Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A Comprehensive Analysis of the Changes in Precipitation Patterns over Beijing during 1960–2012

A Comprehensive Analysis of the Changes in Precipitation Patterns over Beijing during 1960–2012 Hindawi Advances in Meteorology Volume 2019, Article ID 6364040, 22 pages https://doi.org/10.1155/2019/6364040 Research Article A Comprehensive Analysis of the Changes in Precipitation Patterns over Beijing during 1960–2012 1,2,3 2 1 1 Xiaomeng Song , Jianyun Zhang, Chunhua Zhang, and Xianju Zou School of Resources and Geosciences, China University of Mining & Technology, Xuzhou 221116, China State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Nanjing Hydraulic Research Institute, Nanjing 210029, China State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan 430072, China Correspondence should be addressed to Xiaomeng Song; xmsong.cn@gmail.com Received 15 October 2018; Accepted 28 January 2019; Published 6 March 2019 Academic Editor: Efthymios I. Nikolopoulos Copyright © 2019 Xiaomeng Song et al. ,is 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. Precipitation pattern has changed over many regions in recent decades, which may cause the risk of flood or drought. In this study, the main objective is to evaluate the spatiotemporal variability of precipitation in Beijing from 1960 to 2012. First, the mean monthly, seasonal, and annual precipitation series were used to analyze the temporal variation using regression, Mann–Kendall (M-K) test, Sen’s slope, and Pettitt tests. ,e results showed that the annual mean precipitation had a clear decreasing trend, with the statistically significant decrease in summer (especially in July and August) and significant increase in spring (especially in May). Although the decreasing trend is shown in the precipitation concentration indicators, the temporal uneven distribution of precipitation has unchanged. Subsequently, the precipitation time series at 30 stations over Beijing were used to evaluate the changes in precipitation pattern. ,e results showed that the annual series for the most rain gauges had decreasing trends with gradual changes. ,e spatial distribution of precipitation and other indices is geographically consistent, reflecting the principal physiographic and climatic conditions. At the same time, the effects of the terrain and urban development on the precipitation spatial distribution were detected. Generally, the large and heavy precipitations frequently occur in the plain areas, while the precipitation in the mountain areas is dominated by the small and medium precipitation. As a whole, the total precipitation in the plain areas (558.8 mm) was slightly higher than that in the mountainous areas (533.0 mm), while the precipitation in the urban areas (575.9 mm) was much higher than in the surrounding suburb areas (538.9 mm) during 1960–2012. ,e differences between the plain and mountainous areas during the period of 1960–1979, 1980–1999, and 2000–2012 were 24.2 mm, 32.6 mm, and 17.7 mm, respectively. ,e differences in precipitation between the urban and suburb areas for the three periods were 32.9 mm, 45.2 mm, and 31.0 mm, respectively, with the amount accounting for 5.51%, 7.66%, and 5.94% of the mean precipitation in the urban areas for the corresponding periods. transport in rivers and coastal waters. ,ere is a growing 1. Introduction concern in the scientific community over whether there are ,e long-term observation of climatic variables as a practical significant changes in precipitation amount, intensity, du- approach for monitoring climate change is receiving con- ration, and frequency because changes in precipitation siderable attention from researchers throughout the world. patterns may lead to floods or droughts in different areas Of the common climatic variables, precipitation is the most [2–4]. Also, trend and variability analysis of rainfall requires changeable in time and space, which directly affects natural urgent and systematic attention because of significant cycles of water resources. In addition, severe precipitation possible influences on freshwater availability, occurrence of can also cause storm-induced floods [1] and heavy sediment extreme events, food security, and economic activities [5–8]. loads from watersheds which may change sediment ,erefore, the spatial and temporal variability of 2 Advances in Meteorology precipitation time series is important from both the scientific increasing trend from 1724 to 2009; in addition, they found and practical point of view. some periodic characteristics that affected the trends re- garding the amount of annual precipitation. Zhai et al. [4] Linking with industrialization, during the late 20th to the early 21st century, urbanization occurred rapidly across discussed the spatiotemporal variation of precipitation from developing countries, especially in China; this urbanization 1724 to 2010 based on the same meteorological station. ,e is projected to continue over the coming decades. Urbani- annual precipitation presented a slowly increasing trend zation has modified the earth system [9] and local water during 1724–2010, which is consistent with that of the work cycle [10], creating environmental burdens, including urban by Li et al. [15]. Further, Zhai et al. [4] discussed the spatial heat islands (UHI) [11], urban aerosols [12], storm, and variations of precipitation from 1980 to 2010 based on the 20 flooding in urban areas [13, 14]. As a result, it is important to weather stations and showed that annual precipitation has analyze the rainfall variability in urban areas. Several studies decreased significantly during the first decade of the 21st have addressed issues related to urban heat island, impact of century compared to the last two decades of the 20th urbanization on rainfall, and spatiotemporal variations of century. For this study, a comprehensive analysis was conducted rainfall, especially for the metropolitan areas [3, 4, 15–19]. Urban expansion is associated commonly with increased of the changes in precipitation at the regional-scale using 30 amount of precipitation and the frequency of extreme rain gauges in the Beijing metropolitan area. ,is work precipitation in urban areas as well as downwind areas attempted to (1) identify gradual trends and abrupt shifts at because of the effect of urban heat island and high con- various temporal scales (monthly, seasonal, and annual) in centrations of anthropogenic aerosols [20–22]. Precipitation precipitation data series, (2) investigate the temporal and changes and variability in the metropolitan areas could spatial variations of precipitation with the consideration of present a challenge in terms of stormwater management, the urbanization and topography effect, and (3) discuss the water quality, and water supply [23, 24]. changes of precipitation patterns based on the classified precipitation events. ,e findings may be helpful to scien- Beijing, located in northern China and surrounded by mountains with complex terrain, is the capital of China and tifically understand the structural changes in precipitation and to evaluate the water availability and urban water re- one of the most populous cities in the world. It has expe- rienced rapid urbanization in the past few decades due to sources management in the study area. ,is study can also China’s reforms [25]. ,e built-up area of Beijing has grown provide a reference for further studies on precipitation 2 2 from 346 km in 1980 to 1350 km in 2013, and the resident mechanisms in urban areas. population of the city has exceeded 21 million. ,e complex terrain is likely to cause valley wind circulation, which may 2. Materials and Methods have an impact on the local climate, and the changes in the underlying characteristics as a result of rapid urbanization 2.1. Data Sources and Data Preprocessing. ,e daily rain will also influence local precipitation. ,erefore, the study of gauge data from 30 stations in Beijing areas were analyzed in temporal and spatial variations in the local climatic char- this study. ,ese data were provided by the Beijing Hy- acteristics of Beijing is important to our understanding of drological Stations (BHS) of the Beijing Water Authority the impacts of urbanization on the local climate. To date, (BWA). Note that most of the 30 observation stations within numerous studies have analyzed rainfall variability at dif- Beijing were established in the late 1950s and early 1960s and ferent temporal scales from different perspectives. Other formed a fixed network by the end of 1960s. As a result, the studies have been done on the possible effects of urban precipitation data from 1960 to 2012 are used to analyze the expansion on precipitation in the Beijing areas spatial distribution and the urbanization effects. ,e loca- [3, 17, 19, 25–32]. For example, Miao et al. [29] discussed the tions are shown in Figure 1, and details about these stations effects of land use or land cover on the characteristics of are listed in Table 1. For some stations, the missing pre- boundary layer structures of urban areas. ,eir results cipitation due to disruption in data collection requires data showed that the urban-rural circulations induced by topo- infilling. Various infilling methods have been reported in the graphic differences were one of the important causes for the literature, such as nearest neighbor, inverse distance, linear prevalence of mountain-valley flows in the Beijing area. regression, ordinary kriging, multiple linear regression, Zhang et al. [33] found that urbanization predominantly neural networks, and copula-based method [37, 38]. contributed to a reduction in precipitation in Beijing, based Compared to infilling precipitation data on short time scales on a mesoscale model, particularly over the Miyun reservoir (up to several days), the annual time scale is relatively simple. area. Because Beijing has a varied and complex topography As a result, in this work, the nearest neighbor method is used with high mountains in the west and north parts of Beijing, to infill the missing data via the neighbor control station and Sun et al. [34] and Wu et al. [35] pointed out that the terrain its observations. ,e missing data can be estimated by had an important effect on the location and distribution of multiplying with the ratio of the long-term means of the precipitation. Similarly, Sun and Yang [36] found that a target station and its nearest neighbor. In this study, the reduction in rainfall was caused by the joint effects of to- “nearest” neighbor station was selected based on the cor- pography and the urban heat island. In addition, significant relation coefficient and its sufficiently available precipitation spatiotemporal and interseasonal variations in the trends series. For example, the four stations surrounding Gao- and variability of precipitation in the Beijing area were beidian are Tongzhou, Songlinzha, Youanmen, and Maju- noticed. For example, Li et al. [15] revealed that there was an qiao. Figure 2 shows the annual precipitation correlation Advances in Meteorology 3 115°E 116°E 117°E China 40°N 30°N 41°N 20°N THK QJD ZJF XH 10°N YQ HHC MY ZIY HR 90°E 100°E 110°E 120°E TYK GT WJY National boundary PG SH SY SZ YC Basin boundary ZT SJD 40°N SLZ TZ Beijing city YAM GBD YLZ MJQ XYL HC FS ZF NGZ km 010 20 40 60 80 DEM Value Rain gauges –29 to 188.6 Beijing plain 188.7 to 451.5 District 451.6 to 732.6 732.7 to 1,104.3 1,104.4 to 2,283 Figure 1: Location and the distribution of the 30 rain gauges in Beijing that were used in this study. Table 1: Location of the 30 rain gauges and time series information. ° ° Station Abbreviation Longitude ( E) Latitude ( N) Time series Missing data Fangshan FS 115.97 39.70 1950–2012 1968, 1992–1996 Gaobeidian GBD 116.52 39.91 1972–2012 None Guanting GT 115.61 40.23 1950–2012 None Huangcun HC 116.33 39.73 1954–2012 1960–1961 Huanghuacheng HHC 116.34 40.40 1955–2012 None Huairou HR 116.61 40.31 1952–2012 None Majuqiao MJQ 116.55 39.76 1960–2012 1961 Miyun MY 116.84 40.39 1950–2012 1962 Nangezhuang NGZ 116.39 39.50 1965–2012 1976–1978 Pinggu PG 117.10 40.16 1950–2012 1992 Qianjiadian QJD 116.34 40.69 1961–2012 None Shahe SH 116.33 40.13 1959–2012 1979 Sanjiadian SJD 116.10 39.96 1950–2012 None Songlinzha SLZ 116.37 39.95 1963–2012 None Shunyi SY 116.66 40.11 1950–2012 1992–1996 Suzhuang SZ 116.75 40.06 1950–2012 None Tanghekou THK 116.63 40.73 1960–2012 1962 Taoyukou TYK 116.45 40.23 1962–2012 None Tongzhou TZ 116.65 39.93 1950–2012 None Wangjiayuan WJY 115.98 40.20 1962–2012 None Xiahui XH 117.16 40.61 1961–2012 1970–1973 Xiayunling XYL 115.74 39.73 1951–2012 1961, 1992–1996 Youanmen YAM 116.35 39.87 1963–2012 1968 Yanchi YC 115.89 40.03 1963–2012 None Yulinzhuang YLZ 116.79 39.79 1961–2012 None Yanqing YQ 115.99 40.46 1959–2012 1992–1993 Zhangfang ZF 115.69 39.58 1963–2012 None Zhangjiafen ZJF 116.78 40.61 1960–2012 None Zhenluoying ZLY 117.14 40.34 1954–2012 None Zhaitang ZT 115.68 39.96 1951–2012 None 4 Advances in Meteorology 1000 1000 y = 0.9404x + 56.058 y = 1.0539x – 17.093 R = 0.7604 R = 0.8715 800 800 600 600 400 400 200 200 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Gaobeidian (mm) Gaobeidian (mm) (a) (b) y = 0.588x + 26.158 y = 0.7744x + 47.29 R = 0.7574 R = 0.7381 200 200 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Gaobeidian (mm) Gaobeidian (mm) (c) (d) Figure 2: Scatter plots for the correlation relationship between Gaobeidian station and surrounding rain gauges. ,e black solid lines are 1 : 1 lines. ,e red dashed lines mean the linear regression lines. relationships between Gaobeidian and other four stations. (0.1 mm≤ daily precipitation< 10 mm); medium (10 mm≤ daily precipitation< 25 mm); large (25 mm≤ daily With highest correlation, Tongzhou station is the “nearest” neighbor station to infill the missing annual precipitation at precipitation< 50 mm); heavy (50 mm≤ daily precipitation). Gaobeidian station. In addition, there is no missing pre- We also define two indices to assess these changes in pre- cipitation data at Tongzhou station. Once the “nearest” cipitation patterns according to the work from Song et al. neighbor station has been identified for each station with [39]. One is precipitation incidence (PI), which means the missing data, using the linear regression equation provided proportion of precipitation days for every precipitation in Table 2, missing data can be infilled. From Table 2, we find grade to the total days of precipitation in a year; the other is that the precipitation series for the nearest neighbor stations precipitation contribution (PC), which means the pro- are quite strongly correlated with those stations that have portion of precipitation amount for every precipitation grade to the total amount of precipitation in a year. missing data; for most stations, correlation coefficients ex- ceed 0.7. Additionally, the rain gauge stations provide reasonably good coverage of different altitudinal zones, i.e., plain (in- cluding urban and suburb areas) and mountainous areas in 2.2. Classification in Precipitation Grades and Subregions. the study area. By considering the complex surface char- In this study, the standard for records is that precipitation acteristics, 30 observation sites were divided into two cat- amount is greater than or equal to 0.1 mm when rainfall egories, plain and mountain areas (Figure 1). ,ere are 18 occurs. To further understand the change of precipitation sites (SLZ, YAM, GBD, TZ, HC, MJQ, SH, SY, SZ, YLZ, ZF, patterns over Beijing, the observed daily precipitation was FS, SJD, TYK, HR, MY, PG, and NGZ) covering the plain categorized into four grades of intensity according to China areas and 12 sites covering the mountain areas. Among these Meteorological Administration (CMA) standards: small stations in the plain areas, four stations located in urban area Youanmen (mm) Tongzhou (mm) Majuqiao (mm) Songlinzha (mm) Advances in Meteorology 5 Table 2: Information for infilling missing data. Station Nearest neighbor Calibration series Correlation coefficient Equation FS ZF 1963–1967, 1969–1991, 1997–2012 0.78 y � 0.849x + 61.5 GBD TZ 1972–2012 0.93 y � 0.827x + 86.5 HC FS 1962–1967, 1969–1991, 1997–2012 0.68 y � 0.537x + 225.8 MJQ TZ 1960, 1962–2012 0.83 y � 0.738x + 65.4 MY ZLY 1960, 1963–2012 0.64 y � 0.9x + 53.78 NGZ YLZ 1979–2012 0.72 y � 0.657x + 87.1 PG ZLY 1960–1991, 1993–2012 0.66 y � 0.519x + 172.2 QJD YQ 1961–1991, 1994–2012 0.69 y � 0.569x + 173.4 SH SLZ 1963–1978, 1980–2012 0.80 y � 0.673x + 160 SLZ TZ 1963–1967, 1969–2012 0.80 y � 0.795x + 121.5 SY SZ 1960–1991, 1997–2012 0.83 y � 0.831x + 80.8 THK HHC 1960–1961, 1963–2012 0.75 y � 0.504x + 164.4 TYK SH 1962–1978, 1980–2012 0.71 y � 0.713x + 203.1 WJY SH 1962–1978, 1980–2012 0.73 y � 0.7682x + 125.5 XH MY 1963–1969, 1974–2012 0.75 y � 0.570x + 267.2 XYL ZT 1960, 1962–1991, 1997–2012 0.77 y � 1.203x + 72.4 YAM TZ 1963–1967, 1969–2012 0.82 y � 0.814x + 89.7 YC ZT 1963–2012 0.82 y � 0.892x + 67.5 YLZ TZ 1961–2012 0.85 y � 0.861x + 24.2 YQ GT 1960–1991, 1994–2012 0.74 y � 0.974x + 83.9 ZF FS 1963–1967, 1969–1991, 1997–2012 0.78 y � 0.708x + 193.7 for a year can be seen as a circle (360 (SLZ, YAM, GBD, and TZ) were used to estimate the mean ). ,en, the yearly PCP precipitation in the urban areas and six surrounding stations and PCD for a location can be defined as follows: (HC, MJQ, SH, SY, SZ, YLZ) were selected to calculate the P � 􏽘 P , corresponding mean precipitation in the suburb areas. P � 􏽘 P · sin θ , x i i 2.3. Precipitation Concentration Index, Concentration Degree, P � 􏽘 P · cos θ , and Concentration Period. ,e precipitation concentration y i i index (PCI) developed by Oliver [40] was used to explore the (2) changing features of precipitation concentration in Beijing. PCP � arctan􏼠 􏼡, PCI was proposed as an indicator of monthly rainfall het- erogeneity, according to the following equation: 􏽱������ � 2 2 P + P 12 2 x y 􏽐 P i i PCD � , PCI � × 100, (1) 12 P 􏼐􏽐 P 􏼑 where P is the monthly precipitation of the ith month, θ is i i where P is the monthly precipitation in month i. According the azimuth of the ith month, and P is the total precipitation to equation (1), the lowest theoretical value of PCI is 8.3, amount. PCP can reflect which month the maximum which means the perfect uniformity in precipitation dis- monthly precipitation appears in. ,e corresponding re- tribution. Oliver [40] classified PCI values in four categories: lation between PCP and month in a year is as follows: ° ° ° ° (1) PCI≤ 10 means uniform precipitation distribution, January (0 ), February (30 ), March (60 ), April (90 ), May ° ° ° ° i.e., low precipitation concentration; (2) 10< PCI≤ 15 means (120 ), June (150 ), July (180 ), August (210 ), September ° ° ° ° moderate precipitation concentration; (3) 15< PCI≤ 20 (240 ), October (270 ), November (300 ), December (330 ). means irregular precipitation distribution, i.e., high pre- PCD can reflect the degree to which annual total pre- cipitation concentration; and (4) PCI> 20 means strong cipitation is distributed in 12 months, which is ranging from irregular precipitation distribution, i.e., very high pre- 0 to 1. cipitation concentration. In addition, the precipitation concentration degree (PCD) and the precipitation concentration period (PCP) 2.4. Trends Detecting Method. A trend analysis was per- proposed by Zhang and Qian [41] were also used to analyze formed