (16 of 21 stations), which suggests that the peak is a
had a clear afternoon peak in convective precipitation
little earlier during a rainfall event (Figure 5(b)).
whereas those located on the plains had more frequent pre-
cipitation at night in the Beijing-Tianjin-Hebei district [25].
2.3 Diurnal variations in summer precipitation
A general conclusion was drawn that convex topography
readily triggered daytime precipitation, whereas concave
To better understand the causes of the trends of event char-
topography encouraged nocturnal rainfall [26]. The mecha-
acteristics, diurnal variations in the precipitation frequency
nism is described in detail in Yin et al. [23].
were analyzed. The patterns of diurnal variations could be
The third group comprised Cangzhou, Qinhuangdao,
generally divided into three groups. Datong, Zhangbei,
Tangshan, Laoting and Raoyang and had a higher morning
Weixian, Fengning, Zhangjiakou, Huailai and Chengde
peak (Figure 6(e)). The climate of these stations was af-
constitute the first group, having higher afternoon peaks
1581
Yin S Q, et al. Sci China Earth Sci October (2011) Vol.54 No.10
Figure 5 Trends for the normalized time to peak intensity. (a) Long-duration events; (b) short-duration events.
fected by land-ocean circulation. Among the stations, those frequencies for long- and short-duration events decreased
by 0.34 events (10 a) 1 (average of 7.3 events, Table 1) and
that were farthest from the ocean had an afternoon sub-peak,
1.66 events (10 a) 1 (average of 33.7 events, Table 1). The
which seems to be a kind of transition towards inland char-
acteristics with a typical afternoon peak. The morning peak event mean intensity for long-duration events averaged over
could be attributed to diurnal land-sea breezes induced by the whole basin increased slightly by 0.007 mm h 1 (10 a) 1
the land-ocean thermal contrast [23, 27]. (average of 2.2 mm h 1, Table 1), probably because of the
This spatial pattern of diurnal variations in the summer contrary trends in the east and west. The event mean intensity
precipitation in the Haihe River Basin was mainly affected for short-duration events increased more significantly by
by the mountain-valley circulation and the land-ocean cir- 0.044 mm h 1 (10 a) 1 (average of 1.8 mm h 1, Table 1),
culation. Divergences at the 950 hPa level calculated from mainly because of the relatively homogenous positive trends.
the National Centers for Environmental Prediction (NCEP) Rainfall in summer in North China has decreased since
Final Analyses (FNL) data with 1 1 spatial resolution in the late 1970s [28]. Analysis of hourly data obtained from
JJA averaged over 2000 2007 reveal that the convergence the 21 stations shows a sharp decrease in 1980 (data not
area was coincident with the area of frequent rainfall. In the shown); thus, the whole period (1961 2004) was divided
afternoon (Figure 7(a), (b)), the center of negative diver- into two periods (1961 1979 and 1980 2004) to detect the
gence (upward motion) at the 950 hPa level is located over differences in diurnal variations in precipitation. First, for
the western and northern mountainous areas of the basin, all three groups, the precipitation frequency was lower for
which coincides with the afternoon peak of precipitation in the period 1980 2004, especially in the morning, than the
this area. At the same time, the center of positive divergence period 1961 1979 (data not shown). This could be seen
(downward motion) is located over the ocean and plains, more clearly in the normalized frequency of diurnal varia-
which leads to a low probability of precipitation in the af- tions (Figure 6(b), (d), (f)). The normalized morning peak
ternoon for stations close to the ocean and on the plains. for the period 1980 2004 was smaller than that for
Late at night and early in the morning (Figure 7(c), (d)), the 1961 1979, whereas the normalized afternoon peak was
mountainous winds converge over the plains, which is pos- larger. The result is consistent with previous research sug-
sibly responsible for the nocturnal precipitation maximum gesting that although the total rainfall decreased in a recent
of the plains. 40-year period in the Haihe River Basin, the proportion of
extreme precipitation in the total increased [6].
