Changes in daily climate extremes in the eastern and ... · Tank et al., 2006]. Global changes in...

Changes in daily climate extremes in the eastern and central Tibetan

Plateau during 1961–2005

Qinglong You,1,2 Shichang Kang,1,3 Enric Aguilar,4 and Yuping Yan5

Received 14 September 2007; revised 5 December 2007; accepted 8 January 2008; published 1 April 2008.

[1] Changes in indices of climate extremes are analyzed on the basis of daily maximumand minimum surface air temperature and precipitation at 71 meteorological stations withelevation above 2000 m above sea level in the eastern and central Tibetan Plateau (TP)during 1961–2005. Twelve indices of extreme temperature and nine indices of extremeprecipitation are examined. Temperature extremes show patterns consistent with warmingduring the studied period, with a large proportion of stations showing statisticallysignificant trends for all temperature indices. Stations in the northwestern, southwestern,and southeastern TP have larger trend magnitudes. The regional occurrence of extremecold days and nights has decreased by �0.85 and �2.38 d/decade, respectively. Over thesame period, the occurrence of extreme warm days and nights has increased by 1.26 and2.54 d/decade, respectively. The number of frost days and ice days shows statisticallysignificant decreasing at the rate of �4.32 and �2.46 d/decade, respectively. The length ofgrowing season has statistically increased by 4.25 d/decade. The diurnal temperature rangeexhibits a statistically decreasing trend at a rate of �0.20�C per decade. The extremetemperature indices also show statistically significant increasing trends, with larger valuesfor the index describing variations in the lowest minimum temperature. In general,warming trends in minimum temperature indices are of greater magnitude than thosefor maximum temperature. Most precipitation indices exhibit increasing trends in thesouthern and northern TP and show decreasing trends in the central TP. On average,regional annual total precipitation, heavy precipitation days, maximum 1-dayprecipitation, average wet days precipitation, and total precipitation on extreme wet daysshow nonsignificant increases. Decreasing trends are found for maximum 5-dayprecipitation, consecutive wet days, and consecutive dry days, but only the last isstatistically significant.

Citation: You, Q., S. Kang, E. Aguilar, and Y. Yan (2008), Changes in daily climate extremes in the eastern and central Tibetan

Plateau during 1961–2005, J. Geophys. Res., 113, D07101, doi:10.1029/2007JD009389.

1. Introduction

[2] The Fourth Assessment Report of the Intergovern-mental Panel on Climate Change (IPCC) shows an increasein global mean temperature of approximately 0.74�C duringthe latest century [Intergovernmental Panel on ClimateChange, 2007]. In the context of global warming, variationsand trends in extreme climate events have recently receivedmuch attention because extreme climate events are moresensitive to climate change than their mean values [Katz and

Brown, 1992]. At the same time, global climate change isexpected to have a considerable impact on the globalhydrological cycle. It has been seen that the economy,human health and the natural environment are becomingvulnerable to the extreme climate events [Easterling et al.,2000; Kunkel et al., 1999].[3] Precipitation and temperature extremes have been

studied in many regions around the world, such as in theAsia-Pacific region [Griffiths et al., 2005; Manton et al.,2001], Caribbean region [Peterson et al., 2002], southernand west Africa [New et al., 2006], South America [Haylocket al., 2006; Vincent et al., 2005], Middle East [X. Zhang etal., 2005], Central America and northern South American[Aguilar et al., 2005], and central and south Asia [KleinTank et al., 2006]. Global changes in daily climate extremeshave been analyzed [Alexander et al., 2006; Frich et al.,2002]. These studies concluded that widespread significantchanges in temperature extremes are associated with warm-ing, while the changes in precipitation extremes presentmuch less spatially coherence compared with temperaturechange. For China, precipitation has increased by 2% and

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 113, D07101, doi:10.1029/2007JD009389, 2008ClickHere

for

FullArticle

1Institute of Tibetan Plateau Research, Chinese Academy of Sciences,Beijing, China.

2Also at Graduate University of Chinese Academy of Sciences, Beijing,China.

3Also at State Key Laboratory of Cryospheric Science, ChineseAcademy of Sciences, Lanzhou, China.

4Climate Change Research Group, Geography Unit, Universitat RoviraiVirgili de Tarragona, Tarragona, Spain.

5National Climate Center, Beijing, China.