to detect gradual changes or tendencies in the time the interannual variations of precipitation amounts. ,e series data, using the Mann–Kendall (M-K) test and Sen’s basic principle for calculating the PCD and PCP is based on slope method. Both are commonly used tests for trend the vector of monthly total precipitation [42]. ,e as- detection [43–51]. Trends were evaluated using the non- sumptions can be made that monthly total precipitation is a parameteric M-K test [44, 45]. Significance of trend was vector quantity with both magnitude and that the direction evaluated at the 0.05 and 0.10 levels. ,e magnitude of trends 6 Advances in Meteorology was evaluated using Sen’s slope [46]. ,e M-K test and Sen’s during 1960–2012 ranges from 383.9 mm (1965) to slope were chosen because they were widely used to detect 797.9 mm (1964) with an average of 553.6 mm. ,e annual the trends of hydrological and meteorological data time mean precipitation has a significantly decreasing trend at a series. ,e details of the M-K test and Sen’s slope are in- rate of 11.6 mm/10a from 1960 to 2012, as analyzed by the troduced in Appendix. linear regression method. For seasonal precipitation, the spring and autumn precipitations have a slightly increasing trend analyzed (4.7 mm/10a and 4.9 mm/10a); in contrast, 2.5. Pettitt Change Point Analysis. Nonstationarity in time summer precipitation has a visibly decreasing trend series can also be characterized by abrupt shifts in the mean (21.1 mm/10a), which is greater than the decline rate in the or variance of the series. To identify possible shifts, standard annual precipitation series. Unlike the above three seasons, change-point procedures have been applied to multiple time the trend of mean precipitation in winter has been almost series in Earth sciences including precipitation and tem- flat, fluctuating within a range from 0.9 to 27.7 mm. perature [52]. One type of homogeneity (change-point) test ,erefore, decreasing precipitation in summer has been the is the Pettitt test [53], which is a nonparametric most important factor in the overall decrease of annual test—requiring no assumption on the underlying distribu- precipitation in Beijing. Several periods of severe droughts tion [54]. and moisture surpluses are evident in the anomaly series of To perform the two-tailed hypothesis test on the location regional mean precipitation in Figure 3. For example, the parameter (mean), the Pettitt test statistic is calculated as long-term dry period during 1999–2011 in Beijing is ⎪ x < x , characterized by a high number of negative anomalies. Si- ⎧ −1, ⎪ i j multaneously, various temporal variations for interannual D � 0, x � x , (3) i j ij and interdecadal variability of seasonal precipitation were x > x , 1, exhibited during 1960–2012 from the precipitation anomaly i j series and the 11-year moving-average series. For instance, where x and x correspond to the magnitude of the i j in the period of 2000–2012, although the precipitation hydroclimatic variable under consideration and x precedes amount in certain years (2008 and 2012) is on the high side x in time. For evaluation over the entire sample (T years), (larger than the mean value during 1960–2012), the summer these D statistics are combined as follows: total precipitation is relatively lower than its long-term t T mean. In contrast, the precipitation amount in spring and U � 􏽘 􏽘 D . (4) autumn in the most years during 2000–2012 is on the high t,T ij i�1 j�t+1 side. ,e M-K test and Sen’s slope estimation were applied to ,e statistic U is equivalent to a Mann–Whitney statistic detect the trends of the monthly, seasonal, and annual for testing that the two samples X , . . ., X and X , . . ., X 1 t t+1 T precipitation series during 1960–2012, as shown in Table 3. come from the same population. ,e test statistic U is For the monthly scale, the series in May had a statistically evaluated for all possible values of t ranging from 1 to T. ,e significant increasing trend, while that in July and August most probable year of a change-point occurring is evaluated had a significant decreasing trend, both at a level of α � 0.05. using a two-tailed test on the following statistic: ,e series in January and February had a significantly de- 􏼌 􏼌 􏼌 􏼌 􏼌 􏼌 K � max􏼌U 􏼌. (5) creasing trend at a level of α � 0.10. All other series had a T t,T nonsignificant trend at both levels during this study period. If the statistic K is significantly different from 0, then a For the seasonal scale, the precipitation series in spring change-point occurs in the year t corresponding to the point showed a significant increasing trend, while that in summer in time for which the largest absolute value of U , is ob- t,T showed a statistically significant decreasing trend, both at a tained. ,e probability of a shift in a year where |U | is the t,T level of α � 0.05. Both the precipitation series in autumn and maximum is estimated by [55] winter had a nonsignificant trend. For the annual scale, there is a nonsignificant trend for the annual precipitation series −6K P � 2 exp . (6) 􏼠 􏼡 3 2 during 1960–2012 at any level of α � 0.05 or α � 0.10. T + T ,e monthly variation of precipitation amounts is shown Given a certain significance level α, if P< α, we reject the in Figure 4. ,e highest precipitation occurred in the wet season (May to October), especially in the period from June null hypothesis and conclude that X is a significant change point at level α. ,e two significant levels (α � 0.05 and to September (namely flood season in Beijing), while the cold season (November to April) received less precipitation. α � 0.10) were used in this work. In this study, the monthly, seasonal, and annual series of regional mean precipitation ,e precipitation in the period from May to October accounted for 91.8% of the annual mean precipitation. and the annual precipitation series for all the stations were used to detect the change points based on the Pettitt method. Additionally, the precipitation in flood season contributed more than 81.5% to the annual mean precipitation. Further, it is clear that a large percentage of total precipitation occurs 3. Results and Discussion in summer, ranging from 39.5% to 87.4%, especially in July (8.3%–58.2%) and August (6.6%–50.6%). ,e maximum and 3.1. Changes of Regional Mean Precipitation. As seen in Figure 3, the annual mean precipitation in Beijing areas minimum values for average monthly precipitation were Advances in Meteorology 7 120 y = 0.4717x – 873.73 500 y = –2.1143x + 4599.11 0 100 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Summer Spring Linear trend Linear trend (a) (b) 200 30 y = 0.4894x – 891.26 y = 0.0083x – 7.41 0 0 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Winter Autumn Linear trend Linear trend (c) (d) y = –1.156x + 2848.71 –50 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Annual Spring Linear trend 11-Years moving average (e) (f) 300 120 –100 –40 –200 –300 –80 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Summer Autumn 11-Years moving average 11-Years moving average (g) (h) 20 300 –100 –5 –200 –10 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Winter Annual 11-Years moving average 11-Years moving average (i) (j) Figure 3: Time series of precipitation and anomaly in the Beijing metropolitan area from 1960 to 2012. 183.9 mm in July and 2.2 mm in December. To further Beijing has strongly uneven monthly precipitation distri- discuss the variation of monthly precipitation, the time bution, where most precipitation occurred in summer, series of PCI, PCD and PCP are shown in Figure 5. As seen i.e., June, July, and August (as shown in Figure 4). Addi- in Figure 5, the PCI in Beijing during 1960–2012 ranges from tionally, the PCI and PCD both have a slight decreasing 13.1 (2003) to 37.4 (1962) with an average of 23.7. ,e PCD trend at a rate of 1.68/10a and 0.02/10a from 1960 to 2012, as ranges from 0.47 (2003) to 0.86 (1994) with the mean of 0.72 analyzed by the linear regression method. To some extent, over the period of 1960–2012. Overall, the precipitation over the decreasing PCI (PCD) may be caused by the significant Precipitation anomaly (mm) Precipitation anomaly (mm) Precipitation (mm) Precipitation (mm) Precipitation (mm) Precipitation anomaly (mm) Precipitation anomaly (mm) Precipitation anomaly (mm) Precipitation (mm) Precipitation (mm) 8 Advances in Meteorology Table 3: Trend test results of the monthly, seasonal, and annual after the change point were examined, as shown in Table 4. precipitation during 1960–2012 tested by the M-K method and Significant abrupt changes in monthly precipitation were Sen’s slope method. ,e uncertainty range covered from 5% to 95% found in May (47% upward in 1976) and August (46% between the minimum and maximum values of Sen’s slope. downward in 1996) at a level of α � 0.05. A significant abrupt change in monthly precipitation was also found in July (31% Z value Sen’s slope (mm/a) Minimum Maximum ∗ downward in 1998) at a level of α � 0.10. Of them, as shown Jan −1.695 −0.009 −0.018 0 in Table 4, the significant downward change points occurred Feb −1.918 −0.043 −0.113 −0.007 in the end of 1990s with the mean precipitation decreasing Mar −1.036 −0.055 −0.111 0.067 Apr 1.365 0.148 −0.060 0.265 by 46% (1996) and 31% (1998), while the significant upward ∗∗ May 2.179 0.379 0.146 0.476 change point in the mid 1970s (1976) with the mean pre- Jun 1.289 0.465 −0.375 0.724 cipitation increasing by 47%. Other monthly precipitation ∗∗ Jul −1.971 −1.227 −2.200 −0.122 series have no significant abrupt change during 1960–2012. ∗∗ Aug −2.148 −1.401 −3.250 −0.892 For the seasonal scale, only summer precipitation has a Sep 0.905 0.248 −0.350 0.550 significant abrupt change in 1996, and the downward Oct 0.721 0.099 −0.225 0.207 magnitude is approximate 25% compared with the mean Nov 0.744 0.037 −0.077 0.075 value of the entire series. Similar to the other three seasons, Dec 1.204 0.011 −0.005 0.033 ∗∗ the annual precipitation has no significant abrupt change at Spring 2.086 0.535 0.025 1.025 ∗∗ either significant level. Summer −2.202 −2.150 −4.136 −0.344 Autumn 1.243 0.569 −0.304 1.250 Winter 0.184 0.007 −0.134 0.102 Annual −0.767 −0.861 −3.159 1.344 3.2. Changes of Spatial Distribution of Precipitation Amount. ∗,∗∗ Note. Statistically significant trends at the 10% and 5% significance ,e spatial distribution of the annual precipitation and their levels, respectively. trend test results is shown in Figure 6. As shown in Figure 6(a), it can be seen from this figure that the average annual precipitation varies greatly in Beijing and it varies from almost 400 mm in the west to more than 600 mm in the east. From the spatial perspective, the distribution shows that the precipitation is decreasing from east to west due to orographic influence. Apparently, the annual precipitation in the Beijing Plain is greater than in the mountainous areas. But the highest value of precipitation frequently occurred in the front belts (e.g., HR and MY stations) between the plain and mountainous areas. As a whole, there was one local maximum of precipitation with an annual mean pre- cipitation of 650 mm, located in the vicinity of the Huairou 100 and Miyun reservoirs in the northeastern section of Beijing. Another local maximum was centered at the XYL stations in the Fangshan district in the southwestern part of Beijing. With regard to the interannual variability of precipitation, the coefficient of variation and standard deviation of annual Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec precipitation have been identified for the period of analysis, Figure 4: Box-plot for monthly precipitation amount during as shown in Figures 6(b) and 6(c). ,e range of values over 1960–2012. ,e boxes indicate the 25th, 50th, and 75th percentiles Beijing is from 0.199 to 0.380. ,e lowest values appear in the and the whiskers indicate the values with maximum 1.5 northwestern area, and the highest values appear in the front interquartile ranges. ,e hollow square indicates the mean value. belts between the plain and mountainous areas, which is ,e solid circle means the outlier. similar to the spatial pattern of annual mean precipitation. ,e spatial distribution of the standard deviation in pre- decreases in precipitation in summer and the increases in cipitation also shows the similar pattern to the annual precipitation. ,is explains the standard deviation (co- spring and autumn for the Beijing area (as shown in Fig- ure 3). According to the changes of PCP in Figure 5, the efficient of variation) increases with the annual mean pre- cipitation, namely, the higher the annual precipitation is, the range of the mean yearly PCPs in Beijing is 189± 7, implying that annual precipitation mainly falls in summer (June-July- higher the degree of uncertainty (standard deviation and August). ,e time of the PCP mainly appears from the end of coefficient of variation) is. ,is might increase the degree of June (23, June) to the end of July (26, July). ,e maximum uncertainty in the availability of regional water resources, monthly precipitation in most years (46 years) occurs in July due to the close relationship between the precipitation and the other occurs in June (7 years). amount and the regional water resources. Change points in the monthly, seasonal, and annual ,e spatial distribution of Z values of the M-K trend test, Sen’s slope, and linear trend for the individual rain gauges is series over Beijing were detected based on the Pettitt method. ,en, the changes in the two samples before and also shown in Figure 6. ,e annual trends found by the linear Precipitation (mm) Advances in Meteorology 9 y = –0.168x + 357 10 0.9 0.8 0.7 0.6 0.5 y = –0.002x + 4.87 210 0.4 July June 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Years Figure 5: Time series of PCI, PCD, and PCP over the period of 1960–2012. ,e red solid line indicates the linear regression trend. ,e red dashed line means the boundary between the months June and July for the PCP. If the PCP is lower than 180, the maximum monthly precipitation appears in June. If the PCP ranges from 180 to 210, the maximum value appears in July. Table 4: Results of change point analysis for different series based on the Pettitt test. Entire series (mm) Change point p Prechange mean (mm) Postchange mean (mm) Ratio of changes Jan 2.2 2001 0.487 2.3 1.8 −0.23 Feb 4.6 1986 0.166 5.3 3.9 −0.30 Mar 9.0 1991 0.612 9.4 8.4 −0.11 Apr 20.5 1997 0.327 19 24.4 0.26 ∗∗ May 33.6 1976 0.042 22.9 38.6 0.47 Jun 77.2 1975 0.295 62.2 83.7 0.28 Jul 180.7 1998 0.060 195.7 139.2 −0.31 ∗∗ Aug 142.2 1996 0.024 161.8 96.7 −0.46 Sep 50.6 1985 0.540 44.7 56.3 0.23 Oct 22.2 1995 0.424 20 26.9 0.31 Nov 7.8 1967 1.121 3.2 8.6 0.69 Dec 2.2 1973 0.124 0.8 2.7 0.86 Spring 63.1 1976 0.147 49.8 69.4 0.31 ∗∗ Summer 400.1 1996 0.043 430.4 330.1 −0.25 Autumn 80.7 1988 0.690 73 89.9 0.21 Winter 8.9 1996 1.380 8.5 9.8 0.15 Annual 553.0 1996 0.398 573.5 505.6 −0.12 Note. ,e ratio of change is equal to the differences of mean precipitation between prior series and posterior series dividing the mean precipitation at the ∗ ∗∗ entire series. Significant abrupt change at a level of α � 0.10; significant abrupt change at level of α � 0.05. regression were almost similar to the precipitation trends Beijing. From the distribution of the annual precipitation found by the M-K test and Sen’s slope estimator. Both trend, we can see that the highest value of precipitation decline mostly occurred in the southwest part of Beijing with positive and negative trends were identified by the statistical tests in annual precipitation data. Only one station (ZT) a rate of more than 2 mm/a. From the linear trend showed significantly decreasing trend at a level of α � 0.05. (Figure 6(f)), only three stations (NGZ, SZ, and ZJF) have All the other stations showed no significant decreasing or slight increasing trend (less than 1 mm/a) and the other increasing trend at the 95% confidence level. For these stations have decreasing trend with an average of 12.2 mm/ stations with insignificant trend, seven stations showed a 10a. ,e WJY, XYL, and TYK stations have a clear de- slight increasing trend and twenty-two stations showed a creasing trend with a value larger than 20 mm/10a. decreasing trend. ,e declining trend of most stations was Many studies have reported that changes in precipitation similar to the change of regional annual precipitation over pattern are associated with elevation changes, although the PCP PCD PCI 10 Advances in Meteorology 116°E 117°E 116°E 117°E 41°N N N 41°N THK S THK S QJD QJD ZJF XN ZJF XH YQ YQ HHC MY HHC MY ZY ZY HR GT TYK GT TYK WJY WJY PG PG SH SH SY SY SZ SZ YC YC ZT 40°N SJD SLZ ZT 40°N SJD SLZ TZ GBD TZ GBD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF NGZ NGZ 0 10 20 40 0 10 20 40 km km 429.6 to 475.0 559.0 to 582.5 0.199 to 0.237 0.291 to 0.305 475.1 to 506.5 582.6 to 607.8 0.238 to 0.258 0.306 to 0.334 506.6 to 535.3 607.9 to 652.4 0.259 to 0.276 0.335 to 0.380 535.4 to 558.9 0.277 to 0.290 (a) (b) 116°E 117°E 116°E 117°E 41°N 41°N W E THK THK QJD QJD ZJF ZJF XH XH YQ YQ HHC HHC MY MY ZIY ZY HR HR GT GT TYK TYK WJY WJY PG PG SH SH SY SY SZ SZ YC YC ZT ZT SJD 40°N SJD SLZ 40°N SLZ TZ TZ GBD GBD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF NGZ NGZ 0 10 20 40 0 10 20 40 km km –2.09 to –1.96 78.5 to 115.6 173.2 to 191.9 115.7 to 149.6 192.0 to 243.7 –1.95 to 0 149.7 to 173.1 0 to 0.67 (c) (d) Figure 6: Continued. EW EW EW Advances in Meteorology 11 116°E 117°E 116°E 117°E 41°N 41°N W E W E THK THK S QJD QJD ZJF XH ZJF XH YQ YQ HHC MY HHC MY ZIY ZIY HR HR GT TYK TYK WJY GT WJY PG PG SH SY SH SY SZ YC SZ YC ZT ZT 40°N SJD SLZ 40°N SJD SLZ TZ GBD TZ GBD YAM YAM YLZ MJQ YLZ XYL MJQ HC FS XYL HC FS ZF ZF NGZ NGZ 0 10 20 40 0 10 20 40 km km Sen’s slope (mm/a) mm/10a –2.3 to –2.0 –0.9 to 0 –27.4 to –20.0 –9.9 to 0 –1.9 to –1.0 0 to –1.0 –19.9 to –15.0 0 to 9.3 –14.9 to –10.0 (e) (f) Figure 6: ‚e spatial distribution of mean precipitation and their statistical characteristics and trends in Beijing during 1960–2012: (a) the annual mean precipitation, (b) the coeƒcient of variation, (c) the standard deviation, (d) the M-K test, (e) Sen’s slope, and (f) the linear trend. relationships can vary regionally. To investigate the re- series with a negative change for all of these 21 stations. As lationship between the precipitation characteristics and el- discussed in Section 3.1, Beijing su˜ered from the contin- evation, we calculated the correlations between the 53-year uous drought with continued 12 years since the end of 1990s, (1960–2012) average precipitation amount and all station which is consistent with the results of change point analysis. elevations, as shown in Figure 7. ‚e results show that standard deviation and coeƒcient of variation are negatively and signi…cantly correlated with elevation, with Pearson 3.3. Changes of Dierent Precipitation Grades in the Flood correlation coeƒcients higher than 0.5 (p < 0.01). A weak Season. According to the results of PCI (Figure 5), we know that the uneven distribution of precipitation in Beijing is but signi…cant negative relationship is also observed for precipitation amount with correlation of 0.45 (p < 0.05). evident. We can see that the most of precipitation amount in ‚ese strong correlations between the precipitation and Beijing happened in the šood season (June-July-August- elevation indicate