The afternoon peak mainly relating to short-duration
2.4 Comparison of variations in diurnal summer pre-
rainfall events can be attributed to moist convection related
cipitation between two periods
to surface solar heating. Sounding data for Beijing (54511)
and Xingtai (53798) in the Haihe River Basin show that the
The rainfall amounts in summer averaged over the Haihe
surface temperature increased respectively by 0.5 C (10 a) 1
River Basin for long- and short-duration events decreased
by 12.8 mm (10 a) 1 (average of 195.1 mm, Table 1) and (08:00 Local Standard Time, LST hereafter) and 0.3 C (10
a) 1 (20:00 LST), whereas the air temperature at the 300 hPa
2.5 mm (10 a) 1 (average of 156.1 mm, Table 1), whereas
level decreased by 0.2 C (10 a) 1 (08:00 LST) and 0.2 C
the ratio of rainfall for short-duration events to the total in-
creased by 1.3% (10 a) 1 during 1961 2004. The rainfall (10 a) 1 (20:00 LST) during 1966 2007 (Figure 8). The
1582 Yin S Q, et al. Sci China Earth Sci October (2011) Vol.54 No.10
Figure 6 Climatology characteristics ((a), (c) and (e) based on data for 1971 2000) and comparisons of two periods ((b), (d) and (f) based on data for
1961 1979 and 1980 2004) of diurnal variations in the normalized summer precipitation for the three groups of stations. (a) and (b) represent the group
characteristic of a higher afternoon peak in the mountainous area; (c) and (d) represent the group characteristic of typical nighttime rainfall on the plains; (e)
and (f) represent the group characteristic of a higher morning peak near the eastern coast.
higher thermal contrast between lower and upper levels led China were affected by the East Asian summer monsoon
to greater static instability, which increased the relative mainly in July and August (JA). Figure 9 shows the high
probability of late-afternoon convection precipitation [21]. correlation between the rainfall amount for long-duration
For short-duration events, the mean intensity and peak in- events in JA and the East Asian Monsoon (EAM) index,
tensity increased, and the time to peak was earlier, which which is defined as the lower-tropospheric meridional wind
can also be attributed to the greater thermal contrast be- averaged over 105 E to 120 E, 30 N to 40 N, 850 500 hPa,
and July to August, calculated using ERA-40 data from the
tween upper and lower levels.
European Center for Medium range Weather Forecasting
The morning peaks tended to strengthen and weaken
(ECMWF) [24]. The correlation coefficient was 0.63 for the
with the advance and retreat of the East Asian monsoon in
period 1961 2001. Both the rainfall amount for long-dura-
eastern China [23, 29, 30]. The rainfall systems in North
1583
Yin S Q, et al. Sci China Earth Sci October (2011) Vol.54 No.10
Figure 7 Spatial distribution of the 950 hPa divergence (colored) and wind vectors based on the NCEP FNL data for JJA during 2000 2007.
tion events and EAM index had decreasing trends, with precipitation tended to be more concentrated. However, in
rates of 10.8 mm (10 a) 1 and 0.34 m s 1 (10 a) 1. Because the eastern Haihe River Basin, the mean and peak intensities
for long-duration events decreased, as did extreme hourly
of a lack of water vapor, the morning peak resulting mainly
precipitation. This may be because of regulation by the
from long-duration events was remarkably reduced. The
ocean; indeed, in the last section it was suggested that the
opposite trends of intensity in the western and eastern basin
diurnal variations in precipitation in the eastern Haihe River
can be attributed to the mountain-valley circulation and
Basin were influenced by the land-ocean circulation.
land-ocean circulation. In the western Haihe River Basin,
although the rainfall amount for long-duration events de-
creased, the mean intensity and peak intensity for long-
3 Conclusions
duration events increased, as did the extreme hourly precip-
itation. This suggests that the rate of decrease for event du-
Analyses based on hourly summer precipitation data for 21
ration was higher than that for event amount, and that the
1584 Yin S Q, et al. Sci China Earth Sci October (2011) Vol.54 No.10
Figure 8 Variations and trends of air temperature at the surface (solid line with open circle) and at the 300 hPa level (dashed line with open squares) during
1966 2007 based on sounding data averaged over Beijing (54511) and Xingtai (53798) in the Haihe River Basin at 08:00 LST (a) and 20:00 LST (b).