Copyright 2008 by the American Geophysical Union.0148-0227/08/2007JD009389$09.00

D07101 1 of 17

the frequency of precipitation events has decreased by 10%from 1960 to 2000 [B. Liu et al., 2005], mean minimumtemperature has increased significantly and mean maximumtemperatures display no statistically significant trend between1950 and 1995 [Zhai et al., 1999].[4] The Tibetan Plateau (TP), with an average elevation

of over 4000 m above sea level (asl), is the highest andlargest highland in the world and exerts a great influence onregional and global climate through its thermal forcingmechanisms [Duan and Wu, 2005; Yanai et al., 1992; Yehand Gao, 1979]. Previous studies [Duan and Wu, 2006;Kang et al., 2007; Lin and Zhao, 1996; X. Liu and Chen,2000; Niu et al., 2004] showed a significant warming in theTP during the last half century, in phase with the globaltrends derived from the increasing anthropogenic green-house gases emissions. The TP region is expected to be oneof the most seriously impacted areas in the world by globalwarming effects [Chen et al., 2003; Duan et al., 2006].However, there have been few studies in temperature andprecipitation extremes in the TP, primarily owing to the lackof easily available data collection for the region. Forprecipitation a clear regional signal has not been identifiedmainly due to the complex terrain and sparse meteorologicalstations [Du and Ma, 2004; Lin and Zhao, 1996; X Liu andYin, 2001]. Du [2001] analyzed monthly maximum andminimum temperatures for 16 stations in the TP from1961–2000 and found that the magnitude of trend inminimum temperature was greater than that in maximum

temperature. X. Liu et al. [2006] also confirmed the asym-metric pattern of greater warming trends in nighttimetemperatures as compared to the daytime temperatures.[5] The objective of this study is to investigate the

climate change in temperature and precipitation extremesduring the period 1961–2005 in the eastern and central TP,through the analysis of indices generated by the Commis-sion for Climatology (CCl)/Climate Variability and Predict-ability (CLIVAR)/Joint WMO-IOC Technical Commissionfor Oceanography and Marine Meteorology (JCOMM)Expert Team (ET) on Climate Change Detection and Indices(ETCCDI) (http://cccma.seos.uvic.ca/ETCCDI/), a widelyused approach (see section 2 for more). We have analyze therelationship between trends in temperature extremes andelevation using the same method in this region [You et al.,2008]. Analyzing these indices will hopefully lead to abetter understanding of variability and changes in thefrequency, intensity and duration of extreme climate eventsin the TP. Spatial and temporal variability of the changes intemperature and precipitation extremes are discussed in thiswork.

2. Data and Methods

[6] Data including daily precipitation, maximum temper-ature and minimum temperature is provided by the NationalClimate Center, China Meteorological Administration. TheTP in China ranges from 26�0001200 to 39�4605000N and from

Figure 1. The distribution of 71 stations used in this study in the eastern and central Tibetan Plateau(TP) and adjacent territories.

D07101 YOU ET AL.: CHANGES IN CLIMATE EXTREMES IN THE TIBETAN PLATEAU

2 of 17

D07101

Table 1. List of the Selected Stations Above 2000 m Above Sea Level in the Eastern and Central Tibetan Plateau, Including the World

Meteorological Organization (WMO) Number, Station Name, Latitude, Longitude, Elevation, and Data Missing Period During 1961–2005

WMO Number Station Name North Latitude East Longitude Elevation (m) Data Missing Period