that local elevation has important impacts September). In this study, the changes of precipitation on the magnitude of precipitation in Beijing. Box plots show grades were discussed based on the daily precipitation data that the largest mean values of precipitation and standard in the šood season. Summary statistics of precipitation deviation appear at 100∼200 m, while the highest value of frequency, amount, and intensity for di˜erent grades of coeƒcient of variation occur at 0∼100 m. Overall, therefore, precipitation are presented in Table 6. ‚e mean number of the precipitation decreases with altitude, and higher pre- precipitation frequency is largest for small precipitation, with a value of 27, accounting for 67.8% of the total rainy cipitation events occur at lower altitudes in this area with higher coeƒcient of variation and standard deviation. days. ‚e lowest number of days belongs to the heavy precipitation category, at about 1–2 days every year during Regarding the analysis of the change points in the mean, the results for all the 30 rain gauges are summarized in the šood season. ‚e characteristics of precipitation amount Table 5 using the Pettitt test. As shown, change from positive di˜er from that of the frequency for di˜erent grades of to negative direction was detected for the 25 stations with a precipitation. ‚e lowest value of precipitation amount ratio of change from 7% to 22%. Only two stations (PG and belongs to the small precipitation, although its frequency is ZT) presented signi…cant change point at a level of α  0.10, the highest. For the other three categories, their precipitation which occurred in the years 1997 and 1996, respectively. We amounts are almost equivalent, at about 120 mm per year. found most of change points (21 stations) detected based on However, the standard deviation of precipitation amount for Pettitt test occurred in the end of 1990s (1996–1998) without heavy precipitation is largest. ‚is explains the interannual variation for heavy precipitation is relatively larger than any statistical signi…cant levels. ‚e average change was about 16% of the annual mean precipitation in the entire other categories. To a certain extent, the high variation of 12 Advances in Meteorology 700 700 R = –0.447 p = 0.013 600 600 500 500 400 400 300 300 0 100 200 300 400 500 0~100 100~200 200~300 300~400 400~500 Elevation (m) Elevation (m) (a) (b) 0.40 0.40 0.36 0.36 R = –0.643 p < 0.01 0.32 0.32 0.28 0.28 0.24 0.24 0.20 0.20 0 100 200 300 400 500 0~100 100~200 200~300 300~400 400~500 Elevation (m) Elevation (m) (c) (d) 250 250 R = –0.578 200 200 p < 0.01 150 150 100 100 50 50 0 100 200 300 400 500 0~100 100~200 200~300 300~400 400~500 Elevation (m) Elevation (m) (e) (f) Figure 7: Scatter and box-plots in the correlations between precipitation characteristics and elevation. Solid red line is the linear trend. R signifies the Pearson correlation coefficient for each relationship and p indicates the statistical significance. ,e solid square means the mean value during the different elevation bands. heavy precipitation might increase the degree of uncertainty precipitation is larger than that of small precipitation. ,e of precipitation amount. A relatively obvious decline in largest value of PC is the medium precipitation, ranging precipitation amount for the heavy precipitation is shown from 18% to 40% with a mean of 29%. And the second based on the linear regression coefficient. ,is change might belongs to the large precipitation, ranging from 18% to 37% cause the change in precipitation amount in the flood season with a mean of 27%. Additionally, the broad bound of PC over Beijing. (5–45%) for the heavy precipitation explains the relatively Figure 8 also shows the ranges of precipitation frequency larger variation and higher degree of uncertainty. ,e de- and amount for different grades. ,e frequency ranges from tailed information about the PI and PC for 30 stations is shown in Figure S1. 16.9 to 39.4, 3.9 to 10.7, 1.7 to 6.1, and 0.2 to 3.1 for the small, medium, large, and heavy precipitations, respectively. ,e In order to further discuss the spatial patterns of pre- cipitation frequency and amount for the different grades, amount is ranging from 45 to 118.6 mm, 60 to 172.7 mm, 54.7 to 212.7 mm, and 16.7 to 314.2 mm for the different Figure 9 shows the spatial interpolation results of both grades, respectively. In this work, PI and PC are used to indices for different grades. Overall, there are visible dif- further analyze the changes in different grades of pre- ferences among the spatial patterns for the different grades cipitation, as also shown in Figure 8. Overall, the largest of precipitation. For example, the frequency is increasing value of PI belongs to the small precipitation, accounting for from east to west for the small precipitation, but it is in- 67% (54–77%) of all the rainy days. ,e mean values of PI for creasing from south to north for the medium precipitation. large and heavy precipitations are both smaller than 10% From the perspective of precipitation amount, there are (9% and 4%). However, the value of PC for heavy similar distribution patterns for the small and medium Standard deviation Coefficient of variation Annual precipitation (mm) Standard deviation Coefficient of variation Precipitation (mm) Advances in Meteorology 13 Table 5: Results of change-point detection for different stations based on the Pettitt test. Stations Entire series Change point p Prior series Posterior series Ratio of change FS 568.6 1996 0.229 599.0 496.4 −0.17 GBD 570.2 1998 0.494 591.5 509.3 −0.13 GT 374.4 1993 1.553 369.6 383.8 0.03 HR 641.7 1998 0.190 678.4 549.8 −0.19 HC 526.0 1998 0.773 546.3 467.9 −0.14 HHC 593.7 1998 0.465 617.9 524.6 −0.14 MJQ 496.8 1998 0.373 520.9 426.6 −0.17 MY 641.1 1998 0.672 675.1 543.9 −0.19 NGZ 437.1 2006 0.604 422.6 550.2 0.26 PG 511.3 1997 0.060 541.9 431.7 −0.20 QJD 429.7 1992 0.571 440.1 404.8 −0.07 SJD 577.1 1996 0.921 597.6 527.7 −0.11 SH 539.1 1972 0.841 498.6 553.2 0.09 SY 564.3 1998 0.502 598.4 466.9 −0.21 SLZ 586.0 1998 0.344 612.0 511.6 −0.16 SZ 580.8 1998 0.773 602.4 522.1 −0.13 THK 464.0 1998 0.295 484.6 404.9 −0.15 TYK 604.6 1998 0.130 642.1 497.3 −0.22 TZ 584.1 1998 0.487 611.0 507.1 −0.16 WJY 550.0 1979 0.338 622.1 504.1 −0.20 XYL 612.0 1996 0.183 643.8 537.9 −0.16 XH 628.9 1975 1.046 573.6 654.2 0.12 YQ 449.0 1973 0.672 492.4 433.2 −0.12 YC 467.4 1979 0.213 513.8 438.8 −0.14 YAM 565.3 1998 0.285 591.4 490.7 −0.16 YLZ 527.3 1998 0.587 547.6 469.2 −0.14 ZT 448.6 1996 0.094 471.8 393.8 −0.15 ZF 598.2 1996 1.282 614.0 564.6 −0.08 ZJF 618.5 1975 0.451 561.5 644.7 0.12 ZLY 652.4 1998 0.176 689.4 546.9 −0.20 Note. ,e ratio of change is equal to the differences of mean precipitation between prior series and posterior series dividing the mean precipitation at the entire series. Significant abrupt change at a level of α � 0.10. Table 6: Summary statistics for the precipitation frequency, amount, and intensity for the different grades. Frequency (days) Amount (mm) Intensity (mm/day) Mean SD RC (days/10a) Mean SD RC (mm/10a) Mean SD RC (mm/(day·10a)) Small 27.0 3.37 −0.103 80.5 8.50 −0.644 3.0 0.33 −0.033 Medium 7.7 0.76 0.051 122.4 12.03 0.36 15.9 0.23 −0.071 Large 3.6 0.60 −0.148 124.4 21.49 −5.033 34.3 0.55 −0.009 Heavy 1.5 0.50 −0.13 121.7 42.61 −10.506 79.6 5.19 −1.255 Note. SD, standard deviation; RC, regression coefficient. precipitation. ,e precipitation amount of small pre- events, both have negative and significant correlations with cipitation ranges from 65.1 to 95.9 mm, which declines from elevation at a confidence level of 0.01. In conclusion, to a northwest to southeast. However, it declines from north to certain extent, the spatial distributions of precipitation south for the medium precipitation, ranging from 95.9 to frequency and amount are affected by the terrain. ,e large 148.9 mm. Additionally, the spatial distribution of pre- and heavy precipitations frequently occur in the plain areas, cipitation frequency and amount for the large and heavy especially for the piedmont plain areas. But, in contrast, the precipitation of the mountain areas is dominated by the precipitation are almost similar to the distribution of the annual precipitation (Figure 6(a)). ,e relationship between small and medium precipitations both in frequency and the precipitation frequency and amount for different pre- amount. cipitation grades and elevation was also discussed in this work, as shown in Figure 10. For small and medium pre- cipitation events, both frequency and amount have positive 3.4. Differences in Changes of Precipitation in Four Subregions. correlations with elevation, with no significance for fre- ,e variations of annual precipitation series for different quency in small precipitation and amount in medium subregions (i.e., plain area, mountain area, urban area, and precipitation events. While for large and heavy precipitation suburban area) are shown in Figure 11, with the statistical 14 Advances in Meteorology 0.8 0.3 0.2 0.7 0.1 0.6 0 0.0 0.5 Small Medium Large Heavy Small Medium Large Heavy (a) (b) 0.5 0.4 0.3 0.2 0.1 0 0.0 Small Medium Large Heavy Small Medium Large Heavy (c) (d) Figure 8: Box-plots of precipitation frequency (a), precipitation amount (b), precipitation incidence (c), and precipitation contribution (d) for different grades of precipitation over Beijing during 1960–2012. ,e boxes indicate the 25th, 50th, and 75th percentiles and the whiskers indicate the values with maximum 1.5 interquartile ranges. ,e hollow square indicates the mean value. ,e solid circle means the outlier. results given in Table 7. As a whole, we found that the difference has expanded to 32.6 mm (accounting for 5.72% of the total precipitation in the plain areas) between the areas average precipitation amount in the area of the Beijing Plain (urban area) was greater than that in the mountainous in the plain and mountains and 45.2 mm (accounting for (suburban) areas. As seen in Figure 11, the precipitation 7.66% of the total precipitation in the urban areas) between showed slight decreasing trends at the rate of 6.4 and the urban areas surrounding suburb areas. It is clear that the 9.9 mm/decade for the plain and mountain areas average precipitation decreased both in the urban and (Figure 11(a)), 4 and 4.9 mm/decade for the urban and suburb areas through these two periods, while the drying was suburban areas (Figure 11(c)) during 1960–2012, re- larger in the suburb areas, which caused the higher differ- spectively. However, as shown in Figure S2, from the ences in precipitation between the urban and surrounding standpoint of the changes in built-up areas, the urban ex- suburb areas. With rapid development of urbanization, the height and density of buildings are both increasing, and the pansion has experienced three stages in Beijing during the research period: the first stage is the slowest urban expansion rainfall under a weather system in urban areas would stay longer than that in open suburb areas [56]; thus, the total before 1980, the second is the rapid expansion stage from 1980 to 2000, and the third is the fastest urban growing precipitation increased. Additionally, in the process of ur- period during the first decade of the 21st century [3]. banization, population density increases in cities, releasing ,erefore, the changes in precipitation during the three more heat into the atmosphere and changing the impervi- stages for these areas are also discussed. During 1960–1979, ousness of underlying areas. All these factors lead to an the difference of the precipitation amount between areas in urban heat island effect, leading to precipitation variations in the plain and the mountains was about 24.2 mm, which urban areas. However, during 2000–2012, the precipitation accounts for 4.15% of the total precipitation in the plain amount in the above four areas has declined by 49.8– areas, while the difference between urban areas and suburb 67.9 mm compared with the last period due to the long-term drought (1999–2010) in the north China, accounting for areas was 32.9 mm, accounting for 5.51% of the total pre- cipitation in the urban areas. During 1980–1999, the 8.5–11.6% of the average annual precipitation (584.7 mm) Frequency (days) Precipitation amount (mm) PI PC Advances in Meteorology 15 117°E 117°E 116°E 116°E N N 41°N 41°N W E W E THK THK QJD QJD ZJF XH ZJF XH YQ YQ HHC HHC MY MY ZLY ZLY HR HR GT YK YK GT WJY WJY PG PG SH SH SY SY SZ SZ YC YC 40°N 40°N ZT ZT SJD SJD SLZ SLZ TZ TZ GRD GRD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF GZ GZ 0 20 40 10 0 10 20 40 km km Frequency-small (days) Frequency-medium (days) 26.9–28.0 7.9–8.1 17.9–21.6 6.0–6.6 28.1–29.3 8.2–8.5 21.7–23.8 6.7–7.1 29.4–31.2 8.6–9.2 23.9–25.4 7.2–7.5 25.5–26.8 7.6–7.8 (a) (b) Figure 9: Continued. 116°E 117°E 117°E 116°E 41°N 41°N W E W E THK THK QJD QJD ZJF XH ZJF XH YQ YQ HHC MY HHC MY ZLY ZLY HR HR GT YK GT YK WJY WJY PG PG SH SH SY SY SZ SZ YC YC 40°N 40°N ZT SJD ZT SJD SLZ TZ SLZ TZ GRD GRD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF GZ GZ 0 10 20 40 0 10 20 40 km km Frequency-large (days) Frequency-heavy (days) 3.8–3.9 1.5–1.7 2.4–2.8 0.5–0.8 4.0–4.2 1.7–1.9 2.9–3.1 0.9–1.0 4.3–4.9 1.9–2.2 3.2–3.4 1.0–1.3 3.56–3.7 1.3–1.5 (c) (d) Figure 9: Continued. 16 Advances in Meteorology 116°E 117°E 116°E 117°E 41°N 41°N W E W E THK THK QJD QJD ZJF XH ZJF XH YQ YQ HHC HHC MY MY ZLY ZLY HR HR GT YK GT YK WJY WJY mm PG PG SH SH SY SY SZ SZ YC YC 40°N 40°N ZT SJD ZT SJD SLZ TZ SLZ TZ GRD GRD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF GZ GZ 0 20 40 10 0 10 20 40 km km Amount-small (mm) Amount-medium (mm) 83.3–86.5 124.2–129.1 65.1–71.4 95.9–106.3 86.6–89.8 129.2–136.6 71.5–75.1 106.4–114.1 89.9–95.9 136.7–148.9 75.2–79.2 114.2–119.3 79.3–83.2 119.4–124.1 (e) (f) 116°E 117°E 116°E 117°E 41°N 41°N W E W E THK THK QJD QJD ZJF XH ZJF XH YQ YQ HHC MY HHC MY ZLY ZLY HR HR GT YK GT YK WJY WJY PG PG SH SH SY SY SZ SZ YC YC 40°N 40°N ZT ZT SJD SJD SLZ SLZ TZ TZ GRD GRD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF GZ GZ 0 40 0 10 20 40 10 20 km km Amount-large (mm) Amount-heavy (mm) 128.2–137.7 119.4–134.2 79.7–97.9 39.5–66.1 134.3–149.7 137.8–148.8 66.2–84.4 98.0–107.8 148.9–170.5 149.8–180.8 107.9–118.5 84.5–102.6 102.7–119.3 118.6–128.1 (g) (h) Figure 9: Spatial distribution of precipitation frequency and amount for the different grades during 1960–2012. (a, e) Small precipitation. (b, f) Medium precipitation. (c, g) Large precipitation. (d, h) Heavy precipitation. over Beijing. During this period, Table 7 showed that the slightly increasing trend at a rate of 3.49 mm/decade during difference dropped to 17.7 mm (accounting for 3.51% of the the period of 1960–2012 (Figure 12(a)). ,e increasing rate total precipitation in the plain areas) between the plain and for the difference between the plain and mountain areas was mountains and 31.0 mm (accounting for 5.94% of the total larger than that for the difference between urban and precipitation in the urban areas) in the urban and sur- suburban areas (0.87 mm/decade) (Figure 12(b)). However, rounding suburb areas. Moreover, the difference in pre- the results of M-K test and Sen’s slope estimate showed that cipitation between the plain and mountain areas had a there was a nonsignificant decreasing trend for the Advances in Meteorology 17 35 5.0 2.5 4.5 2.0 4.0 1.5 3.5 3.0 1.0 2.5 R = –0.62 R = –0.722 R = 0.388 0.5 R = 0.348 p < 0.01 p < 0.01 2.0 p = 0.034 p = 0.060 15 1.5 0.0 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 Elevation (m) Elevation (m) Elevation (m) Elevation (m) (a) (b) (c) (d) 100 200 R = –0.65 40 R = 0.302 80 R = –0.706 R = 0.755 p < 0.01 p = 0.105 p < 0.01 p < 0.01 60 90 60 0 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 Elevation (m) Elevation (m) Elevation (m) Elevation (m) (e) (f) (g) (h) Figure 10: Spatial distribution of precipitation frequency and amount for the different grades during 1960–2012. (a, e) Small precipitation. (b, f) Medium precipitation. (c, g) Large precipitation. (d, h) Heavy precipitation. 800 800 y = 6.64x – 12501.04 y = 7.64x – 14629.68 y = 21.34x – 42302.58 y = –0.64x + 1834.38 600 600 y = –0.99x + 2502.44 y = 5.06x – 9524.98 y = 4.94x – 9162.75 300 300 y = 6.99x – 13526.99 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Plain area Linear trend (plain area) Plain area Linear trend (plain area) Mountain area Linear trend (mountain area) Mountain area Linear trend (mountain area) (a) (b) Figure 11: Continued. Precipitation amount (mm) Precipitation frequency (days) Precipitation (mm) Precipitation amount (mm) Precipitation frequency (days) Precipitation (mm) Precipitation amount (mm) Precipitation frequency (days) Precipitation amount (mm) Precipitation frequency (days) 18 Advances in Meteorology 900 900 y = 4.08x – 7435.81 800 800 y = –0.40x + 1369.77 y = 10.50x – 20307.50 y = 29.32x – 58289.38 700 700 600 600 500 500 400 400 y = –0.49x + 1505.99 y = 7.62x – 14618.98 300 300 y = 22.05x – 43737.46 y = 7.89x – 14969.22 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Urban area Linear trend (urban area) Urban area Linear trend (urban area) Suburban area Linear trend (suburban area) Suburban area Linear trend (suburban area) (c) (d) Figure 11: Time series of annual mean precipitation in the plain and mountain areas (top panel) and the urban and suburb areas (bottom panel). ,e linear trend analyses are discussed based on the different periods: one (left panel) is based on the whole period (1960–2012) and another one (right panel) is based on the three subperiods (1960–1979, 1980–2000, and 2001–2012) considering the different urban expansion stages. Table 7: Comparison of mean precipitation in the different subregions and periods. 