area with a higher afternoon peak, stations on the plains
with typical nighttime rainfall, and stations close to the
ocean with a higher morning peak. When precipitation di-
urnal variations for two different periods (1961 1979 and
1980 2004) were compared, it was found that the morning
peak mainly related to long-duration events decreased and
the normalized afternoon peak mainly related to short-du-
ration events increased. Higher thermal contrast between
lower and upper levels may be responsible for the trend for
increasing intensity for short-duration events. The trend for
the decreasing amount of rainfall for long-duration events
was highly correlated with the trend for a weakening East
Asian monsoon in North China. The intensity for
long-duration events, as well as for extreme hourly precipi-
tation, increased in the western basin and decreased in the
eastern: this may be attributable to the mountain-plain-
ocean topography of the basin. The mechanism for how
Figure 9 Correlation between rainfall amount for long-duration events
(solid line with filled circles) in JA and the EAM index (dashed line with topography affects rainfall intensity trends requires further
open circles), which was defined as the lower-tropospheric meridional
study.
wind averaged over 105 E to 120 E, 30 N to 40 N, 850 500 hPa, and July
The decreasing annual rainfall, together with increasing
to August, calculated from ECMWF reanalysis data.
intensity, is not good for the absorption of water by soil and
may increase the risk of drought. On the other hand, the
stations during 1961 2004 in the Haihe River Basin greater intensity of short-duration events and the earlier
demonstrated that the summer total rainfall amounts for peak during a rainfall event may increase the risk of urban
long- and short-duration events averaged over the basin waterlogging, local flooding and soil erosion. It will be in-
decreased by 12.8 and 2.5 mm (10 a) 1 respectively, where- teresting to assess the importance of trends for extremes as
as the ratio of rainfall for short-duration events to the total reflected by hourly scale precipitation data in future work
increased by 1.3% (10 a) 1 in a recent 40-year period. In focused on flood risk assessment and infrastructure design.
respect to spatial distribution, hourly precipitation frequen-
cy and amount decreased with high spatial coherency in the
The authors would like to thank the Hebei, Henan, Shanxi and Beijing
basin; extreme hourly precipitation (intensity no less than
Meteorological Bureaus for supplying hourly precipitation data, academ-
10 mm h 1), as well as mean and peak intensities for ician Wang Hao and two anonymous reviewers for their valuable com-
long-duration events decreased in the eastern basin but in- ments, and Dr. Han Rongqing, Dr. Ou Tinghai and Zuo Jinqing for analy-
sis of the EAM index and divergence field. This work was supported by the
creased in the western; the mean and peak intensity for
National Basic Research Program of China (Grant No. 2006CB403404),
short-duration events increased at most stations in the basin. the National Key Technology R&D Program of China (Grant No. 2007
The patterns of diurnal variations could generally be di- BAC29B04) and the National Science Foundation Program for Post-
vided into three groups: stations located in the mountainous doctoral Scientists of China (Grant No. 200********).
1585
Yin S Q, et al. Sci China Earth Sci October (2011) Vol.54 No.10
1 Dore M H I. Climate change and changes in global precipitation pat- 16 Oostindie K, Bronswijk J J B. FLOCR-A Simulation Model for the
Calculation of Water Balance, Cracking and Surface Subsidence of
terns: What do we know? Environ Int, 2005, 31: 1167 1181
Clay Soils. DLO Winand Staring Centre Report No. 47. Nether-
2 IPCC. Summary for policymakers. In: Solomon S, Qin D H, Manning
lands: Wageningen, 1992
M, et al, eds. Climate Change 2007: The Physical Science Basis,
17 Vanclooster M, Viaene P, Diels J. WAVE: A Mathematical Model
Contribution of Working Group I to the Fourth Assessment Report of
for Simulating Water and Agrochemicals in the Vadose Environment:
the Intergovernmental Panel on Climate Change. Cambridge: Cam-
Reference and Users Manual (Release 2.0). Leuven: Katholieke Uni-
bridge University Press, 2007
versiteit Leuven, 1994
3 Zhai P M, Sun A J, Ren F M, et al. Changes of climate extremes in
18 Lenderink G, van Meijgaard E. Increase in hourly precipitation ex-
China. Clim Change, 1999, 42: 203 218
tremes beyond expectations from temperature changes. Nat Geosci,
4 Zhai P M, Zhang X B, Wan H, et al. Trends in total precipitation and
2008, 1: 511 514
frequency of daily precipitation extremes over China. J Clim, 2005,
19 Re M, Barros V R. Extreme rainfalls in SE South America. Clim
18: 1096 1108
Change, 2009, 96: 119 136
5 Qian W H, Lin X. Regional trends in recent precipitation indices in
20 Fujibe F, Yamazaki N, Kobayashi K. Long-term changes in the diur-
China. Meteorol Atmos Phys, 2005, 90: 193 207
nal precipitation cycles in Japan for 106 years (1898 2003). J Mete-
6 Zhai P M, Wang C C, Li W. A review on study of change in precipi-
orol Soc Jpn, 2006, 84: 311 317
tation extremes (in Chinese). Adv Clim Change Res, 2007, 3:
21 Li J, Yu R C, Wang J J. Diurnal variations of summer precipitation in
144 148
Beijing. Chin Sci Bull, 2008, 53: 1933 1936
7 Yao C, Yang S, Qian W H, et al. Regional summer precipitation
22 Wang B M, Lu Y P, Zhang Q. The color scanning digitizing pro-
events in Asia and their changes in the past decades. J Geophys Res,
cessing system of precipitation autographic record paper (in Chinese).