52787 Wuqiaoling 37�120 102�520 3045.156080 Hezuo 35� 102�540 291056093 Minxian 34�260 104�010 231551886 Mangya 38�150 90�510 2944.852602 Lenghu 38�450 93�20 277052633 Tuole 38�480 98�250 336752645 Yeniugou 38�250 99�350 332052657 Qilian 38�110 100�150 2787.452707 Xiaozaohuo 36�480 93�410 2767 Apr–Dec 197452713 Dachaidan 37�510 95�220 3173.252737 Delingha 37�220 97�220 2981.552754 Gangcha 37�20 100�080 3301.552765 Menyuan 37�230 101�370 285052818 Germu 36�250 94�540 2807.652825 Nuomuhong 36�260 96�250 2790.452836 Dulan 36�180 98�060 3191.152856 Gonghe 36�160 100�370 283552866 Xinning 36�430 101�450 2295.252868 Guide 36�020 101�260 2237.152908 Wudaoliang 35�130 93�050 4612.252943 Xinghai 35�350 99�590 3323.256004 Tuotuohe 34�130 92�260 4533.156018 Zaduo 32�540 95�180 4066.456021 Qumalai 34�080 95�470 4175 Aug–Dec 196256029 Yushu 33�010 97�010 3681.256033 Maduo 34�550 98�130 4272.356034 Qingshuihe 33�480 97�080 4415.456046 Dari 33�450 99�390 3967.556067 Jiuzhi 33�260 101�290 3628.5 Apr–May 196256125 Nangqian 32�120 96�290 3643.756151 Banma 32�560 100�450 3530 Apr 1962–196555279 Bange 31�230 90�010 4700 Apr 196555299 Naqu 31�290 92�040 450755472 Shenzha 30�570 88�380 467255578 Rikeze 29�150 88�530 383655591 Lhasa 29�40 91�080 3648.7 Jun–Oct 196855598 Zedang 29�150 91�460 3551.755664 Dingri 28�380 87�050 4300 Nov 1968–1969, Aug 1969–Sep 197055680 Jiangzi 28�550 89�360 404055696 Longzi 28�250 92�280 386055773 Pali 27�440 89�050 430056106 Suoxian 31�530 93�470 4022.856116 Dingqing 31�250 95�360 3873.1 Jun–Aug 196956137 Changdu 31�090 97�10 330656227 Bomi 29�520 95�460 273656312 Linzi 29�40 94�20 2991.856038 Shiqu 32�590 98�060 420056079 Ruoergai 33�350 102�580 3439.656144 Dege 31�480 98�350 318456146 Ganzi 31�370 100� 3393.556152 Seda 32�170 100�20 3893.956167 Daofu 30�590 101�070 2957.256172 Maerkang 31�540 102�140 2664.456173 Hongyuan 32�480 102�330 3491.656178 Xiaojing 31� 102�210 2369.256182 Songpan 32�390 103�340 2850.756247 Batang 30� 99�060 2589.2 May–Dec 196856251 Xinlong 30�560 100�190 300056257 Litang 30� 100�160 3948.9 Sep 1967, Jan–Jul 1968, May–Aug 196956357 Daocheng 29�030 100�180 3727.7 May 196856374 Kangding 30�030 101�580 2615.756385 Emeishan 29�310 103�20 3047.456459 Muli 27�560 101�160 2426.556462 Jiulong 29� 101�30 2987.356479 Zhaojue 28� 102�510 2132.456565 Yanyuan 27�260 101�310 254556444 Deqin 28�290 98�550 331956543 Zhongdian 27�50 99�420 3276.156548 Weixi 27�10 99�170 2325.656651 Lijiang 26�520 100�130 2392.456684 Huize 26�250 103�170 2109.5


3 of 17

D07101

73�1805200 to 104�4605900E, and is distributed in 6 provinces,namely, the Tibet Autonomous Region, the Qinghai Prov-ince, Yunnan Province, Sichuan Province, Gansu Provinceand Xinjiang Uigur Autonomous Region [Y. Zhang et al.,2002]. There are 156 stations in the original data. A total of124 stations maintain daily data since 1961, of thesestations, 38 stations are excluded owing to the elevationbelow 2000 m asl, then 12 stations are also excluded owingto problems in data (10 stations stopped operation duringthe 1980s—1990s; 2 stations showed abnormity due to thediscontinuity of data record). The distribution of the stationsis uneven and very sparse in the western TP, which mayinfluence the regional trends. Therefore 3 stations in thewestern TP are also excluded. The remaining 71 stationswith elevation above 2000 m asl were selected, whichobservation started no later than 1961 and they were locatedin the eastern and central TP and in very close places whichare not administratively in the region, but are relevant to thisstudy (Figure 1).[7] Stations are identified by their World Meteorological

Organization (WMO) number and the stations name, alongwith longitude, latitude, elevation and missing period during1961–2005 (Table 1). Most meteorological stations wereestablished during the 1950s, and the selected 71 stationsare located with station altitudes varying between 2109.5 m(56684-Huize) and 4700 m (55279-Bange). Fourteen sta-tions are located above 4000 m and nine stations aresituated between 2000 m and 2500 m (Figure 2). In orderto obtain comparable time series we select data onlycovering the period of 1961 to 2005, excluding the spares

data available for earlier periods. The stations with shorterrecords are not selected in this study but they are stillavailable for assessing data quality and homogeneity atnearby stations.[8] Data quality control is a necessary step before the