1 1 Plain area Mountainous area Percent of differences (%) Urban area Suburb area Percent of differences (%) 1960–1979 583.2 559.0 4.15 597.0 564.1 5.51 1980–1999 569.6 537.0 5.72 589.9 544.7 7.66 2000–2012 504.8 487.1 3.51 522.0 491.0 5.94 1960–2012 558.8 533.0 4.62 575.9 538.9 6.42 Note. ,e percent of differences � the differences in precipitation between the plain (urban) and mountainous (suburb) areas divided by the total amount of precipitation in the plain (urban) areas. differences in precipitation between the urban and suburban ,ese changes, to some extent, lead to the significant change areas. in spring and summer. Overall, there is no change in a strongly uneven spa- tiotemporal distribution in precipitation within the Beijing 4. Conclusions metropolis. Although the PCI and PCD both have a de- In this study, a comprehensive investigation was performed creasing trend, they are both in a relatively high level, which on the precipitation changes across Beijing based on the is an inherent characteristic for the East Asian Monsoon monthly, seasonal, and annual mean precipitation from 30 areas. ,ese declines may be dominantly caused by the rain gauges from 1960 to 2012. Many time series analysis significant decrease in precipitation in summer and the methods, evaluating indicators, and tools were used to assess increases in spring and autumn. Simultaneously, the pre- the variations of precipitation structure. Based on our cipitation roughly decreases from east to west due to the analysis, main findings are summarized as follows. orographic effects, with the highest rain occurring at the ,e annual mean precipitation has a clear decreasing Miyun and Huairou reservoirs in the northeast and the lowest in the northwestern mountain areas. In addition, the trend at a rate of 11.6 mm/10a during the period of 1960– 2012. Seasonally, a significant decrease in precipitation effects of urban development on precipitation become more and more obvious. occurred in summer (21.1 mm/10a). In contrast, the pre- cipitation in spring shows a significant increasing trend with ,e spatiotemporal changes for the four precipitation a rate of 4.7 mm/10a. ,ere are nonsignificant trends in grades (small, median, large, and heavy) are different. In autumn and winter. However, this increase in spring is general, the frequency for the small precipitation is the unable to offset the remarkable decrease in summer, which is largest, followed by the median and large precipitations, and a dominant driver for the decline in annual precipitation. that for the heavy precipitation is the lowest. On average, the Similarly, the monthly series of precipitation in January, amount for the small precipitation is the lowest and the February, July, and August show a significant decreasing other three grades are almost equivalent. ,e spatial dis- trend while that in May has a statistically significant increase. tribution for different grades are affected by the terrain; the Precipitation (mm) Precipitation (mm) Advances in Meteorology 19 250 250 y = 0.349x – 668.06 y = 0.087x – 136.22 –50 –50 –100 –100 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years (a) (b) Figure 12: Differences in precipitation and their linear trends: (a) between the plain and mountain areas, (b) between the urban and suburban areas. n−1 n large and heavy precipitation frequently occur in the plain S � 􏽘 􏽘 sgn x − x , areas, while the precipitation in the mountain areas is 􏼐 􏼑 (A.1) 0 j k k�1 j�k+1 dominated by the small and medium precipitations. In the current context of global climate change and rapid where n is the length of the series, k � 1, 2, . . ., n− 1, j � 2, urban development in Beijing, adaptations for avoiding or 3, . . ., n, and minimising their impacts on water resources require analysis of the change in precipitation patterns on a fine spatio- ⎧ ⎪ 1, 􏼐x − x 􏼑> 0, ⎪ j k temporal scale. ,e adverse natural conditions and the rapid ⎨ sgn􏼐x − x 􏼑 � 0, 􏼐x − x 􏼑 � 0, (A.2) j k j k urban development in Beijing lead to the high risk in the ⎪ urban water resources management. For example, Beijing ⎩ −1, 􏼐x − x 􏼑< 0. j k suffers the problems both in water deficit and urban flood and waterlogging in recent years. To some extent, these It has been proven that when n≥ 8, S follows approx- problems are probably caused by the change in precipitation imately the normal distribution with 0 mean and the var- pattern. ,erefore, the future work should be paid attention iance as to the relationship between the change in precipitation and [n(n− 1)(2n + 5)] the urban water resources safety, in order to obtain deep (A.3) Var S � . understanding and knowledge. Standardized statistic Z can be Calculated as Appendix S − 1 ⎧ ⎪ 􏽱������� ⎪ , S > 0, Var S 􏼁 Mann–Kendall trend test and Sen’s slope estimate method ⎪ were widely used to detect trends in hydrological and me- teorological time series. ,e detailed information was in- Z � 0, S � 0, (A.4) troduced as follows: In the M-K trend test, the null hypothesis (H ) is that the 0 ⎪ S + 1 data in a time series are independent and identically distributed ⎪ 􏽱������� , S < 0. ⎪ 0 random variables and the hypothesis (H ) is that there is a trend 1 ⎩ Var S 􏼁 in the series. Statistical parameter (S ) is defined as Difference in precipitation (mm) Difference in precipitation (mm) 20 Advances in Meteorology Negative Z value denotes downward trend and positive Z National Natural Science Foundation of China (51609242), value shows upward trend. ,e trend is significant at the 95% the National Key Research and Development Program confidence level |Z|> 1.96 and vice versa. (2017YFC1502701), the China Postdoctoral Science Foun- In Sen’s slope estimator method, the slope of trend in the dation (2018M632333), the Open Research Fund Program of sample of N pairs of data is calculated by State Key Laboratory of Water Resources and Hydropower x − x Engineering Science (2015SWG02), and the Open Research j k Q � , for i � 1, 2, . . . , N, (A.5) i Fund Program of State Key Laboratory of Hydrology-Water j− k Resources and Hydraulic Engineering (2015490411). ,e authors would like to thank the Beijing Hydrologic Stations where x and x are the data values at times j and k (j> k), j k respectively. If there is only one datum in each time period, of Beijing Water Authority in China for providing all of the precipitation data. then N � n(n− 1)/2, where n is the number of time periods. If there are multiple observations in one or more time periods, then N< n(n− 1)/2, where n is the total number of Supplementary Materials observations. ,e N values of Q are ranked from smallest to largest Figure S1: the incidence (PI) and contribute (PC) of different and the median of slope is computed as precipitation grades to the total frequency and amount for the 30 stations during 1960–2012. Figure S2: the urban built- Q , if N is odd, ⎧ ⎪ [(N+1)/2] up areas in Beijing during 1950–2012. (Supplementary Q � (A.6) med Materials) ⎪ Q + Q [N/2] [(N+2)/2] , if N is even. References ,e Q sign reflects data trend reflection, while its med value indicates the steepness of the trend. To determine [1] C. Chen and W. Huang, “Hydrological modeling of typhoon- whether the median slope is statistically different from zero, induced extreme storm runoffs from Shihmen watershed to one should obtain the confidence interval of Q at specific reservoir, Taiwan,” Natural Hazards, vol. 67, no. 2, pp. 747– med probability. ,e confidence interval about the time slope can 761, 2013. [2] H. Chang and W. T. Kwon, “Spatial variations of summer be computed as follows: 􏽱������� precipitation trends in South Korea, 1973–2005,” Environ- C � Z Var S , (A.7) 􏼁 mental Research Letters, vol. 2, no. 4, article 045012, 2007. α 1−α/2 0 [3] X. Song, J. Zhang, A. AghaKouchak et al., “Rapid urbanization where Z is obtained from the standard normal distri- and changes in spatiotemporal characteristics of precipitation 1−α/2 in Beijing metropolitan area,” Journal of Geophysical Research: bution table. In this study, the confidence interval was Atmospheres, vol. 119, no. 19, pp. 11250–11271, 2014. computed at a significant level of α � 0.05. ,en, M � [4] Y. Zhai, Y. Guo, J. Zhou, N. Guo, J. Wang, and Y. Teng, “,e (N− C )/2 and M � (N + C )/2 are computed. ,e lower and α 2 α spatio-temporal variability of annual precipitation and its upper limits of the confidence interval, Q and Q , are min max local impact factors during 1724-2010 in Beijing, China,” th th the M largest and the (M + 1) largest of the N ordered 1 2 Hydrological Processes, vol. 28, no. 4, pp. 2192–2201, 2014. slope estimates. ,e slope Q is statistically different from med [5] S. Dawadi and S. Ahmad, “Changing climatic conditions in zero if the two limits have similar sign. the Colorado River Basin: implications for water resources management,” Journal of Hydrology, vol. 430-431, pp. 127– Data Availability 141, 2012. [6] V. Kumar, S. K. Jain, and Y. Singh, “Analysis of long-term Rainfall data supporting this article are available from the rainfall trends in India,” Hydrological Sciences Journal, vol. 55, Hydrological Data of Haihe River Basin (in Chinese) and the no. 4, pp. 484–496, 2010. [7] A. Kalra and S. Ahmad, “Evaluating changes and estimating Annual Hydrological Report of China (Volume III), released seasonal precipitation for the Colorado River Basin using a by Ministry of Water Resources of China. For further in- stochastic nonparametric disaggregation technique,” Water formation or right to access to the material used in this Resources Research, vol. 47, no. 5, article W05555, 2011. paper, readers can also contact the Beijing Hydrological [8] S. Sagarika, A. Kalra, and S. Ahmad, “Evaluating the effect of Stations of the Beijing Water Authority (http://www.bjwater. persistence on long-term trends and analyzing step changes in gov.cn). For the other data used in this article, readers can streamflows of the continental United States,” Journal of contact the corresponding author X. Song. Hydrology, vol. 517, pp. 36–53, 2014. [9] J. Wang, J. Feng, Z. Yan, Y. Hu, and G. Jia, “Nested high- Conflicts of Interest resolution modeling of the impact of urbanization on re- gional climate in three vast urban agglomerations in China,” ,e authors declare that there are no conflicts of interest Journal of Geophysical Research: Atmospheres, vol. 117, regarding the publication of this paper. no. D21, 2012. [10] J. Zhang, X. Song, G. Wang et al., “Development and chal- Acknowledgments lenges of urban hydrology in a changing environment: I. Hydrological response to urbanization,” Advances in Water ,is work was supported by the Fundamental Research Science, vol. 25, no. 4, pp. 594–605, 2014, in Chinese with Funds for the Central Universities (2015XKMS034), the English abstract. Advances in Meteorology 21 [11] E. Kalnay and M. Cai, “Impact of urbanization and land-use homogeneized observations,” Advances in Atmospheric Sci- change on climate,” Nature, vol. 423, no. 6939, pp. 528–531, ences, vol. 32, no. 9, pp. 1173–1185, 2015. 2003. [28] S. Wu, F. Chen, M. A. LeMone, M. Tewari, Q. Li, and Y. Wang, “An observational and modeling study of characteristics of [12] D. Rosenfeld, “Suppression of rain and snow by urban and industrial air pollution,” Science, vol. 287, no. 5459, urban heat island and boundary layer structures in Beijing,” pp. 1793–1796, 2000. Journal of Applied Meteorology and Climatology, vol. 48, no. 3, [13] D. Rosenfeld, U. Lohmann, G. B. Raga et al., “Flood or pp. 484–501, 2009. drought: how do aerosols affect precipitation?,” Science, [29] S. Miao, F. Chen, and Q. Li, “Month-averaged impacts of vol. 321, no. 5894, pp. 1309–1313, 2008. urbanization on atmospheric boundary layer structure and [14] Y. Huang, S. Chen, Q. Cao et al., “Evaluation of version-7 precipitation in summer in Beijing area,” Chinese Journal of TRMM multi-satellite precipitation analysis product during Geophysics, vol. 53, no. 7, pp. 1580–1593, 2010, in Chinese the Beijing extreme heavy rainfall events of 21 July 2012,” with English abstract. Water, vol. 6, pp. 32–44, 2014. [30] Z. Xu, L. Zhang, and B. Ruan, “Analysis on the spatio- [15] M. Li, J. Xia, Z. Chen et al., “Variation analysis of precipitation temporal distribution of precipitation in the Beijing re- during past 286 years in Beijing area, China, using non- gion,” Arid Land Geography, vol. 29, no. 2, pp. 186–192, 2006, parametric test and wavelet analysis,” Hydrological Pro- in Chinese with English abstract. [31] S. Yu, “Decadal rainfall variability over Beijing and the urban cesses, vol. 27, pp. 2934–2943, 2013. [16] J.-Y. Han, J.-J. Baik, and H. Lee, “Urban impacts on pre- effect,” Advances in Nature Science, vol. 17, no. 5, pp. 632–638, cipitation,” Asia-Pacific Journal of Atmospheric Sciences, 2007, in Chinese with English abstract. vol. 50, no. 1, pp. 17–30, 2014. [32] Y. Zhang, J. A. Smith, L. Luo, Z. Wang, and M. L. Baeck, [17] Z. Han, Z. Yan, Z. Li, W. Liu, and Y. Wang, “Impact of “Urbanization and rainfall variability in the Beijing metro- urbanization on low-temperature precipitation in Beijing politan region,” Journal of Hydrometeorology, vol. 15, no. 6, during 1960-2008,” Advances in Atmospheric Sciences, vol. 31, pp. 2219–2235, 2014. no. 1, pp. 48–56, 2014. [33] C. Zhang, C. Ji, Y. Kuo et al., “Numerical simulation of to- [18] J. Wang, R. Zhang, and Y. Wang, “Areal differences in diurnal pography effects on the “00.7” severe rainfall in Beijing,” variations in summer precipitation over Beijing metropolitan Progress in Natural Science, vol. 15, no. 9, pp. 818–826, 2005. region,” Beoretical and Applied Climatology, vol. 110, no. 3, [34] J. Sun, H. Wang, L. Wang et al., “,e role of urban boundary pp. 395–408, 2012. layer in local convective torrential rain happening in Beijing [19] L. Yang, F. Tian, J. A. Smith, and H. Hu, “Urban signatures in on 10 July 2004,” Chinese Journal of Atmospheric Sciences, vol. 30, pp. 221–234, 2006, in Chinese with English abstract. the spatial clustering of summer heavy rainfall events over the Beijing metropolitan region,” Journal of Geophysical Research: [35] Q. Wu, H. Guo, B. Yang et al., “Effects of topography and Atmospheres, vol. 119, no. 3, pp. 1203–1217, 2014. urban heat circulation on a meso-β torrential rain in Beijing [20] N. Molders ¨ and M. A. Olson, “Impact of urban effects on area,” Meteorology Monthly, vol. 35, pp. 58–64, 2009, in precipitation in high latitudes,” Journal of Hydrometeorology, Chinese with English abstract. vol. 5, no. 3, pp. 409–429, 2004. [36] J. Sun and B. Yang, “Meso-β scale torrential rain affected by [21] J. M. Shepherd, H. Pierce, and A. J. Negri, “Rainfall modi- topography and the urban circulation,” Chinese Journal of fication by major urban areas: observations from spaceborne Atmospheric Sciences, vol. 32, pp. 1352–1364, 2008, in Chinese rain radar on the TRMM satellite,” Journal of Applied Me- with English abstract. teorology, vol. 41, no. 7, pp. 689–701, 2002. [37] A. Bardossy ´ and G. Pegram, “Infilling missing precipitation [22] C. L. Zhang, F. Chen, S. G. Miao, Q. C. Li, X. A. Xia, and records - a comparison of a new copula-based method with other techniques,” Journal of Hydrology, vol. 519, pp. 1162– C. Y. Xuan, “Impacts of urban expansion and future green planting on summer precipitation in the Beijing metropolitan 1170, 2014. area,” Journal of Geophysical Research, vol. 114, no. D2, [38] R. S. V. Teegavarapu and V. Chandramouli, “Improved p. D02116, 2009. weighting methods, deterministic and stochastic data-driven [23] S. Dawadi and S. Ahmad, “Evaluating the impact of demand- models for estimation of missing precipitation records,” side management on water resources under changing climatic Journal of Hydrology, vol. 312, no. 1–4, pp. 191–206, 2005. conditions and increasing population,” Journal of Environ- [39] X. M. Song, J. Y. Zhang, J. F. Liu et al., “Spatial-temporal mental Management, vol. 114, pp. 261–275, 2013. variation characteristics of precipitation pattern in Beijing,” [24] A. K. Venkatesan, S. Ahmad, W. Johnson, and J. R. Batista, Journal of Hydraulic Engineering, vol. 46, no. 5, pp. 525–535, “Systems dynamic model to forecast salinity load to the 2015, in Chinese with English abstract. Colorado River due to urbanization within the Las Vegas [40] J. E. Oliver, “Monthly precipitation distribution: a compar- Valley,” Science of the Total Environment, vol. 409, no. 13, ative index,” Professional Geographer, vol. 32, no. 3, pp. 2616–2625, 2011. pp. 300–309, 1980. [41] L. J. Zhang and Y. F. Qian, “Annual distribution features of [25] X. Batista, Z. Wang, Y. Qi, and H. Guo, “Effect of urbanization on the winter precipitation distribution in Beijing area,” precipitation in China and their interannual variations,” Acta Science in China Series D: Earth Sciences, vol. 52, no. 2, Meteorologica Sinica, vol. 17, no. 2, pp. 146–163, 2003. pp. 250–256, 2009. [42] X. M. Li, F. Q. Jiang, L. H. Li et al., “Spatial and temporal [26] S. Guo and J. Ma, “Impact of urbanization on precipitation in variability of precipitation concentration index, concentration Beijing area,” Journal of the Meteorological Sciences, vol. 31, degree and concentration period in Xinjiang, China,” In- no. 4, pp. 414–421, 2011, in Chinese with English abstract. ternational Journal of Climatology, vol. 31, pp. 1679–1693, [27] Z. Li, Z. Yan, K. Tu, and H. Wu, “Changes of precipitation and 2011. extremes and the possible effect of urbanization in the Beijing [43] S. Fatichi, S. M. Barbosa, E. Caporali, and M. E. Silva, “De- Metropolitan region during 1960-2012 based on terministic versus stochastic trends: detection and 22 Advances in Meteorology challenges,” Journal of Geophysical Research, vol. 114, no. D18, article D18121, 2009. [44] M. G. Silva, Rank Correlation Methods, Griffin, London, UK, [45] H. B. Mann, “Nonparametric tests against trend,” Econo- metrica, vol. 13, no. 3, pp. 245–259, 1945. [46] P. K. Sen, “Estimates of the regression coefficient based on Kendall’s Tau,” Journal of the American Statistical Association, vol. 63, no. 324, pp. 1379–1389, 1968. [47] Z. Bao, J. Zhang, G. Wang et al., “Attribution for decreasing streamflow of the Haihe River basin, northern China: climate variability or human activities?,” Journal of Hydrology, vol. 460-461, pp. 117–129, 2012. [48] W. Shi, X. Yu, W. Liao, Y. Wang, and B. Jia, “Spatial and temporal variability of daily precipitation concentration in the Lancang River basin, China,” Journal of Hydrology, vol. 495, pp. 197–207, 2013. [49] 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. [50] Y. Dinpashoh, D. Jhajharia, A. Fakheri-Fard et al., “Trends in reference crop evapotranspiration over Iran,” Journal of Hydrology, vol. 399, no. 3-4, pp. 422–433, 2011. [51] T. B. M. J. Ouarda, C. Charron, K. Niranjan Kumar et al., “Evolution of the rainfall regime in the United Arab Emir- ates,” Journal of Hydrology, vol. 514, pp. 258–270, 2014. [52] A. R. Tome and P. M. A. Miranda, “Piecewise linear fitting and trend changing points of climate parameters,” Geophysical Research Letters, vol. 31, no. 2, p. L02207, 2004. [53] A. N. Pettitt, “A non-parametric approach to the change- point problem,” Applied Statistics, vol. 28, no. 2, pp. 126–135, [54] G. Villarini, F. Serinaldi, J. A. Smith, and W. F. Krajewski, “On the stationarity of annual flood peaks in the continental United States during the 20th century,” Water Resources Research, vol. 45, no. 8, article W08417, 2009. [55] C. Rouge, ´ Y. Ge, and X. Cai, “Detecting gradual and abrupt changes in hydrological records,” Advances in Water Re- sources, vol. 53, pp. 33–44, 2013. [56] M.-N. Yang, Y.-P. Xu, G.-B. Pan, and L.-F. Han, “Impacts of urbanization on precipitation in Taihu Lake basin, China,” Journal of Hydrologic Engineering, vol. 19, no. 4, pp. 739–746, 2014. International Journal of The Scientific Advances in Advances in Geophysics Chemistry Scientica World Journal Public Health Hindawi Hindawi Hindawi Hindawi Publishing Corporation Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 Journal of Environmental and Public Health Advances in Meteorology Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Applied & Environmental Journal of Soil Science Geological Research Hindawi Volume 2018 Hindawi www.hindawi.com www.hindawi.com Volume 2018 International Journal of International Journal of Agronomy Ecology International Journal of Advances in International Journal of Forestry Research Microbiology Agriculture Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 International Journal of Journal of Journal of International Journal of Biodiversity Archaea Analytical Chemistry Chemistry Marine Biology Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Advances in Meteorology Hindawi Publishing Corporation

A Comprehensive Analysis of the Changes in Precipitation Patterns over Beijing during 1960–2012

Loading next page...
 
/lp/hindawi-publishing-corporation/a-comprehensive-analysis-of-the-changes-in-precipitation-patterns-over-IHWEiSvawP

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2019 Xiaomeng Song et al. 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.