2008, 113: D17107, doi: 10.1029/2007JD009603 J Appl Meteorol Clim, 2004, 15: 737 744
8 Fu J L, Qian W H, Lin X, et al. Trends in graded precipitation in 23 Yin S Q, Chen D, Xie Y. Diurnal variations of precipitation during
China from 1961 to 2000. Adv Atmos Sci, 2008, 25: 267 278 the warm season over China. Int J Climatol, 2009, 29: 1154 1170
9 Qian Y, Gong D Y, Fan J W, et al. Heavy pollution suppresses light 24 Yu R C, Wang B, Zhou T J. Tropospheric cooling and summer mon-
rain in China: Observations and modeling. J Geophys Res, 2009, 114: soon weakening trend over East Asia. Geophys Res Lett, 2004, 31:
D00K02 L22212, doi: 10.1029/2004GL021270
10 Qian W H, Fu J L, Yan Z W. Decrease of light rain events in summer 25 Wu Z H, Chen M X. Climatic analysis of summer (JJA) short-period
associated with a warming environment in China during 1961 2005. precipitation in Beijing-Tianjin-Hebei area (in Chinese). In: Proceed-
Geophys Res Lett, 2007, 34: L11705, doi: 10.1029/2007GL029631 ings of International Symposium on Sino-US Nowcasting of Severe
11 Zhang J Y, Zhang S L, Wang J X, et al. Study on runoff trends of the Convection Weather. Beijing: China Meteorological Press, 2004.
six larger basins in China over the past 50 years (in Chinese). Adv 264 269
Water Sci, 2007, 18: 230 234 26 Lin Z G. Topography-Induced Precipitation Climatology (in Chinese).
12 Wang J X, Li H, Xia J, et al. Water shortage and adaptation measures Beijing: Science Press, 1995
27 Dai A. Global precipitation and thunderstorm frequencies. Part II:
under climate change: Simulation analysis in the Haihe River Basin
Diurnal variations. J Clim, 2001, 14: 1112 1128
(in Chinese). Adv Clim Change Res, 2008, 4: 336 341
28 Lu R Y. Linear relationship between the interdecadal and interannual
13 Kandel D D, Western A W, Grayson R B, et al. Process parameteri-
variabilities of North China rainfall in rainy season. Chin Sci Bull,
zation and temporal scaling in surface runoff and erosion modeling.
2003, 48: 1040 1044
Hydrol Process, 2004, 18: 1423 1446
29 Chen G, Sha W M, Iwasaki T. Diurnal variation of precipitation over
14 Ross P J. SWIM A Simulation Model for Soil Water Infiltration
southeastern China: Spatial distribution and its seasonality. J Ge-
and Movement, CSIRO Division of Soils. Queensland: Townsville,
ophys Res, 2009, 114: D13103, doi: 10.1029/2008JD011103
1990
30 Chen G, Sha W M, Iwasaki T. Diurnal variation of precipitation over
15 Lane L J, Nearing M A. USDA-Water Erosion Prediction Project:
southeastern China: 2. Impact of the diurnal monsoon variability. J
Hillslope Profile Model Documentation. NSERL Report No. 2. Indi-
Geophys Res, 2009, 114: D21105, doi: 10.1029/2009JD012181
ana: West Lafayette, 1989
Contact this candidate