calculation of indices because erroneous outliers can seri-ously impact the indices calculation and their trends. Dataquality control and calculation of the indices are performedusing the computer program RClimDex, which is developedand maintained by Xuebin Zhang and Feng Yang at theClimate Research Branch of Meteorological Service ofCanada. Software and documentation are available onlinefor downloading (http://cccma.seos.uvic.ca/ETCCDI/). Thesoftware identifies on a first run erroneous temperature andprecipitation data, such as precipitation values below 0 mmor days with Tmax< Tmin. Additional execution, identifypotential outliers, which have to be manually checked,validated, corrected or removed. For temperature, they aredefined as values outlying a user-defined threshold deter-mined by mean plus/minus a number of standard deviations.In our case, we choose 3 standard deviations as the thresh-olds for a finer quality control of the data. Both forprecipitation and temperature, data plots are available forvisual inspections to reveal more outliers as well as a varietyof problems that cause changes in the seasonal cycle orvariance of the data. Also, histograms of the data are createdwhich reveal problems that show up when looking at thedata set as a whole [Aguilar et al., 2005; New et al., 2006].Figure 3 is an example of the plots used to quality controlprecipitation data. It explains the data density in twodifferent ways: a histogram (bars) and a Kernel-filtered(line) which is a nonparametric approach to density fitting[Aguilar et al., 2005]. Both show that precipitation data inthe station is fine.[9] Homogeneity assessment and adjustment can be quite

complex and it often requires close neighbor stations,detailed station history and a great amount of time [Vincentet al., 2005]. Data homogeneity is assessed using the RHtest

Figure 2. Number of selected stations with observationstart year (top) and number of selected stations above thecategorized elevation (bottom) in the eastern and central TP.

Figure 3. Example of precipitation successful qualitycontrol procedures using RClimdex. Histogram (verticalbars) and Kernel-filtered density (line).


4 of 17

D07101

software (available from the ETCCDI Web site), which usesa two-phase regression model to check for multiple step-change points that could exits in a time series [Wang, 2003;Wang and Zhou, 2005]. The two-phase regression modelwas applied to annual mean daily maximum, minimumtemperature as well as daily temperature range, to identifypotential inhomogeneities in the data [X. Zhang et al.,2005]. Once a possible step change is identified in theannual series, it is also checked against the station history.There are 13 stations with a potential step in annualmaximum temperature and 8 stations with a potential stepin annual minimum temperature. Historical explanations forthe cause of the step, such as the relocation, are found foronly two stations. Therefore, we removed them from ourfinal data set. Figure 4 shows an example where a stepchange has been detected in the Tibet Autonomous Region.The station shows a large inhomogeneity in 1983, corrob-

orated by the station history, which shows a piece ofmetadata saying that it relocated that year.[10] After data quality control and homogeneity assess-

ment, RClimDex is used to calculate climate indices fromthe daily data. Expert Team for Climate Change Detectionand Indices (ETCCDI) has been coordinating a suite of 11precipitation and 16 temperature indices. For percentileindices a bootstrap procedure has been implemented toensure that the percentile-based temperature indices donot have artificial jumps at the boundaries of the in-baseand out-of-base period [X. Zhang et al., 2004]. Some ofthe indices, such as the number of tropical nights, thenumber of warm or cold duration and so on, are notrelevant to the studied region and have not been used,leading to a final selection of 12 temperature indices and9 precipitation indices (Table 2). They have been calcu-lated over the quality controlled data of the stations that

Figure 4. Homogeneity assessment results for annual mean daily minimum (top) and maximum(bottom) temperature for Jiali station (30�400, 93�170, 4488.8 m above sea level). The largest, statisticallysignificant discontinuity around 1983 is verified by the original station data, which indicate that thestation relocated in 1983.