ISSN
1687-9309
eISSN
1687-9317
DOI
10.1155/2019/6364040
Publisher site
See Article on Publisher Site

Abstract

Hindawi Advances in Meteorology Volume 2019, Article ID 6364040, 22 pages https://doi.org/10.1155/2019/6364040 Research Article A Comprehensive Analysis of the Changes in Precipitation Patterns over Beijing during 1960–2012 1,2,3 2 1 1 Xiaomeng Song , Jianyun Zhang, Chunhua Zhang, and Xianju Zou School of Resources and Geosciences, China University of Mining & Technology, Xuzhou 221116, China State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Nanjing Hydraulic Research Institute, Nanjing 210029, China State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan 430072, China Correspondence should be addressed to Xiaomeng Song; xmsong.cn@gmail.com Received 15 October 2018; Accepted 28 January 2019; Published 6 March 2019 Academic Editor: Efthymios I. Nikolopoulos Copyright © 2019 Xiaomeng Song et al. ,is 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. Precipitation pattern has changed over many regions in recent decades, which may cause the risk of flood or drought. In this study, the main objective is to evaluate the spatiotemporal variability of precipitation in Beijing from 1960 to 2012. First, the mean monthly, seasonal, and annual precipitation series were used to analyze the temporal variation using regression, Mann–Kendall (M-K) test, Sen’s slope, and Pettitt tests. ,e results showed that the annual mean precipitation had a clear decreasing trend, with the statistically significant decrease in summer (especially in July and August) and significant increase in spring (especially in May). Although the decreasing trend is shown in the precipitation concentration indicators, the temporal uneven distribution of precipitation has unchanged. Subsequently, the precipitation time series at 30 stations over Beijing were used to evaluate the changes in precipitation pattern. ,e results showed that the annual series for the most rain gauges had decreasing trends with gradual changes. ,e spatial distribution of precipitation and other indices is geographically consistent, reflecting the principal physiographic and climatic conditions. At the same time, the effects of the terrain and urban development on the precipitation spatial distribution were detected. Generally, the large and heavy precipitations frequently occur in the plain areas, while the precipitation in the mountain areas is dominated by the small and medium precipitation. As a whole, the total precipitation in the plain areas (558.8 mm) was slightly higher than that in the mountainous areas (533.0 mm), while the precipitation in the urban areas (575.9 mm) was much higher than in the surrounding suburb areas (538.9 mm) during 1960–2012. ,e differences between the plain and mountainous areas during the period of 1960–1979, 1980–1999, and 2000–2012 were 24.2 mm, 32.6 mm, and 17.7 mm, respectively. ,e differences in precipitation between the urban and suburb areas for the three periods were 32.9 mm, 45.2 mm, and 31.0 mm, respectively, with the amount accounting for 5.51%, 7.66%, and 5.94% of the mean precipitation in the urban areas for the corresponding periods. transport in rivers and coastal waters. ,ere is a growing 1. Introduction concern in the scientific community over whether there are ,e long-term observation of climatic variables as a practical significant changes in precipitation amount, intensity, du- approach for monitoring climate change is receiving con- ration, and frequency because changes in precipitation siderable attention from researchers throughout the world. patterns may lead to floods or droughts in different areas Of the common climatic variables, precipitation is the most [2–4]. Also, trend and variability analysis of rainfall requires changeable in time and space, which directly affects natural urgent and systematic attention because of significant cycles of water resources. In addition, severe precipitation possible influences on freshwater availability, occurrence of can also cause storm-induced floods [1] and heavy sediment extreme events, food security, and economic activities [5–8]. loads from watersheds which may change sediment ,erefore, the spatial and temporal variability of 2 Advances in Meteorology precipitation time series is important from both the scientific increasing trend from 1724 to 2009; in addition, they found and practical point of view. some periodic characteristics that affected the trends re- garding the amount of annual precipitation. Zhai et al. [4] Linking with industrialization, during the late 20th to the early 21st century, urbanization occurred rapidly across discussed the spatiotemporal variation of precipitation from developing countries, especially in China; this urbanization 1724 to 2010 based on the same meteorological station. ,e is projected to continue over the coming decades. Urbani- annual precipitation presented a slowly increasing trend zation has modified the earth system [9] and local water during 1724–2010, which is consistent with that of the work cycle [10], creating environmental burdens, including urban by Li et al. [15]. Further, Zhai et al. [4] discussed the spatial heat islands (UHI) [11], urban aerosols [12], storm, and variations of precipitation from 1980 to 2010 based on the 20 flooding in urban areas [13, 14]. As a result, it is important to weather stations and showed that annual precipitation has analyze the rainfall variability in urban areas. Several studies decreased significantly during the first decade of the 21st have addressed issues related to urban heat island, impact of century compared to the last two decades of the 20th urbanization on rainfall, and spatiotemporal variations of century. For this study, a comprehensive analysis was conducted rainfall, especially for the metropolitan areas [3, 4, 15–19]. Urban expansion is associated commonly with increased of the changes in precipitation at the regional-scale using 30 amount of precipitation and the frequency of extreme rain gauges in the Beijing metropolitan area. ,is work precipitation in urban areas as well as downwind areas attempted to (1) identify gradual trends and abrupt shifts at because of the effect of urban heat island and high con- various temporal scales (monthly, seasonal, and annual) in centrations of anthropogenic aerosols [20–22]. Precipitation precipitation data series, (2) investigate the temporal and changes and variability in the metropolitan areas could spatial variations of precipitation with the consideration of present a challenge in terms of stormwater management, the urbanization and topography effect, and (3) discuss the water quality, and water supply [23, 24]. changes of precipitation patterns based on the classified precipitation events. ,e findings may be helpful to scien- Beijing, located in northern China and surrounded by mountains with complex terrain, is the capital of China and tifically understand the structural changes in precipitation and to evaluate the water availability and urban water re- one of the most populous cities in the world. It has expe- rienced rapid urbanization in the past few decades due to sources management in the study area. ,is study can also China’s reforms [25]. ,e built-up area of Beijing has grown provide a reference for further studies on precipitation 2 2 from 346 km in 1980 to 1350 km in 2013, and the resident mechanisms in urban areas. population of the city has exceeded 21 million. ,e complex terrain is likely to cause valley wind circulation, which may 2. Materials and Methods have an impact on the local climate, and the changes in the underlying characteristics as a result of rapid urbanization 2.1. Data Sources and Data Preprocessing. ,e daily rain will also influence local precipitation. ,erefore, the study of gauge data from 30 stations in Beijing areas were analyzed in temporal and spatial variations in the local climatic char- this study. ,ese data were provided by the Beijing Hy- acteristics of Beijing is important to our understanding of drological Stations (BHS) of the Beijing Water Authority the impacts of urbanization on the local climate. To date, (BWA). Note that most of the 30 observation stations within numerous studies have analyzed rainfall variability at dif- Beijing were established in the late 1950s and early 1960s and ferent temporal scales from different perspectives. Other formed a fixed network by the end of 1960s. As a result, the studies have been done on the possible effects of urban precipitation data from 1960 to 2012 are used to analyze the expansion on precipitation in the Beijing areas spatial distribution and the urbanization effects. ,e loca- [3, 17, 19, 25–32]. For example, Miao et al. [29] discussed the tions are shown in Figure 1, and details about these stations effects of land use or land cover on the characteristics of are listed in Table 1. For some stations, the missing pre- boundary layer structures of urban areas. ,eir results cipitation due to disruption in data collection requires data showed that the urban-rural circulations induced by topo- infilling. Various infilling methods have been reported in the graphic differences were one of the important causes for the literature, such as nearest neighbor, inverse distance, linear prevalence of mountain-valley flows in the Beijing area. regression, ordinary kriging, multiple linear regression, Zhang et al. [33] found that urbanization predominantly neural networks, and copula-based method [37, 38]. contributed to a reduction in precipitation in Beijing, based Compared to infilling precipitation data on short time scales on a mesoscale model, particularly over the Miyun reservoir (up to several days), the annual time scale is relatively simple. area. Because Beijing has a varied and complex topography As a result, in this work, the nearest neighbor method is used with high mountains in the west and north parts of Beijing, to infill the missing data via the neighbor control station and Sun et al. [34] and Wu et al. [35] pointed out that the terrain its observations. ,e missing data can be estimated by had an important effect on the location and distribution of multiplying with the ratio of the long-term means of the precipitation. Similarly, Sun and Yang [36] found that a target station and its nearest neighbor. In this study, the reduction in rainfall was caused by the joint effects of to- “nearest” neighbor station was selected based on the cor- pography and the urban heat island. In addition, significant relation coefficient and its sufficiently available precipitation spatiotemporal and interseasonal variations in the trends series. For example, the four stations surrounding Gao- and variability of precipitation in the Beijing area were beidian are Tongzhou, Songlinzha, Youanmen, and Maju- noticed. For example, Li et al. [15] revealed that there was an qiao. Figure 2 shows the annual precipitation correlation Advances in Meteorology 3 115°E 116°E 117°E China 40°N 30°N 41°N 20°N THK QJD ZJF XH 10°N YQ HHC MY ZIY HR 90°E 100°E 110°E 120°E TYK GT WJY National boundary PG SH SY SZ YC Basin boundary ZT SJD 40°N SLZ TZ Beijing city YAM GBD YLZ MJQ XYL HC FS ZF NGZ km 010 20 40 60 80 DEM Value Rain gauges –29 to 188.6 Beijing plain 188.7 to 451.5 District 451.6 to 732.6 732.7 to 1,104.3 1,104.4 to 2,283 Figure 1: Location and the distribution of the 30 rain gauges in Beijing that were used in this study. Table 1: Location of the 30 rain gauges and time series information. ° ° Station Abbreviation Longitude ( E) Latitude ( N) Time series Missing data Fangshan FS 115.97 39.70 1950–2012 1968, 1992–1996 Gaobeidian GBD 116.52 39.91 1972–2012 None Guanting GT 115.61 40.23 1950–2012 None Huangcun HC 116.33 39.73 1954–2012 1960–1961 Huanghuacheng HHC 116.34 40.40 1955–2012 None Huairou HR 116.61 40.31 1952–2012 None Majuqiao MJQ 116.55 39.76 1960–2012 1961 Miyun MY 116.84 40.39 1950–2012 1962 Nangezhuang NGZ 116.39 39.50 1965–2012 1976–1978 Pinggu PG 117.10 40.16 1950–2012 1992 Qianjiadian QJD 116.34 40.69 1961–2012 None Shahe SH 116.33 40.13 1959–2012 1979 Sanjiadian SJD 116.10 39.96 1950–2012 None Songlinzha SLZ 116.37 39.95 1963–2012 None Shunyi SY 116.66 40.11 1950–2012 1992–1996 Suzhuang SZ 116.75 40.06 1950–2012 None Tanghekou THK 116.63 40.73 1960–2012 1962 Taoyukou TYK 116.45 40.23 1962–2012 None Tongzhou TZ 116.65 39.93 1950–2012 None Wangjiayuan WJY 115.98 40.20 1962–2012 None Xiahui XH 117.16 40.61 1961–2012 1970–1973 Xiayunling XYL 115.74 39.73 1951–2012 1961, 1992–1996 Youanmen YAM 116.35 39.87 1963–2012 1968 Yanchi YC 115.89 40.03 1963–2012 None Yulinzhuang YLZ 116.79 39.79 1961–2012 None Yanqing YQ 115.99 40.46 1959–2012 1992–1993 Zhangfang ZF 115.69 39.58 1963–2012 None Zhangjiafen ZJF 116.78 40.61 1960–2012 None Zhenluoying ZLY 117.14 40.34 1954–2012 None Zhaitang ZT 115.68 39.96 1951–2012 None 4 Advances in Meteorology 1000 1000 y = 0.9404x + 56.058 y = 1.0539x – 17.093 R = 0.7604 R = 0.8715 800 800 600 600 400 400 200 200 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Gaobeidian (mm) Gaobeidian (mm) (a) (b) y = 0.588x + 26.158 y = 0.7744x + 47.29 R = 0.7574 R = 0.7381 200 200 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 Gaobeidian (mm) Gaobeidian (mm) (c) (d) Figure 2: Scatter plots for the correlation relationship between Gaobeidian station and surrounding rain gauges. ,e black solid lines are 1 : 1 lines. ,e red dashed lines mean the linear regression lines. relationships between Gaobeidian and other four stations. (0.1 mm≤ daily precipitation< 10 mm); medium (10 mm≤ daily precipitation< 25 mm); large (25 mm≤ daily With highest correlation, Tongzhou station is the “nearest” neighbor station to infill the missing annual precipitation at precipitation< 50 mm); heavy (50 mm≤ daily precipitation). Gaobeidian station. In addition, there is no missing pre- We also define two indices to assess these changes in pre- cipitation data at Tongzhou station. Once the “nearest” cipitation patterns according to the work from Song et al. neighbor station has been identified for each station with [39]. One is precipitation incidence (PI), which means the missing data, using the linear regression equation provided proportion of precipitation days for every precipitation in Table 2, missing data can be infilled. From Table 2, we find grade to the total days of precipitation in a year; the other is that the precipitation series for the nearest neighbor stations precipitation contribution (PC), which means the pro- are quite strongly correlated with those stations that have portion of precipitation amount for every precipitation grade to the total amount of precipitation in a year. missing data; for most stations, correlation coefficients ex- ceed 0.7. Additionally, the rain gauge stations provide reasonably good coverage of different altitudinal zones, i.e., plain (in- cluding urban and suburb areas) and mountainous areas in 2.2. Classification in Precipitation Grades and Subregions. the study area. By considering the complex surface char- In this study, the standard for records is that precipitation acteristics, 30 observation sites were divided into two cat- amount is greater than or equal to 0.1 mm when rainfall egories, plain and mountain areas (Figure 1). ,ere are 18 occurs. To further understand the change of precipitation sites (SLZ, YAM, GBD, TZ, HC, MJQ, SH, SY, SZ, YLZ, ZF, patterns over Beijing, the observed daily precipitation was FS, SJD, TYK, HR, MY, PG, and NGZ) covering the plain categorized into four grades of intensity according to China areas and 12 sites covering the mountain areas. Among these Meteorological Administration (CMA) standards: small stations in the plain areas, four stations located in urban area Youanmen (mm) Tongzhou (mm) Majuqiao (mm) Songlinzha (mm) Advances in Meteorology 5 Table 2: Information for infilling missing data. Station Nearest neighbor Calibration series Correlation coefficient Equation FS ZF 1963–1967, 1969–1991, 1997–2012 0.78 y � 0.849x + 61.5 GBD TZ 1972–2012 0.93 y � 0.827x + 86.5 HC FS 1962–1967, 1969–1991, 1997–2012 0.68 y � 0.537x + 225.8 MJQ TZ 1960, 1962–2012 0.83 y � 0.738x + 65.4 MY ZLY 1960, 1963–2012 0.64 y � 0.9x + 53.78 NGZ YLZ 1979–2012 0.72 y � 0.657x + 87.1 PG ZLY 1960–1991, 1993–2012 0.66 y � 0.519x + 172.2 QJD YQ 1961–1991, 1994–2012 0.69 y � 0.569x + 173.4 SH SLZ 1963–1978, 1980–2012 0.80 y � 0.673x + 160 SLZ TZ 1963–1967, 1969–2012 0.80 y � 0.795x + 121.5 SY SZ 1960–1991, 1997–2012 0.83 y � 0.831x + 80.8 THK HHC 1960–1961, 1963–2012 0.75 y � 0.504x + 164.4 TYK SH 1962–1978, 1980–2012 0.71 y � 0.713x + 203.1 WJY SH 1962–1978, 1980–2012 0.73 y � 0.7682x + 125.5 XH MY 1963–1969, 1974–2012 0.75 y � 0.570x + 267.2 XYL ZT 1960, 1962–1991, 1997–2012 0.77 y � 1.203x + 72.4 YAM TZ 1963–1967, 1969–2012 0.82 y � 0.814x + 89.7 YC ZT 1963–2012 0.82 y � 0.892x + 67.5 YLZ TZ 1961–2012 0.85 y � 0.861x + 24.2 YQ GT 1960–1991, 1994–2012 0.74 y � 0.974x + 83.9 ZF FS 1963–1967, 1969–1991, 1997–2012 0.78 y � 0.708x + 193.7 for a year can be seen as a circle (360 (SLZ, YAM, GBD, and TZ) were used to estimate the mean ). ,en, the yearly PCP precipitation in the urban areas and six surrounding stations and PCD for a location can be defined as follows: (HC, MJQ, SH, SY, SZ, YLZ) were selected to calculate the P � 􏽘 P , corresponding mean precipitation in the suburb areas. P � 􏽘 P · sin θ , x i i 2.3. Precipitation Concentration Index, Concentration Degree, P � 􏽘 P · cos θ , and Concentration Period. ,e precipitation concentration y i i index (PCI) developed by Oliver [40] was used to explore the (2) changing features of precipitation concentration in Beijing. PCP � arctan􏼠 􏼡, PCI was proposed as an indicator of monthly rainfall het- erogeneity, according to the following equation: 􏽱������ � 2 2 P + P 12 2 x y 􏽐 P i i PCD � , PCI � × 100, (1) 12 P 􏼐􏽐 P 􏼑 where P is the monthly precipitation of the ith month, θ is i i where P is the monthly precipitation in month i. According the azimuth of the ith month, and P is the total precipitation to equation (1), the lowest theoretical value of PCI is 8.3, amount. PCP can reflect which month the maximum which means the perfect uniformity in precipitation dis- monthly precipitation appears in. ,e corresponding re- tribution. Oliver [40] classified PCI values in four categories: lation between PCP and month in a year is as follows: ° ° ° ° (1) PCI≤ 10 means uniform precipitation distribution, January (0 ), February (30 ), March (60 ), April (90 ), May ° ° ° ° i.e., low precipitation concentration; (2) 10< PCI≤ 15 means (120 ), June (150 ), July (180 ), August (210 ), September ° ° ° ° moderate precipitation concentration; (3) 15< PCI≤ 20 (240 ), October (270 ), November (300 ), December (330 ). means irregular precipitation distribution, i.e., high pre- PCD can reflect the degree to which annual total pre- cipitation concentration; and (4) PCI> 20 means strong cipitation is distributed in 12 months, which is ranging from irregular precipitation distribution, i.e., very high pre- 0 to 1. cipitation concentration. In addition, the precipitation concentration degree (PCD) and the precipitation concentration period (PCP) 2.4. Trends Detecting Method. A trend analysis was per- proposed by Zhang and Qian [41] were also used to analyze formed to detect gradual