5 of 17

D07101

passed the homogeneity assessment. Table 2 providestheir descriptions.[11] The selected indices are calculated on a monthly

and/or annual basis. Some indices are based on thresholddefined as percentiles. The percentiles are calculated fromthe reference period 1961–1990, which is a climatenormal period widely used. Monthly indices are obtainedif no more than 3 days are missing in a month and annualvalues are calculated if no more than 15 days are missingin a year. Threshold indices are computed if at least 70%of the data are present in the reference period.[12] In this study, linear trends for indices are calculated

using a nonparametric approach, a Kendall’s t-based Sen’srobust slope estimator [Sen, 1968], which is adapted andapplied in a study of annual temperature and precipitationchange over Canada [ X Zhang et al., 2000], and in extremewave heights over Northern Hemisphere oceans [Wang andSwail, 2001]. Annual missing value is excluded from theanalysis when calculating the linear trend. The 95%confidence intervals are calculated from tabulated values[Kendall, 1955]. The significance of the trends is deter-mined using an iterative procedure [Wang and Swail, 2001;X. Zhang et al., 2000] to compute the trends and to test thetrends significance taking account of a lag-1 autocorrelationeffect. For the eastern and central TP as a whole, the regionalseries are calculated by averaging anomalies relative to1961–2005. In order to avoid average series being domi-nated by those stations with high precipitation, regionalseries for precipitation indices are calculated again: stan-dardizing the simple anomaly through dividing by thestation standard deviation during the studied period. Allthe regional series are converted into trends per decade whendescribing linear regression trends. A trend is considered tobe statistically significant if it is significant at the 5% level.

3. Results

[13] The analysis of temperature and precipitation reveala variety of changes in extreme values during 1961–2005 inthe eastern and central TP. Spatial patterns of trends intemperature extremes have a much higher degree of coher-ence while precipitation in the region has more variability.The results for indices in climate extremes are describedalong this section.

3.1. Temperature

3.1.1. Cold Extremes (TX10, TN10, TXn, TNn, FD, ID)[14] Figure 5 shows the spatial distribution pattern of the

temporal trends in cold extremes for the 71 meteorologicalstations and Figure 6 demonstrates the regional annualanomalies series for indices of cold extremes in the easternand central TP. The regional trends in indices of coldextremes are in Table 3. Table 4 shows the number ofstations with significant negative, nonsignificant, and sig-nificant positive trends for cold extremes indices during1961–2005. For cold nights (TN10) and cold days (TX10),about 77% and 44% of stations have decreasing trends thatare statistically significant. Stations in the northern andsouthwestern TP, especially around the Qaidam Basin, havelarger trend magnitudes, while there are still a few stationsthat have increasing trends for cold days (TX10) and occurmainly in the southeastern TP. The cold days (TX10) havefluctuant variations before the mid-1980s and decreaseannually after that, but the cold nights (TN10) have contin-ually decreasing trends during the period of 1961–2005.The regional trends (in percentage of days) for these twoindices are �2.38 and �0.85 d/decade, respectively.[15] Similarly the temperatures of coldest days and cold-

est nights in each year (TXn and TNn) show increasingtrends at approximately 80–90% of stations. But only 21%

Table 2. Definitions of 12 Temperature Indices and 9 Precipitation Indices Used in This Studya

Index Descriptive Name Definition Units

TemperatureTXx warmest day annual highest TX �CTNx warmest night annual highest TN �CTXn coldest day annual lowest TX �CTNn coldest night annual lowest TN �CTN10 cold night frequency percentage of days when TN < 10th percentile of 1961–1990 %TX10 cold day frequency percentage of days when TX < 10th percentile of 1961–1990 %TN90 warm night frequency percentage of days when TN > 90th percentile of 1961–1990 %TX90 warm day frequency percentage of days when TX > 90th percentile of 1961–1990 %DTR diurnal temperature range annual mean difference between TX and TN �CID ice days annual count when TX < 0�C dFD frost days annual count when TN < 0�C dGSL growing season length annual count between first span of at least 6 days with TG > 5�C after winter

and first span after summer of 6 days with TG < 5�Cd

PrecipitationPRCPTOT wet day precipitation annual total precipitation from wet days mmSDII simple daily intensity index average precipitation on wet days mm/dCDD consecutive dry days maximum number of consecutive dry days dCWD consecutive wet days maximum number of consecutive wet days dR10mm number of heavy precipitation days annual count of days when RR � 10 mm dR95 very wet day precipitation annual total precipitation when RR > 95th percentile of 1961–1990 daily precipitation mmR99 extremely wet day precipitation annual total precipitation when RR > 99th percentile of 1961–1990 daily precipitation mmRX1day maximum 1-day precipitation annual maximum 1-day precipitation mmRX5day maximum 5-day precipitation annual maximum consecutive 5-day precipitation mm

aAll the indices are calculated by RClimDEX. Abbreviations are as follows: TX, daily maximum temperature; TN, daily minimum temperature;TG, daily mean temperature; RR, daily precipitation. A wet day is defined when RR � 1 mm, and a dry day is defined when RR < 1 mm. Indices areincluded for completeness but are not analyzed further in this article.