changes or tendencies in the time the interannual variations of precipitation amounts. ,e series data, using the Mann–Kendall (M-K) test and Sen’s basic principle for calculating the PCD and PCP is based on slope method. Both are commonly used tests for trend the vector of monthly total precipitation [42]. ,e as- detection [43–51]. Trends were evaluated using the non- sumptions can be made that monthly total precipitation is a parameteric M-K test [44, 45]. Significance of trend was vector quantity with both magnitude and that the direction evaluated at the 0.05 and 0.10 levels. ,e magnitude of trends 6 Advances in Meteorology was evaluated using Sen’s slope [46]. ,e M-K test and Sen’s during 1960–2012 ranges from 383.9 mm (1965) to slope were chosen because they were widely used to detect 797.9 mm (1964) with an average of 553.6 mm. ,e annual the trends of hydrological and meteorological data time mean precipitation has a significantly decreasing trend at a series. ,e details of the M-K test and Sen’s slope are in- rate of 11.6 mm/10a from 1960 to 2012, as analyzed by the troduced in Appendix. linear regression method. For seasonal precipitation, the spring and autumn precipitations have a slightly increasing trend analyzed (4.7 mm/10a and 4.9 mm/10a); in contrast, 2.5. Pettitt Change Point Analysis. Nonstationarity in time summer precipitation has a visibly decreasing trend series can also be characterized by abrupt shifts in the mean (21.1 mm/10a), which is greater than the decline rate in the or variance of the series. To identify possible shifts, standard annual precipitation series. Unlike the above three seasons, change-point procedures have been applied to multiple time the trend of mean precipitation in winter has been almost series in Earth sciences including precipitation and tem- flat, fluctuating within a range from 0.9 to 27.7 mm. perature [52]. One type of homogeneity (change-point) test ,erefore, decreasing precipitation in summer has been the is the Pettitt test [53], which is a nonparametric most important factor in the overall decrease of annual test—requiring no assumption on the underlying distribu- precipitation in Beijing. Several periods of severe droughts tion [54]. and moisture surpluses are evident in the anomaly series of To perform the two-tailed hypothesis test on the location regional mean precipitation in Figure 3. For example, the parameter (mean), the Pettitt test statistic is calculated as long-term dry period during 1999–2011 in Beijing is ⎪ x < x , characterized by a high number of negative anomalies. Si- ⎧ −1, ⎪ i j multaneously, various temporal variations for interannual D � 0, x � x , (3) i j ij and interdecadal variability of seasonal precipitation were x > x , 1, exhibited during 1960–2012 from the precipitation anomaly i j series and the 11-year moving-average series. For instance, where x and x correspond to the magnitude of the i j in the period of 2000–2012, although the precipitation hydroclimatic variable under consideration and x precedes amount in certain years (2008 and 2012) is on the high side x in time. For evaluation over the entire sample (T years), (larger than the mean value during 1960–2012), the summer these D statistics are combined as follows: total precipitation is relatively lower than its long-term t T mean. In contrast, the precipitation amount in spring and U � 􏽘 􏽘 D . (4) autumn in the most years during 2000–2012 is on the high t,T ij i�1 j�t+1 side. ,e M-K test and Sen’s slope estimation were applied to ,e statistic U is equivalent to a Mann–Whitney statistic detect the trends of the monthly, seasonal, and annual for testing that the two samples X , . . ., X and X , . . ., X 1 t t+1 T precipitation series during 1960–2012, as shown in Table 3. come from the same population. ,e test statistic U is For the monthly scale, the series in May had a statistically evaluated for all possible values of t ranging from 1 to T. ,e significant increasing trend, while that in July and August most probable year of a change-point occurring is evaluated had a significant decreasing trend, both at a level of α � 0.05. using a two-tailed test on the following statistic: ,e series in January and February had a significantly de- 􏼌 􏼌 􏼌 􏼌 􏼌 􏼌 K � max􏼌U 􏼌. (5) creasing trend at a level of α � 0.10. All other series had a T t,T nonsignificant trend at both levels during this study period. If the statistic K is significantly different from 0, then a For the seasonal scale, the precipitation series in spring change-point occurs in the year t corresponding to the point showed a significant increasing trend, while that in summer in time for which the largest absolute value of U , is ob- t,T showed a statistically significant decreasing trend, both at a tained. ,e probability of a shift in a year where |U | is the t,T level of α � 0.05. Both the precipitation series in autumn and maximum is estimated by [55] winter had a nonsignificant trend. For the annual scale, there is a nonsignificant trend for the annual precipitation series −6K P � 2 exp . (6) 􏼠 􏼡 3 2 during 1960–2012 at any level of α � 0.05 or α � 0.10. T + T ,e monthly variation of precipitation amounts is shown Given a certain significance level α, if P< α, we reject the in Figure 4. ,e highest precipitation occurred in the wet season (May to October), especially in the period from June null hypothesis and conclude that X is a significant change point at level α. ,e two significant levels (α � 0.05 and to September (namely flood season in Beijing), while the cold season (November to April) received less precipitation. α � 0.10) were used in this work. In this study, the monthly, seasonal, and annual series of regional mean precipitation ,e precipitation in the period from May to October accounted for 91.8% of the annual mean precipitation. and the annual precipitation series for all the stations were used to detect the change points based on the Pettitt method. Additionally, the precipitation in flood season contributed more than 81.5% to the annual mean precipitation. Further, it is clear that a large percentage of total precipitation occurs 3. Results and Discussion in summer, ranging from 39.5% to 87.4%, especially in July (8.3%–58.2%) and August (6.6%–50.6%). ,e maximum and 3.1. Changes of Regional Mean Precipitation. As seen in Figure 3, the annual mean precipitation in Beijing areas minimum values for average monthly precipitation were Advances in Meteorology 7 120 y = 0.4717x – 873.73 500 y = –2.1143x + 4599.11 0 100 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Summer Spring Linear trend Linear trend (a) (b) 200 30 y = 0.4894x – 891.26 y = 0.0083x – 7.41 0 0 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Winter Autumn Linear trend Linear trend (c) (d) y = –1.156x + 2848.71 –50 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Annual Spring Linear trend 11-Years moving average (e) (f) 300 120 –100 –40 –200 –300 –80 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Summer Autumn 11-Years moving average 11-Years moving average (g) (h) 20 300 –100 –5 –200 –10 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Winter Annual 11-Years moving average 11-Years moving average (i) (j) Figure 3: Time series of precipitation and anomaly in the Beijing metropolitan area from 1960 to 2012. 183.9 mm in July and 2.2 mm in December. To further Beijing has strongly uneven monthly precipitation distri- discuss the variation of monthly precipitation, the time bution, where most precipitation occurred in summer, series of PCI, PCD and PCP are shown in Figure 5. As seen i.e., June, July, and August (as shown in Figure 4). Addi- in Figure 5, the PCI in Beijing during 1960–2012 ranges from tionally, the PCI and PCD both have a slight decreasing 13.1 (2003) to 37.4 (1962) with an average of 23.7. ,e PCD trend at a rate of 1.68/10a and 0.02/10a from 1960 to 2012, as ranges from 0.47 (2003) to 0.86 (1994) with the mean of 0.72 analyzed by the linear regression method. To some extent, over the period of 1960–2012. Overall, the precipitation over the decreasing PCI (PCD) may be caused by the significant Precipitation anomaly (mm) Precipitation anomaly (mm) Precipitation (mm) Precipitation (mm) Precipitation (mm) Precipitation anomaly (mm) Precipitation anomaly (mm) Precipitation anomaly (mm) Precipitation (mm) Precipitation (mm) 8 Advances in Meteorology Table 3: Trend test results of the monthly, seasonal, and annual after the change point were examined, as shown in Table 4. precipitation during 1960–2012 tested by the M-K method and Significant abrupt changes in monthly precipitation were Sen’s slope method. ,e uncertainty range covered from 5% to 95% found in May (47% upward in 1976) and August (46% between the minimum and maximum values of Sen’s slope. downward in 1996) at a level of α � 0.05. A significant abrupt change in monthly precipitation was also found in July (31% Z value Sen’s slope (mm/a) Minimum Maximum ∗ downward in 1998) at a level of α � 0.10. Of them, as shown Jan −1.695 −0.009 −0.018 0 in Table 4, the significant downward change points occurred Feb −1.918 −0.043 −0.113 −0.007 in the end of 1990s with the mean precipitation decreasing Mar −1.036 −0.055 −0.111 0.067 Apr 1.365 0.148 −0.060 0.265 by 46% (1996) and 31% (1998), while the significant upward ∗∗ May 2.179 0.379 0.146 0.476 change point in the mid 1970s (1976) with the mean pre- Jun 1.289 0.465 −0.375 0.724 cipitation increasing by 47%. Other monthly precipitation ∗∗ Jul −1.971 −1.227 −2.200 −0.122 series have no significant abrupt change during 1960–2012. ∗∗ Aug −2.148 −1.401 −3.250 −0.892 For the seasonal scale, only summer precipitation has a Sep 0.905 0.248 −0.350 0.550 significant abrupt change in 1996, and the downward Oct 0.721 0.099 −0.225 0.207 magnitude is approximate 25% compared with the mean Nov 0.744 0.037 −0.077 0.075 value of the entire series. Similar to the other three seasons, Dec 1.204 0.011 −0.005 0.033 ∗∗ the annual precipitation has no significant abrupt change at Spring 2.086 0.535 0.025 1.025 ∗∗ either significant level. Summer −2.202 −2.150 −4.136 −0.344 Autumn 1.243 0.569 −0.304 1.250 Winter 0.184 0.007 −0.134 0.102 Annual −0.767 −0.861 −3.159 1.344 3.2. Changes of Spatial Distribution of Precipitation Amount. ∗,∗∗ Note. Statistically significant trends at the 10% and 5% significance ,e spatial distribution of the annual precipitation and their levels, respectively. trend test results is shown in Figure 6. As shown in Figure 6(a), it can be seen from this figure that the average annual precipitation varies greatly in Beijing and it varies from almost 400 mm in the west to more than 600 mm in the east. From the spatial perspective, the distribution shows that the precipitation is decreasing from east to west due to orographic influence. Apparently, the annual precipitation in the Beijing Plain is greater than in the mountainous areas. But the highest value of precipitation frequently occurred in the front belts (e.g., HR and MY stations) between the plain and mountainous areas. As a whole, there was one local maximum of precipitation with an annual mean pre- cipitation of 650 mm, located in the vicinity of the Huairou 100 and Miyun reservoirs in the northeastern section of Beijing. Another local maximum was centered at the XYL stations in the Fangshan district in the southwestern part of Beijing. With regard to the interannual variability of precipitation, the coefficient of variation and standard deviation of annual Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec precipitation have been identified for the period of analysis, Figure 4: Box-plot for monthly precipitation amount during as shown in Figures 6(b) and 6(c). ,e range of values over 1960–2012. ,e boxes indicate the 25th, 50th, and 75th percentiles Beijing is from 0.199 to 0.380. ,e lowest values appear in the and the whiskers indicate the values with maximum 1.5 northwestern area, and the highest values appear in the front interquartile ranges. ,e hollow square indicates the mean value. belts between the plain and mountainous areas, which is ,e solid circle means the outlier. similar to the spatial pattern of annual mean precipitation. ,e spatial distribution of the standard deviation in pre- decreases in precipitation in summer and the increases in cipitation also shows the similar pattern to the annual precipitation. ,is explains the standard deviation (co- spring and autumn for the Beijing area (as shown in Fig- ure 3). According to the changes of PCP in Figure 5, the efficient of variation) increases with the annual mean pre- cipitation, namely, the higher the annual precipitation is, the range of the mean yearly PCPs in Beijing is 189± 7, implying that annual precipitation mainly falls in summer (June-July- higher the degree of uncertainty (standard deviation and August). ,e time of the PCP mainly appears from the end of coefficient of variation) is. ,is might increase the degree of June (23, June) to the end of July (26, July). ,e maximum uncertainty in the availability of regional water resources, monthly precipitation in most years (46 years) occurs in July due to the close relationship between the precipitation and the other occurs in June (7 years). amount and the regional water resources. Change points in the monthly, seasonal, and annual ,e spatial distribution of Z values of the M-K trend test, Sen’s slope, and linear trend for the individual rain gauges is series over Beijing were detected based on the Pettitt method. ,en, the changes in the two samples before and also shown in Figure 6. ,e annual trends found by the linear Precipitation (mm) Advances in Meteorology 9 y = –0.168x + 357 10 0.9 0.8 0.7 0.6 0.5 y = –0.002x + 4.87 210 0.4 July June 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Years Figure 5: Time series of PCI, PCD, and PCP over the period of 1960–2012. ,e red solid line indicates the linear regression trend. ,e red dashed line means the boundary between the months June and July for the PCP. If the PCP is lower than 180, the maximum monthly precipitation appears in June. If the PCP ranges from 180 to 210, the maximum value appears in July. Table 4: Results of change point analysis for different series based on the Pettitt test. Entire series (mm) Change point p Prechange mean (mm) Postchange mean (mm) Ratio of changes Jan 2.2 2001 0.487 2.3 1.8 −0.23 Feb 4.6 1986 0.166 5.3 3.9 −0.30 Mar 9.0 1991 0.612 9.4 8.4 −0.11 Apr 20.5 1997 0.327 19 24.4 0.26 ∗∗ May 33.6 1976 0.042 22.9 38.6 0.47 Jun 77.2 1975 0.295 62.2 83.7 0.28 Jul 180.7 1998 0.060 195.7 139.2 −0.31 ∗∗ Aug 142.2 1996 0.024 161.8 96.7 −0.46 Sep 50.6 1985 0.540 44.7 56.3 0.23 Oct 22.2 1995 0.424 20 26.9 0.31 Nov 7.8 1967 1.121 3.2 8.6 0.69 Dec 2.2 1973 0.124 0.8 2.7 0.86 Spring 63.1 1976 0.147 49.8 69.4 0.31 ∗∗ Summer 400.1 1996 0.043 430.4 330.1 −0.25 Autumn 80.7 1988 0.690 73 89.9 0.21 Winter 8.9 1996 1.380 8.5 9.8 0.15 Annual 553.0 1996 0.398 573.5 505.6 −0.12 Note. ,e ratio of change is equal to the differences of mean precipitation between prior series and posterior series dividing the mean precipitation at the ∗ ∗∗ entire series. Significant abrupt change at a level of α � 0.10; significant abrupt change at level of α � 0.05. regression were almost similar to the precipitation trends Beijing. From the distribution of the annual precipitation found by the M-K test and Sen’s slope estimator. Both trend, we can see that the highest value of precipitation decline mostly occurred in the southwest part of Beijing with positive and negative trends were identified by the statistical tests in annual precipitation data. Only one station (ZT) a rate of more than 2 mm/a. From the linear trend showed significantly decreasing trend at a level of α � 0.05. (Figure 6(f)), only three stations (NGZ, SZ, and ZJF) have All the other stations showed no significant decreasing or slight increasing trend (less than 1 mm/a) and the other increasing trend at the 95% confidence level. For these stations have decreasing trend with an average of 12.2 mm/ stations with insignificant trend, seven stations showed a 10a. ,e WJY, XYL, and TYK stations have a clear de- slight increasing trend and twenty-two stations showed a creasing trend with a value larger than 20 mm/10a. decreasing trend. ,e declining trend of most stations was Many studies have reported that changes in precipitation similar to the change of regional annual precipitation over pattern are associated with elevation changes, although the PCP PCD PCI 10 Advances in Meteorology 116°E 117°E 116°E 117°E 41°N N N 41°N THK S THK S QJD QJD ZJF XN ZJF XH YQ YQ HHC MY HHC MY ZY ZY HR GT TYK GT TYK WJY WJY PG PG SH SH SY SY SZ SZ YC YC ZT 40°N SJD SLZ ZT 40°N SJD SLZ TZ GBD TZ GBD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF NGZ NGZ 0 10 20 40 0 10 20 40 km km 429.6 to 475.0 559.0 to 582.5 0.199 to 0.237 0.291 to 0.305 475.1 to 506.5 582.6 to 607.8 0.238 to 0.258 0.306 to 0.334 506.6 to 535.3 607.9 to 652.4 0.259 to 0.276 0.335 to 0.380 535.4 to 558.9 0.277 to 0.290 (a) (b) 116°E 117°E 116°E 117°E 41°N 41°N W E THK THK QJD QJD ZJF ZJF XH XH YQ YQ HHC HHC MY MY ZIY ZY HR HR GT GT TYK TYK WJY WJY PG PG SH SH SY SY SZ SZ YC YC ZT ZT SJD 40°N SJD SLZ 40°N SLZ TZ TZ GBD GBD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF NGZ NGZ 0 10 20 40 0 10 20 40 km km –2.09 to –1.96 78.5 to 115.6 173.2 to 191.9 115.7 to 149.6 192.0 to 243.7 –1.95 to 0 149.7 to 173.1 0 to 0.67 (c) (d) Figure 6: Continued. EW EW EW Advances in Meteorology 11 116°E 117°E 116°E 117°E 41°N 41°N W E W E THK THK S QJD QJD ZJF XH ZJF XH YQ YQ HHC MY HHC MY ZIY ZIY HR HR GT TYK TYK WJY GT WJY PG PG SH SY SH SY SZ YC SZ YC ZT ZT 40°N SJD SLZ 40°N SJD SLZ TZ GBD TZ GBD YAM YAM YLZ MJQ YLZ XYL MJQ HC FS XYL HC FS ZF ZF NGZ NGZ 0 10 20 40 0 10 20 40 km km Sen’s slope (mm/a) mm/10a –2.3 to –2.0 –0.9 to 0 –27.4 to –20.0 –9.9 to 0 –1.9 to –1.0 0 to –1.0 –19.9 to –15.0 0 to 9.3 –14.9 to –10.0 (e) (f) Figure 6: ‚e spatial distribution of mean precipitation and their statistical characteristics and trends in Beijing during 1960–2012: (a) the annual mean precipitation, (b) the coeƒcient of variation, (c) the standard deviation, (d) the M-K test, (e) Sen’s slope, and (f) the linear trend. relationships can vary regionally. To investigate the re- series with a negative change for all of these 21 stations. As lationship between the precipitation characteristics and el- discussed in Section 3.1, Beijing su˜ered from the contin- evation, we calculated the correlations between the 53-year uous drought with continued 12 years since the end of 1990s, (1960–2012) average precipitation amount and all station which is consistent with the results of change point analysis. elevations, as shown in Figure 7. ‚e results show that standard deviation and coeƒcient of variation are negatively and signi…cantly correlated with elevation, with Pearson 3.3. Changes of Dierent Precipitation Grades in the Flood correlation coeƒcients higher than 0.5 (p < 0.01). A weak Season. According to the results of PCI (Figure 5), we know that the uneven distribution of precipitation in Beijing is but signi…cant negative relationship is also observed for precipitation amount with correlation of 0.45 (p < 0.05). evident. We can see that the most of precipitation amount in ‚ese strong correlations between the precipitation and Beijing happened in the šood season (June-July-August- elevation indicate that local elevation has important impacts September). In this study, the changes of precipitation on the magnitude of precipitation in Beijing. Box plots show grades were discussed based on the daily precipitation data that the largest mean values of precipitation and standard in the šood season. Summary statistics of precipitation deviation appear at 100∼200 m, while the highest value of frequency, amount, and intensity for di˜erent grades of coeƒcient of variation occur at 0∼100 m. Overall, therefore, precipitation are presented in Table 6. ‚e mean number of the precipitation decreases with altitude, and higher pre- precipitation frequency is largest for small precipitation, with a value of 27, accounting for 67.8% of the total rainy cipitation events occur at lower altitudes in this area with higher coeƒcient of variation and standard deviation. days. ‚e lowest number of days belongs to the heavy precipitation category, at about 1–2 days every year during Regarding the analysis of the change points in the mean, the results for all the 30 rain gauges are summarized in the šood season. ‚e characteristics of precipitation amount Table 5 using the Pettitt test. As shown, change from positive di˜er from that of the frequency for di˜erent grades of to negative direction was detected for the 25 stations with a precipitation. ‚e lowest value of precipitation amount ratio of change from 7% to 22%. Only two stations (PG and belongs to the small precipitation, although its frequency is ZT) presented signi…cant change point at a level of α  0.10, the highest. For the other three categories, their precipitation which occurred in the years 1997 and 1996, respectively. We amounts are almost equivalent, at about 120 mm per year. found most of change points (21 stations) detected based on However, the standard deviation of precipitation amount for Pettitt test occurred in the end of 1990s (1996–1998) without heavy precipitation is largest. ‚is explains the interannual variation for heavy precipitation is relatively larger than any statistical signi…cant levels. ‚e average change was about 16% of the annual mean precipitation in the entire other categories. To a certain extent, the high variation of 12 Advances in Meteorology 700 700 R = –0.447 p = 0.013 600 600 500 500 400 400 300 300 0 100 200 300 400 500 0~100 100~200 200~300 300~400 400~500 Elevation (m) Elevation (m) (a) (b) 0.40 0.40 0.36 0.36 R = –0.643 p < 0.01 0.32 0.32 0.28 0.28 0.24 0.24 0.20 0.20 0 100 200 300 400 500 0~100 100~200 200~300 300~400 400~500 Elevation (m) Elevation (m) (c) (d) 250 250 R = –0.578 200 200 p < 0.01 150 150 100 100 50 50 0 100 200 300 400 500 0~100 100~200 200~300 300~400 400~500 Elevation (m) Elevation (m) (e) (f) Figure 7: Scatter and box-plots in the correlations between precipitation characteristics and elevation. Solid red line is the linear trend. R signifies the Pearson correlation coefficient for each relationship and p indicates the statistical significance. ,e solid square means the mean value during the different elevation bands. heavy precipitation might increase the degree of uncertainty precipitation is larger than that of small precipitation. ,e of precipitation amount. A relatively obvious decline in largest value of PC is the medium precipitation, ranging precipitation amount for the heavy precipitation is shown from 18% to 40% with a mean of 29%. And the second based on the linear regression coefficient. ,is change might belongs to the large precipitation, ranging from 18% to 37% cause the change in precipitation amount in the flood season with a mean of 27%. Additionally, the broad bound of PC over Beijing. (5–45%) for the heavy precipitation explains the relatively Figure 8 also shows the ranges of precipitation frequency larger variation and higher degree of uncertainty. ,e de- and amount for different grades. ,e frequency ranges from tailed information about the PI and PC for 30 stations is shown in Figure S1. 16.9 to 39.4, 3.9 to 10.7, 1.7 to 6.1, and 0.2 to 3.1 for the small, medium, large, and heavy precipitations, respectively. ,e In order to further discuss the spatial patterns of pre- cipitation frequency and amount for the different grades, amount is ranging from 45 to 118.6 mm, 60 to 172.7 mm, 54.7 to 212.7 mm, and 16.7 to 314.2 mm for the different Figure 9 shows the spatial interpolation results of both grades, respectively. In this work, PI and PC are used to indices for different grades. Overall, there are visible dif- further analyze the changes in different grades of pre- ferences among the spatial patterns for the different grades cipitation, as also shown in Figure 8. Overall, the largest of precipitation. For example, the frequency is increasing value of PI belongs to the small precipitation, accounting for from east to west for the small precipitation, but it is in- 67% (54–77%) of all the rainy days. ,e mean values of PI for creasing from south to north for the medium precipitation. large and heavy precipitations are both smaller than 10% From the perspective of precipitation amount, there are (9% and 4%). However, the value of PC for heavy similar distribution patterns for the small and medium Standard deviation Coefficient of variation Annual precipitation (mm) Standard deviation Coefficient of variation Precipitation (mm) Advances in Meteorology 13 Table 5: Results of change-point detection for different stations based on the Pettitt test. Stations Entire series Change point p Prior series Posterior series Ratio of change FS 568.6 1996 0.229 599.0 496.4 −0.17 GBD 570.2 1998 0.494 591.5 509.3 −0.13 GT 374.4 1993 1.553 369.6 383.8 0.03 HR 641.7 1998 0.190 678.4 549.8 −0.19 HC 526.0 1998 0.773 546.3 467.9 −0.14 HHC 593.7 1998 0.465 617.9 524.6 −0.14 MJQ 496.8 1998 0.373 520.9 426.6 −0.17 MY 641.1 1998 0.672 675.1 543.9 −0.19 NGZ 437.1 2006 0.604 422.6 550.2 0.26 PG 511.3 1997 0.060 541.9 431.7 −0.20 QJD 429.7 1992 0.571 440.1 404.8 −0.07 SJD 577.1 1996 0.921 597.6 527.7 −0.11 SH 539.1 1972 0.841 498.6 553.2 0.09 SY 564.3 1998 0.502 598.4 466.9 −0.21 SLZ 586.0 1998 0.344 612.0 511.6 −0.16 SZ 580.8 1998 0.773 602.4 522.1 −0.13 THK 464.0 1998 0.295 484.6 404.9 −0.15 TYK 604.6 1998 0.130 642.1 497.3 −0.22 TZ 584.1 1998 0.487 611.0 507.1 −0.16 WJY 550.0 1979 0.338 622.1 504.1 −0.20 XYL 612.0 1996 0.183 643.8 537.9 −0.16 XH 628.9 1975 1.046 573.6 654.2 0.12 YQ 449.0 1973 0.672 492.4 433.2 −0.12 YC 467.4 1979 0.213 513.8 438.8 −0.14 YAM 565.3 1998 0.285 591.4 490.7 −0.16 YLZ 527.3 1998 0.587 547.6 469.2 −0.14 ZT 448.6 1996 0.094 471.8 393.8 −0.15 ZF 598.2 1996 1.282 614.0 564.6 −0.08 ZJF 618.5 1975 0.451 561.5 644.7 0.12 ZLY 652.4 1998 0.176 689.4 546.9 −0.20 Note. ,e ratio of change is equal to the differences of mean precipitation between prior series and posterior series dividing the mean precipitation at the entire series. Significant abrupt change at a level of α � 0.10. Table 6: Summary statistics for the precipitation frequency, amount, and intensity for the different grades. Frequency (days) Amount (mm) Intensity (mm/day) Mean SD RC (days/10a) Mean SD RC (mm/10a) Mean SD RC (mm/(day·10a)) Small 27.0 3.37 −0.103 80.5 8.50 −0.644 3.0 0.33 −0.033 Medium 7.7 0.76 0.051 122.4 12.03 0.36 15.9 0.23 −0.071 Large 3.6 0.60 −0.148 124.4 21.49 −5.033 34.3 0.55 −0.009 Heavy 1.5 0.50 −0.13 121.7 42.61 −10.506 79.6 5.19 −1.255 Note. SD, standard deviation; RC, regression coefficient. precipitation. ,e precipitation amount of small pre- events, both have negative and significant correlations with cipitation ranges from 65.1 to 95.9 mm, which declines from elevation at a confidence level of 0.01. In conclusion, to a northwest to southeast. However, it declines from north to certain extent, the spatial distributions of precipitation south for the medium precipitation, ranging from 95.9 to frequency and amount are affected by the terrain. ,e large 148.9 mm. Additionally, the spatial distribution of pre- and heavy precipitations frequently occur in the plain areas, cipitation frequency and amount for the large and heavy especially for the piedmont plain areas. But, in contrast, the precipitation of the mountain areas is dominated by the precipitation are almost similar to the distribution of the annual precipitation (Figure 6(a)). ,e relationship between small and medium precipitations both in frequency and the precipitation frequency and amount for different pre- amount. cipitation grades and elevation was also discussed in this work, as shown in Figure 10. For small and medium pre- cipitation events, both frequency and amount have positive 3.4. Differences in Changes of Precipitation in Four Subregions. correlations with elevation, with no significance for fre- ,e variations of annual precipitation series for different quency in small precipitation and amount in medium subregions (i.e., plain area, mountain area, urban area, and precipitation events. While for large and heavy precipitation suburban area) are shown in Figure 11, with the statistical 14 Advances in Meteorology 0.8 0.3 0.2 0.7 0.1 0.6 0 0.0 0.5 Small Medium Large Heavy Small Medium Large Heavy (a) (b) 0.5 0.4 0.3 0.2 0.1 0 0.0 Small Medium Large Heavy Small Medium Large Heavy (c) (d) Figure 8: Box-plots of precipitation frequency (a), precipitation amount (b), precipitation incidence (c), and precipitation contribution (d) for different grades of precipitation over Beijing during 1960–2012. ,e boxes indicate the 25th, 50th, and 75th percentiles and the whiskers indicate the values with maximum 1.5 interquartile ranges. ,e hollow square indicates the mean value. ,e solid circle means the outlier. results given in Table 7. As a whole, we found that the difference has expanded to 32.6 mm (accounting for 5.72% of the total precipitation in the plain areas) between the areas average precipitation amount in the area of the Beijing Plain (urban area) was greater than that in the mountainous in the plain and mountains and 45.2 mm (accounting for (suburban) areas. As seen in Figure 11, the precipitation 7.66% of the total precipitation in the urban areas) between showed slight decreasing trends at the rate of 6.4 and the urban areas surrounding suburb areas. It is clear that the 9.9 mm/decade for the plain and mountain areas average precipitation decreased both in the urban and (Figure 11(a)), 4 and 4.9 mm/decade for the urban and suburb areas through these two periods, while the drying was suburban areas (Figure 11(c)) during 1960–2012, re- larger in the suburb areas, which caused the higher differ- spectively. However, as shown in Figure S2, from the ences in precipitation between the urban and surrounding standpoint of the changes in built-up areas, the urban ex- suburb areas. With rapid development of urbanization, the height and density of buildings are both increasing, and the pansion has experienced three stages in Beijing during the research period: the first stage is the slowest urban expansion rainfall under a weather system in urban areas would stay longer than that in open suburb areas [56]; thus, the total before 1980, the second is the rapid expansion stage from 1980 to 2000, and the third is the fastest urban growing precipitation increased. Additionally, in the process of ur- period during the first decade of the 21st century [3]. banization, population density increases in cities, releasing ,erefore, the changes in precipitation during the three more heat into the atmosphere and changing the impervi- stages for these areas are also discussed. During 1960–1979, ousness of underlying areas. All these factors lead to an the difference of the precipitation amount between areas in urban heat island effect, leading to precipitation variations in the plain and the mountains was about 24.2 mm, which urban areas. However, during 2000–2012, the precipitation accounts for 4.15% of the total precipitation in the plain amount in the above four areas has declined by 49.8– areas, while the difference between urban areas and suburb 67.9 mm compared with the last period due to the long-term drought (1999–2010) in the north China, accounting for areas was 32.9 mm, accounting for 5.51% of the total pre- cipitation in the urban areas. During 1980–1999, the 8.5–11.6% of the average annual precipitation (584.7 mm) Frequency (days) Precipitation amount (mm) PI PC Advances in Meteorology 15 117°E 117°E 116°E 116°E N N 41°N 41°N W E W E THK THK QJD QJD ZJF XH ZJF XH YQ YQ HHC HHC MY MY ZLY ZLY HR HR GT YK YK GT WJY WJY PG PG SH SH SY SY SZ SZ YC YC 40°N 40°N ZT ZT SJD SJD SLZ SLZ TZ TZ GRD GRD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF GZ GZ 0 20 40 10 0 10 20 40 km km Frequency-small (days) Frequency-medium (days) 26.9–28.0 7.9–8.1 17.9–21.6 6.0–6.6 28.1–29.3 8.2–8.5 21.7–23.8 6.7–7.1 29.4–31.2 8.6–9.2 23.9–25.4 7.2–7.5 25.5–26.8 7.6–7.8 (a) (b) Figure 9: Continued. 116°E 117°E 117°E 116°E 41°N 41°N W E W E THK THK QJD QJD ZJF XH ZJF XH YQ YQ HHC MY HHC MY ZLY ZLY HR HR GT YK GT YK WJY WJY PG PG SH SH SY SY SZ SZ YC YC 40°N 40°N ZT SJD ZT SJD SLZ TZ SLZ TZ GRD GRD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF GZ GZ 0 10 20 40 0 10 20 40 km km Frequency-large (days) Frequency-heavy (days) 3.8–3.9 1.5–1.7 2.4–2.8 0.5–0.8 4.0–4.2 1.7–1.9 2.9–3.1 0.9–1.0 4.3–4.9 1.9–2.2 3.2–3.4 1.0–1.3 3.56–3.7 1.3–1.5 (c) (d) Figure 9: Continued. 16 Advances in Meteorology 116°E 117°E 116°E 117°E 41°N 41°N W E W E THK THK QJD QJD ZJF XH ZJF XH YQ YQ HHC HHC MY MY ZLY ZLY HR HR GT YK GT YK WJY WJY mm PG PG SH SH SY SY SZ SZ YC YC 40°N 40°N ZT SJD ZT SJD SLZ TZ SLZ TZ GRD GRD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF GZ GZ 0 20 40 10 0 10 20 40 km km Amount-small (mm) Amount-medium (mm) 83.3–86.5 124.2–129.1 65.1–71.4 95.9–106.3 86.6–89.8 129.2–136.6 71.5–75.1 106.4–114.1 89.9–95.9 136.7–148.9 75.2–79.2 114.2–119.3 79.3–83.2 119.4–124.1 (e) (f) 116°E 117°E 116°E 117°E 41°N 41°N W E W E THK THK QJD QJD ZJF XH ZJF XH YQ YQ HHC MY HHC MY ZLY ZLY HR HR GT YK GT YK WJY WJY PG PG SH SH SY SY SZ SZ YC YC 40°N 40°N ZT ZT SJD SJD SLZ SLZ TZ TZ GRD GRD YAM YAM YLZ YLZ MJQ MJQ XYL HC XYL HC FS FS ZF ZF GZ GZ 0 40 0 10 20 40 10 20 km km Amount-large (mm) Amount-heavy (mm) 128.2–137.7 119.4–134.2 79.7–97.9 39.5–66.1 134.3–149.7 137.8–148.8 66.2–84.4 98.0–107.8 148.9–170.5 149.8–180.8 107.9–118.5 84.5–102.6 102.7–119.3 118.6–128.1 (g) (h) Figure 9: Spatial distribution of precipitation frequency and amount for the different grades during 1960–2012. (a, e) Small precipitation. (b, f) Medium precipitation. (c, g) Large precipitation. (d, h) Heavy precipitation. over Beijing. During this period, Table 7 showed that the slightly increasing trend at a rate of 3.49 mm/decade during difference dropped to 17.7 mm (accounting for 3.51% of the the period of 1960–2012 (Figure 12(a)). ,e increasing rate total precipitation in the plain areas) between the plain and for the difference between the plain and mountain areas was mountains and 31.0 mm (accounting for 5.94% of the total larger than that for the difference between urban and precipitation in the urban areas) in the urban and sur- suburban areas (0.87 mm/decade) (Figure 12(b)). However, rounding suburb areas. Moreover, the difference in pre- the results of M-K test and Sen’s slope estimate showed that cipitation between the plain and mountain areas had a there was a nonsignificant decreasing trend for the Advances in Meteorology 17 35 5.0 2.5 4.5 2.0 4.0 1.5 3.5 3.0 1.0 2.5 R = –0.62 R = –0.722 R = 0.388 0.5 R = 0.348 p < 0.01 p < 0.01 2.0 p = 0.034 p = 0.060 15 1.5 0.0 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 Elevation (m) Elevation (m) Elevation (m) Elevation (m) (a) (b) (c) (d) 100 200 R = –0.65 40 R = 0.302 80 R = –0.706 R = 0.755 p < 0.01 p = 0.105 p < 0.01 p < 0.01 60 90 60 0 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 Elevation (m) Elevation (m) Elevation (m) Elevation (m) (e) (f) (g) (h) Figure 10: Spatial distribution of precipitation frequency and amount for the different grades during 1960–2012. (a, e) Small precipitation. (b, f) Medium precipitation. (c, g) Large precipitation. (d, h) Heavy precipitation. 800 800 y = 6.64x – 12501.04 y = 7.64x – 14629.68 y = 21.34x – 42302.58 y = –0.64x + 1834.38 600 600 y = –0.99x + 2502.44 y = 5.06x – 9524.98 y = 4.94x – 9162.75 300 300 y = 6.99x – 13526.99 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Plain area Linear trend (plain area) Plain area Linear trend (plain area) Mountain area Linear trend (mountain area) Mountain area Linear trend (mountain area) (a) (b) Figure 11: Continued. Precipitation amount (mm) Precipitation frequency (days) Precipitation (mm) Precipitation amount (mm) Precipitation frequency (days) Precipitation (mm) Precipitation amount (mm) Precipitation frequency (days) Precipitation amount (mm) Precipitation frequency (days) 18 Advances in Meteorology 900 900 y = 4.08x – 7435.81 800 800 y = –0.40x + 1369.77 y = 10.50x – 20307.50 y = 29.32x – 58289.38 700 700 600 600 500 500 400 400 y = –0.49x + 1505.99 y = 7.62x – 14618.98 300 300 y = 22.05x – 43737.46 y = 7.89x – 14969.22 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years Urban area Linear trend (urban area) Urban area Linear trend (urban area) Suburban area Linear trend (suburban area) Suburban area Linear trend (suburban area) (c) (d) Figure 11: Time series of annual mean precipitation in the plain and mountain areas (top panel) and the urban and suburb areas (bottom panel). ,e linear trend analyses are discussed based on the different periods: one (left panel) is based on the whole period (1960–2012) and another one (right panel) is based on the three subperiods (1960–1979, 1980–2000, and 2001–2012) considering the different urban expansion stages. Table 7: Comparison of mean precipitation in the different subregions and periods. 