6 of 17

D07101

and 56% of stations for these two indices have statisticallysignificant trend due to the higher variance of this index.The stations presenting larger trend magnitudes are alsosituated in the northern and southeastern TP for these twoindices, while there are a few stations with decreasing trendsin the middle and southeastern TP only for TXn. It can beseen that TXn has a slight decreasing trend from 1961 to1980 then turns to increasing trend after 1980, while TNnhas a clear decreasing trend during 1961–2005, the regionaltrends in TXn and TNn are 0.30 and 0.69�C/decade,respectively, which is compatible with decreasing trend inTX10 and TN10.[16] The number of ice days (ID) has decreased at a rate

of �2.46 d/decade. The overall trend is significant at the0.001 level and intensifies after 1990. Around 30% ofstations have a statistically significant decreasing trendmainly occurring in Qinghai Province. In the southern TP,

the number of ice days is very little because of the lowlatitude, resulting to the feeble trend magnitudes comparedwith the high latitude. Frost days (FD) has also generallydecreased over the analysis period, at a regional rate of�4.32 d/decade, significant at the 0.001 level. About 39%of stations show a statistically significant decreasing trendand stations with larger trend magnitudes are distributed inthe southern and northwestern TP.[17] Table 5 shows the proportion of stations where trends

in indices are of a particular relative magnitude. About 85%of stations show larger trend magnitudes in TN10 thanTX10. For TXn and TNn, 75% of stations have greatertrend magnitudes in TNn. About 70% of stations showlarger trend magnitudes in FD than ID.3.1.2. Diurnal Temperature Range (DTR)[18] Many previous studies in the TP [Du, 2001; Duan

et al., 2006; X. Liu et al., 2006] show that the maximum

Figure 5. Spatial patterns of trends per decade for indices of cold extremes. Positive trends are shownas solid dots, negative trends as open dots. The size of the dot is proportional to the magnitude of thetrends.


7 of 17

D07101

Figure 6. Regional annual anomalies series relative to 1961–2005 for indices of cold extremes. Thesmoother line is the 9-year smoothing average.


8 of 17

D07101

and minimum temperatures both have increasing trends,but minimum temperature increases more rapidly thanmaximum temperature in the recent decades. Larger trendsin minimum temperature than maximum temperature

should bring declining trends in DTR. The regional trendis �0.20�C/decade with significant at the 0.001 level(Table 3), which drastically declines from 1961 to 1980. Incontrast, there are comparable increases in minimum andmaximum temperature since the 1980s, altering recent DTRtrend (Figure 7). Compared with the global, the tendency ofdecrease in the study conforms to it but the rate of decline ismuch higher [Vose et al., 2005].[19] Approximately 83% (44% statistically significant) of

stations show a decrease in DTR (Table 4), the largest DTRdiminished areas, such as the northwestern and southeasternTP, are in accordance with the areas of the strongestwarming (Figure 8).3.1.3. Warm Extremes (TX90, TN90, TXx, TNx, GSL)[20] For warm extremes indices during 1961–2005 in the

eastern and central TP, spatial distribution of temporaltrends and the regional anomalies series are shown inFigures 9 and 10. The regional trends and number ofstations with significant negative, nonsignificant, and sig-nificant positive trends are also listed in Tables 3 and 4.[21] For the percentage of days exceeding the 90th

percentiles (TX90 and TN90), about 49% and 85% ofstations show statistically significant increasing trends,respectively. Areas in the northern and southwestern TPhave larger trend magnitudes, while a few stations havedecreasing trends for warm days (TX90) and mainly occurin the west of Sichuan Province. The regional trends forthese two indices are 1.26 and 2.54 d/decade, respectively.About 39% and 51% of stations have statistically significantincreasing trends for extreme temperatures (TXx and TNx),which areas of larger trend magnitudes are accordance withTX90 and TN90. The regional trends for these two indicesalso show statistically increasing trends with the rate equiv-alent to 0.28 and 0.25�C/decade, respectively. For TX90and TN90, about 77% of stations have greater magnitude inTN90, and approximately 61% of stations have greatermagnitude in TXx than in TNx. (Table 5).[22] The regional trend for growing season length (GSL)

is 4.25 d/decade, although the GSL is not monotonic, as theincreasing trend observed during the 1960s and the 1980sonward, was reversed during the 1970s. About 39% ofstations show statistically significant increasing trends, withlarger values at the stations in the northern, southwestern

Table 3. Trends Per Decade for Regional Indices of Temperature

and Precipitation Extremesa

Index Units 1961–2005

TemperatureTN10 d/decade �2.38 (�2.85 to �1.91)TX10 d/decade �0.85 (�1.35 to �0.37)TN90 d/decade 2.54 (1.84–3.11)TX90 d/decade 1.26 (0.62–1.87)DTR �C/decade �0.20 (�0.26 to –0.14)TNn �C/decade 0.69 (0.51–0.87)TNx �C/decade 0.25 (0.17–0.34)TXn �C/decade 0.30 (0.06–0.53)TXx �C/decade 0.28 (0.12–0.42)FD d/decade �4.32 (�5.53 to �3.28)GSL d/decade 4.25 (2.87–5.69)ID d/decade �2.46 (�3.45 to �1.23)

PrecipitationPRCPTOT mm/decade 6.66 (�0.08–12.54)SDII mm/decade 0.03 (�0.01–0.07)RX1day mm/decade 0.27 (�0.03–0.63)RX5day mm/decade �0.08 (�0.92–0.70)R10mm d/decade 0.23 (�0.02–0.50)R95 mm/decade 1.28 (�1.55–4.15)R99 mm/decade 1.09 (�0.30–2.33)CDD d/decade �4.64 (�7.21 to �2.33)CWD d/decade �0.07 (�0.22–0.08)R95/RR %/decade 0.15 (�0.26–0.56)R99/RR %/decade 0.17 (0–0.41)

aParentheses are 95% confidence intervals. Values for trends significantat the 5% level (t test) are set in bold. The bottom rows give the trends forthe ratios R95/RR and R99/RR.

Table 4. Number of Stations With Significant Negative, Nonsigni-

ficant, and Significant Positive Trends for the Annual Temperature

and Precipitation Indices During 1961–2005a

Index Negative Nonsignificant Positive

TemperatureTN10 55 (70) 16 0 (1)TX10 31 (63) 39 1 (8)TN90 0 (1) 11 60 (70)TX90 1 (6) 35 35 (65)DTR 31 (59) 39 1 (12)TNn 0 (5) 31 40 (66)TNx 0 (4) 35 36 (67)TXn 0 (14) 56 15 (56)TXx 1 (11) 42 28 (59)FD 52 (70) 19 0 (1)GSL 0 (4) 43 28 (67)ID 21 (62) 50 0 (3)

PrecipitationPRCPTOT 1 (19) 63 7 (52)SDII 0 (26) 68 3 (44)RX1day 1 (29) 67 3 (41)RX5day 2 (34) 68 1 (36)R10mm 1 (20) 64 6 (49)R95 0 (24) 67 4 (46)R99 0 (23) 69 2 (48)CDD 13 (62) 57 1 (8)CWD 2 (39) 68 1 (31)

aSignificant at the 0.05 level. Numbers of stations with negative andpositive are also known in parentheses.

Table 5. Number and Proportion of Individual Stations Where the

Trend in One Index Is of Greater Magnitude Than the Trend in a

Seconda

Index Comparison Number Proportion

TX90 > TX10 abs 56 0.79TN90 > TN10 abs 41 0.58TXx > TXn rel 34 0.48TNx > TNn rel 11 0.15TXx > TNx rel 43 0.61TXn > TNn rel 18 0.25ID > FD abs 21 0.30TX90 > TN90 abs 16 0.23TX90 > TN10 abs 20 0.28TX10 > TN10 abs 11 0.15TN90 > TX10 abs 64 0.90

aAbbreviations are as follows: abs, indicates that the absolute magnitudesof trends are compared; rel, indicates that the signs of trends are retainedduring comparison.