1 1 Plain area Mountainous area Percent of differences (%) Urban area Suburb area Percent of differences (%) 1960–1979 583.2 559.0 4.15 597.0 564.1 5.51 1980–1999 569.6 537.0 5.72 589.9 544.7 7.66 2000–2012 504.8 487.1 3.51 522.0 491.0 5.94 1960–2012 558.8 533.0 4.62 575.9 538.9 6.42 Note. ,e percent of differences � the differences in precipitation between the plain (urban) and mountainous (suburb) areas divided by the total amount of precipitation in the plain (urban) areas. differences in precipitation between the urban and suburban ,ese changes, to some extent, lead to the significant change areas. in spring and summer. Overall, there is no change in a strongly uneven spa- tiotemporal distribution in precipitation within the Beijing 4. Conclusions metropolis. Although the PCI and PCD both have a de- In this study, a comprehensive investigation was performed creasing trend, they are both in a relatively high level, which on the precipitation changes across Beijing based on the is an inherent characteristic for the East Asian Monsoon monthly, seasonal, and annual mean precipitation from 30 areas. ,ese declines may be dominantly caused by the rain gauges from 1960 to 2012. Many time series analysis significant decrease in precipitation in summer and the methods, evaluating indicators, and tools were used to assess increases in spring and autumn. Simultaneously, the pre- the variations of precipitation structure. Based on our cipitation roughly decreases from east to west due to the analysis, main findings are summarized as follows. orographic effects, with the highest rain occurring at the ,e annual mean precipitation has a clear decreasing Miyun and Huairou reservoirs in the northeast and the lowest in the northwestern mountain areas. In addition, the trend at a rate of 11.6 mm/10a during the period of 1960– 2012. Seasonally, a significant decrease in precipitation effects of urban development on precipitation become more and more obvious. occurred in summer (21.1 mm/10a). In contrast, the pre- cipitation in spring shows a significant increasing trend with ,e spatiotemporal changes for the four precipitation a rate of 4.7 mm/10a. ,ere are nonsignificant trends in grades (small, median, large, and heavy) are different. In autumn and winter. However, this increase in spring is general, the frequency for the small precipitation is the unable to offset the remarkable decrease in summer, which is largest, followed by the median and large precipitations, and a dominant driver for the decline in annual precipitation. that for the heavy precipitation is the lowest. On average, the Similarly, the monthly series of precipitation in January, amount for the small precipitation is the lowest and the February, July, and August show a significant decreasing other three grades are almost equivalent. ,e spatial dis- trend while that in May has a statistically significant increase. tribution for different grades are affected by the terrain; the Precipitation (mm) Precipitation (mm) Advances in Meteorology 19 250 250 y = 0.349x – 668.06 y = 0.087x – 136.22 –50 –50 –100 –100 1960 1970 1980 1990 2000 2010 1960 1970 1980 1990 2000 2010 Years Years (a) (b) Figure 12: Differences in precipitation and their linear trends: (a) between the plain and mountain areas, (b) between the urban and suburban areas. n−1 n large and heavy precipitation frequently occur in the plain S � 􏽘 􏽘 sgn x − x , areas, while the precipitation in the mountain areas is 􏼐 􏼑 (A.1) 0 j k k�1 j�k+1 dominated by the small and medium precipitations. In the current context of global climate change and rapid where n is the length of the series, k � 1, 2, . . ., n− 1, j � 2, urban development in Beijing, adaptations for avoiding or 3, . . ., n, and minimising their impacts on water resources require analysis of the change in precipitation patterns on a fine spatio- ⎧ ⎪ 1, 􏼐x − x 􏼑> 0, ⎪ j k temporal scale. ,e adverse natural conditions and the rapid ⎨ sgn􏼐x − x 􏼑 � 0, 􏼐x − x 􏼑 � 0, (A.2) j k j k urban development in Beijing lead to the high risk in the ⎪ urban water resources management. For example, Beijing ⎩ −1, 􏼐x − x 􏼑< 0. j k suffers the problems both in water deficit and urban flood and waterlogging in recent years. To some extent, these It has been proven that when n≥ 8, S follows approx- problems are probably caused by the change in precipitation imately the normal distribution with 0 mean and the var- pattern. ,erefore, the future work should be paid attention iance as to the relationship between the change in precipitation and [n(n− 1)(2n + 5)] the urban water resources safety, in order to obtain deep (A.3) Var S � . understanding and knowledge. Standardized statistic Z can be Calculated as Appendix S − 1 ⎧ ⎪ 􏽱������� ⎪ , S > 0, Var S 􏼁 Mann–Kendall trend test and Sen’s slope estimate method ⎪ were widely used to detect trends in hydrological and me- teorological time series. ,e detailed information was in- Z � 0, S � 0, (A.4) troduced as follows: In the M-K trend test, the null hypothesis (H ) is that the 0 ⎪ S + 1 data in a time series are independent and identically distributed ⎪ 􏽱������� , S < 0. ⎪ 0 random variables and the hypothesis (H ) is that there is a trend 1 ⎩ Var S 􏼁 in the series. Statistical parameter (S ) is defined as Difference in precipitation (mm) Difference in precipitation (mm) 20 Advances in Meteorology Negative Z value denotes downward trend and positive Z National Natural Science Foundation of China (51609242), value shows upward trend. ,e trend is significant at the 95% the National Key Research and Development Program confidence level |Z|> 1.96 and vice versa. (2017YFC1502701), the China Postdoctoral Science Foun- In Sen’s slope estimator method, the slope of trend in the dation (2018M632333), the Open Research Fund Program of sample of N pairs of data is calculated by State Key Laboratory of Water Resources and Hydropower x − x Engineering Science (2015SWG02), and the Open Research j k Q � , for i � 1, 2, . . . , N, (A.5) i Fund Program of State Key Laboratory of Hydrology-Water j− k Resources and Hydraulic Engineering (2015490411). ,e authors would like to thank the Beijing Hydrologic Stations where x and x are the data values at times j and k (j> k), j k respectively. If there is only one datum in each time period, of Beijing Water Authority in China for providing all of the precipitation data. then N � n(n− 1)/2, where n is the number of time periods. If there are multiple observations in one or more time periods, then N< n(n− 1)/2, where n is the total number of Supplementary Materials observations. ,e N values of Q are ranked from smallest to largest Figure S1: the incidence (PI) and contribute (PC) of different and the median of slope is computed as precipitation grades to the total frequency and amount for the 30 stations during 1960–2012. Figure S2: the urban built- Q , if N is odd, ⎧ ⎪ [(N+1)/2] up areas in Beijing during 1950–2012. (Supplementary Q � (A.6) med Materials) ⎪ Q + Q [N/2] [(N+2)/2] , if N is even. References ,e Q sign reflects data trend reflection, while its med value indicates the steepness of the trend. To determine [1] C. Chen and W. Huang, “Hydrological modeling of typhoon- whether the median slope is statistically different from zero, induced extreme storm runoffs from Shihmen watershed to one should obtain the confidence interval of Q at specific reservoir, Taiwan,” Natural Hazards, vol. 67, no. 2, pp. 747– med probability. ,e confidence interval about the time slope can 761, 2013. [2] H. Chang and W. T. Kwon, “Spatial variations of summer be computed as follows: 􏽱������� precipitation trends in South Korea, 1973–2005,” Environ- C � Z Var S , (A.7) 􏼁 mental Research Letters, vol. 2, no. 4, article 045012, 2007. α 1−α/2 0 [3] X. Song, J. Zhang, A. AghaKouchak et al., “Rapid urbanization where Z is obtained from the standard normal distri- and changes in spatiotemporal characteristics of precipitation 1−α/2 in Beijing metropolitan area,” Journal of Geophysical Research: bution table. In this study, the confidence interval was Atmospheres, vol. 119, no. 19, pp. 11250–11271, 2014. computed at a significant level of α � 0.05. ,en, M � [4] Y. Zhai, Y. Guo, J. Zhou, N. Guo, J. Wang, and Y. Teng, “,e (N− C )/2 and M � (N + C )/2 are computed. ,e lower and α 2 α spatio-temporal variability of annual precipitation and its upper limits of the confidence interval, Q and Q , are min max local impact factors during 1724-2010 in Beijing, China,” th th the M largest and the (M + 1) largest of the N ordered 1 2 Hydrological Processes, vol. 28, no. 4, pp. 2192–2201, 2014. slope estimates. ,e slope Q is statistically different from med [5] S. Dawadi and S. Ahmad, “Changing climatic conditions in zero if the two limits have similar sign. the Colorado River Basin: implications for water resources management,” Journal of Hydrology, vol. 430-431, pp. 127– Data Availability 141, 2012. [6] V. Kumar, S. K. Jain, and Y. Singh, “Analysis of long-term Rainfall data supporting this article are available from the rainfall trends in India,” Hydrological Sciences Journal, vol. 55, Hydrological Data of Haihe River Basin (in Chinese) and the no. 4, pp. 484–496, 2010. [7] A. Kalra and S. Ahmad, “Evaluating changes and estimating Annual Hydrological Report of China (Volume III), released seasonal precipitation for the Colorado River Basin using a by Ministry of Water Resources of China. For further in- stochastic nonparametric disaggregation technique,” Water formation or right to access to the material used in this Resources Research, vol. 47, no. 5, article W05555, 2011. paper, readers can also contact the Beijing Hydrological [8] S. Sagarika, A. Kalra, and S. Ahmad, “Evaluating the effect of Stations of the Beijing Water Authority (http://www.bjwater. persistence on long-term trends and analyzing step changes in gov.cn). For the other data used in this article, readers can streamflows of the continental United States,” Journal of contact the corresponding author X. Song. Hydrology, vol. 517, pp. 36–53, 2014. [9] J. Wang, J. Feng, Z. Yan, Y. Hu, and G. Jia, “Nested high- Conflicts of Interest resolution modeling of the impact of urbanization on re- gional climate in three vast urban agglomerations in China,” ,e authors declare that there are no conflicts of interest Journal of Geophysical Research: Atmospheres, vol. 117, regarding the publication of this paper. no. D21, 2012. [10] J. Zhang, X. Song, G. Wang et al., “Development and chal- Acknowledgments lenges of urban hydrology in a changing environment: I. Hydrological response to urbanization,” Advances in Water ,is work was supported by the Fundamental Research Science, vol. 25, no. 4, pp. 594–605, 2014, in Chinese with Funds for the Central Universities (2015XKMS034), the English abstract. Advances in Meteorology 21 [11] E. Kalnay and M. Cai, “Impact of urbanization and land-use homogeneized observations,” Advances in Atmospheric Sci- change on climate,” Nature, vol. 423, no. 6939, pp. 528–531, ences, vol. 32, no. 9, pp. 1173–1185, 2015. 2003. [28] S. Wu, F. Chen, M. A. LeMone, M. Tewari, Q. Li, and Y. Wang, “An observational and modeling study of characteristics of [12] D. Rosenfeld, “Suppression of rain and snow by urban and industrial air pollution,” Science, vol. 287, no. 5459, urban heat island and boundary layer structures in Beijing,” pp. 1793–1796, 2000. Journal of Applied Meteorology and Climatology, vol. 48, no. 3, [13] D. Rosenfeld, U. Lohmann, G. B. Raga et al., “Flood or pp. 484–501, 2009. drought: how do aerosols affect precipitation?,” Science, [29] S. Miao, F. Chen, and Q. Li, “Month-averaged impacts of vol. 321, no. 5894, pp. 1309–1313, 2008. urbanization on atmospheric boundary layer structure and [14] Y. Huang, S. Chen, Q. Cao et al., “Evaluation of version-7 precipitation in summer in Beijing area,” Chinese Journal of TRMM multi-satellite precipitation analysis product during Geophysics, vol. 53, no. 7, pp. 1580–1593, 2010, in Chinese the Beijing extreme heavy rainfall events of 21 July 2012,” with English abstract. Water, vol. 6, pp. 32–44, 2014. [30] Z. Xu, L. Zhang, and B. Ruan, “Analysis on the spatio- [15] M. Li, J. Xia, Z. Chen et al., “Variation analysis of precipitation temporal distribution of precipitation in the Beijing re- during past 286 years in Beijing area, China, using non- gion,” Arid Land Geography, vol. 29, no. 2, pp. 186–192, 2006, parametric test and wavelet analysis,” Hydrological Pro- in Chinese with English abstract. [31] S. Yu, “Decadal rainfall variability over Beijing and the urban cesses, vol. 27, pp. 2934–2943, 2013. [16] J.-Y. Han, J.-J. Baik, and H. Lee, “Urban impacts on pre- effect,” Advances in Nature Science, vol. 17, no. 5, pp. 632–638, cipitation,” Asia-Pacific Journal of Atmospheric Sciences, 2007, in Chinese with English abstract. vol. 50, no. 1, pp. 17–30, 2014. [32] Y. Zhang, J. A. Smith, L. Luo, Z. Wang, and M. L. Baeck, [17] Z. Han, Z. Yan, Z. Li, W. Liu, and Y. Wang, “Impact of “Urbanization and rainfall variability in the Beijing metro- urbanization on low-temperature precipitation in Beijing politan region,” Journal of Hydrometeorology, vol. 15, no. 6, during 1960-2008,” Advances in Atmospheric Sciences, vol. 31, pp. 2219–2235, 2014. no. 1, pp. 48–56, 2014. [33] C. Zhang, C. Ji, Y. Kuo et al., “Numerical simulation of to- [18] J. Wang, R. Zhang, and Y. Wang, “Areal differences in diurnal pography effects on the “00.7” severe rainfall in Beijing,” variations in summer precipitation over Beijing metropolitan Progress in Natural Science, vol. 15, no. 9, pp. 818–826, 2005. region,” Beoretical and Applied Climatology, vol. 110, no. 3, [34] J. Sun, H. Wang, L. Wang et al., “,e role of urban boundary pp. 395–408, 2012. layer in local convective torrential rain happening in Beijing [19] L. Yang, F. Tian, J. A. Smith, and H. Hu, “Urban signatures in on 10 July 2004,” Chinese Journal of Atmospheric Sciences, vol. 30, pp. 221–234, 2006, in Chinese with English abstract. the spatial clustering of summer heavy rainfall events over the Beijing metropolitan region,” Journal of Geophysical Research: [35] Q. Wu, H. Guo, B. Yang et al., “Effects of topography and Atmospheres, vol. 119, no. 3, pp. 1203–1217, 2014. urban heat circulation on a meso-β torrential rain in Beijing [20] N. Molders ¨ and M. A. Olson, “Impact of urban effects on area,” Meteorology Monthly, vol. 35, pp. 58–64, 2009, in precipitation in high latitudes,” Journal of Hydrometeorology, Chinese with English abstract. vol. 5, no. 3, pp. 409–429, 2004. [36] J. Sun and B. Yang, “Meso-β scale torrential rain affected by [21] J. M. Shepherd, H. Pierce, and A. J. Negri, “Rainfall modi- topography and the urban circulation,” Chinese Journal of fication by major urban areas: observations from spaceborne Atmospheric Sciences, vol. 32, pp. 1352–1364, 2008, in Chinese rain radar on the TRMM satellite,” Journal of Applied Me- with English abstract. teorology, vol. 41, no. 7, pp. 689–701, 2002. [37] A. Bardossy ´ and G. Pegram, “Infilling missing precipitation [22] C. L. Zhang, F. Chen, S. G. Miao, Q. C. Li, X. A. Xia, and records - a comparison of a new copula-based method with other techniques,” Journal of Hydrology, vol. 519, pp. 1162– C. Y. Xuan, “Impacts of urban expansion and future green planting on summer precipitation in the Beijing metropolitan 1170, 2014. area,” Journal of Geophysical Research, vol. 114, no. D2, [38] R. S. V. Teegavarapu and V. Chandramouli, “Improved p. D02116, 2009. weighting methods, deterministic and stochastic data-driven [23] S. Dawadi and S. Ahmad, “Evaluating the impact of demand- models for estimation of missing precipitation records,” side management on water resources under changing climatic Journal of Hydrology, vol. 312, no. 1–4, pp. 191–206, 2005. conditions and increasing population,” Journal of Environ- [39] X. M. Song, J. Y. Zhang, J. F. Liu et al., “Spatial-temporal mental Management, vol. 114, pp. 261–275, 2013. variation characteristics of precipitation pattern in Beijing,” [24] A. K. Venkatesan, S. Ahmad, W. Johnson, and J. R. Batista, Journal of Hydraulic Engineering, vol. 46, no. 5, pp. 525–535, “Systems dynamic model to forecast salinity load to the 2015, in Chinese with English abstract. Colorado River due to urbanization within the Las Vegas [40] J. E. Oliver, “Monthly precipitation distribution: a compar- Valley,” Science of the Total Environment, vol. 409, no. 13, ative index,” Professional Geographer, vol. 32, no. 3, pp. 2616–2625, 2011. pp. 300–309, 1980. [41] L. J. Zhang and Y. F. Qian, “Annual distribution features of [25] X. Batista, Z. Wang, Y. Qi, and H. Guo, “Effect of urbanization on the winter precipitation distribution in Beijing area,” precipitation in China and their interannual variations,” Acta Science in China Series D: Earth Sciences, vol. 52, no. 2, Meteorologica Sinica, vol. 17, no. 2, pp. 146–163, 2003. pp. 250–256, 2009. [42] X. M. Li, F. Q. Jiang, L. H. Li et al., “Spatial and temporal [26] S. Guo and J. Ma, “Impact of urbanization on precipitation in variability of precipitation concentration index, concentration Beijing area,” Journal of the Meteorological Sciences, vol. 31, degree and concentration period in Xinjiang, China,” In- no. 4, pp. 414–421, 2011, in Chinese with English abstract. ternational Journal of Climatology, vol. 31, pp. 1679–1693, [27] Z. Li, Z. Yan, K. Tu, and H. Wu, “Changes of precipitation and 2011. extremes and the possible effect of urbanization in the Beijing [43] S. Fatichi, S. M. Barbosa, E. Caporali, and M. E. Silva, “De- Metropolitan region during 1960-2012 based on terministic versus stochastic trends: detection and 22 Advances in Meteorology challenges,” Journal of Geophysical Research, vol. 114, no. D18, article D18121, 2009. [44] M. G. Silva, Rank Correlation Methods, Griffin, London, UK, [45] H. B. Mann, “Nonparametric tests against trend,” Econo- metrica, vol. 13, no. 3, pp. 245–259, 1945. [46] P. K. Sen, “Estimates of the regression coefficient based on Kendall’s Tau,” Journal of the American Statistical Association, vol. 63, no. 324, pp. 1379–1389, 1968. [47] Z. Bao, J. Zhang, G. Wang et al., “Attribution for decreasing streamflow of the Haihe River basin, northern China: climate variability or human activities?,” Journal of Hydrology, vol. 460-461, pp. 117–129, 2012. [48] W. Shi, X. Yu, W. Liao, Y. Wang, and B. Jia, “Spatial and temporal variability of daily precipitation concentration in the Lancang River basin, China,” Journal of Hydrology, vol. 495, pp. 197–207, 2013. [49] 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. [50] Y. Dinpashoh, D. Jhajharia, A. Fakheri-Fard et al., “Trends in reference crop evapotranspiration over Iran,” Journal of Hydrology, vol. 399, no. 3-4, pp. 422–433, 2011. [51] T. B. M. J. Ouarda, C. Charron, K. Niranjan Kumar et al., “Evolution of the rainfall regime in the United Arab Emir- ates,” Journal of Hydrology, vol. 514, pp. 258–270, 2014. [52] A. R. Tome and P. M. A. Miranda, “Piecewise linear fitting and trend changing points of climate parameters,” Geophysical Research Letters, vol. 31, no. 2, p. L02207, 2004. [53] A. N. Pettitt, “A non-parametric approach to the change- point problem,” Applied Statistics, vol. 28, no. 2, pp. 126–135, [54] G. Villarini, F. Serinaldi, J. A. Smith, and W. F. Krajewski, “On the stationarity of annual flood peaks in the continental United States during the 20th century,” Water Resources Research, vol. 45, no. 8, article W08417, 2009. [55] C. Rouge, ´ Y. Ge, and X. Cai, “Detecting gradual and abrupt changes in hydrological records,” Advances in Water Re- sources, vol. 53, pp. 33–44, 2013. [56] M.-N. Yang, Y.-P. Xu, G.-B. Pan, and L.-F. Han, “Impacts of urbanization on precipitation in Taihu Lake basin, China,” Journal of Hydrologic Engineering, vol. 19, no. 4, pp. 739–746, 2014. International Journal of The Scientific Advances in Advances in Geophysics Chemistry Scientica World Journal Public Health Hindawi Hindawi Hindawi Hindawi Publishing Corporation Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 Journal of Environmental and Public Health Advances in Meteorology Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Applied & Environmental Journal of Soil Science Geological Research Hindawi Volume 2018 Hindawi www.hindawi.com www.hindawi.com Volume 2018 International Journal of International Journal of Agronomy Ecology International Journal of Advances in International Journal of Forestry Research Microbiology Agriculture Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 International Journal of Journal of Journal of International Journal of Biodiversity Archaea Analytical Chemistry Chemistry Marine Biology Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal

Advances in MeteorologyHindawi Publishing Corporation

Published: Mar 6, 2019

References