9 of 17

D07101

and southeastern TP. This spatial pattern is similar to theobserved in other warm indices.3.1.4. Comparison of Warm and Cold Extremes[23] In order to learn more about the relative changes in

the daily temperature distribution, it is necessary to comparetrends in warm and cold indices. Comparison betweenwarm and cold extremes is shown in Table 5.[24] For TX90 and TX10, about 79% of stations have

larger trend magnitudes in TX90 than in TX10, and theregional trend in TX90 is more than 1.5 times that ofTX10. For minimum temperature, the regional trend inTN90 (2.54 d/decade) is of greater magnitude than that ofTN10 (�2.38 d/decade), but the difference is not as markedas for maximum temperature. When looking at individualstations a greater proportion (58%) of stations have highertrend magnitudes in TN90 than in TN10.[25] For TXx and TXn, regional trend in TXn is higher

than in TXx (0.30 and 0.28�C/decade, respectively), androughly half of the stations show larger trends in TXn.The magnitude of the regional trend in TNn is morethan 2.8 time that of TNx. At individual stations, about85% of stations have greater trend magnitudes in TNn.Therefore, we can conclude that changes in some warmextremes (TN90 and TX90) seem to be larger thanchanges in some cold extremes (TN10 and TX10), whilesome warm extremes (TNx and TXx) seem to havesmaller trend magnitudes than that in some cold extremes(TNn and TXn).

3.2. Precipitation

[26] In contrast to the temperature extremes, the signifi-cance of changes in precipitation extremes during 1961–2005 is low as trends are difficult to detect against the largerinterannual and decadal-scale variability of precipitation in

the eastern and central TP. The spatial distribution oftemporal trends and the regional annual standardizedanomalies series of precipitation indices are shown in

Figure 7. Same as Figure 6, but for trends in diurnal temperature range.

Figure 8. Same as Figure 5, but for trends in diurnaltemperature range.


10 of 17

D07101

Figures 11 and 12. The regional trends for precipitationindices are also listed in Table 3.[27] For the eastern and central TP as a whole, annual

total precipitation (PRCPTOT) shows positive correlationswith precipitation indices except consecutive dry days(CDD) (Table 6). When looking at regional trends, onlyconsecutive dry days (CDD) have statistically significanttrend (�4.64 d/decade). Two other indices show nonsignif-icant decreasing trends: maximum 5-day precipitation(RX5day) and consecutive wet days (CWD) (Table 3).For these three indices, 87%, 49% and 55% of stationshave decreasing trends and most of them are located in thecentral TP.[28] PRCPTOT shows larger trend magnitudes and the

regional increasing trend is 6.66 mm/decade with significantat the 0.1 level, with a decreasing trend in the 1960s andincreases slightly since the 1970s. About 73% of stationshave increasing trends mostly occurring in the southern TP

and the north of Qinghai Province while 27% of stationshave decreasing trends located in the central TP.[29] About 69% of stations for heavy precipitation days

(R10mm) have increasing trends which the distributions aresimilar to PRCPTOT. About 87% of stations for CDD havedecreasing trends and stations in the northeastern andsouthwestern TP have larger trend magnitudes.[30] In additions, maximum 1-day precipitation (RX1day),

average wet days precipitation (SDII), total precipitation onextreme wet days (R95 and R99) show nonsignificantincreasing regional trends during the period. The proportionof stations with positive trends for these indices is 58%,62%, 65% and 68%, respectively. These indices almosthave the similar distributions that stations located in north-ern, southeastern and southwestern TP show a pattern ofincreasing trends.[31] Figure 13 presents the regional series of the ratio

between the precipitation amount on very (extremely) wet

Figure 9. Same as Figure 5, but for warm extreme indices.


11 of 17

D07101

Figure 10. Same as Figure 6, but for warm extreme indices.


12 of 17

D07101

Figure 11. Same as Figure 5, but for trends in precipitation indices.


13 of 17

D07101

Figure 12. Regional annual standardized anomalies series relative to 1961–2005 for precipitationindices. The smoother line is the 9-year smoothing average.


14 of 17

D07101

days and total precipitation. The contribution of very wetdays (above the 95th percentile) to total amounts variesbetween 15%—24% and increases slightly over time. Thetrend in this ratio is 0.15%/decade (not significant at the 5%level). The contribution of extremely wet days (above the99th percentile) to the total amounts varies between 4%—9%, and the trend (0.17%/decade) is not significant at the5% level.

4. Discussion and Conclusions

[32] With the help of a set of widely spread descriptiveindices, a better understanding of observed change intemperature and precipitation extremes is gained for theeastern and central TP during 1961–2005. For most sta-tions, statistically significant increases in the percentage ofwarm nights/days and decrease in the percentage of cold

nights/days are observed during the period 1961–2005, thetrend magnitudes in cold/warm nights are larger than thosein cold/warm days. Therefore, the (daytime) trends inmaximum temperature extremes are smaller than the (night-time) trends in minimum temperature extremes, which canbe in line with the observed decrease in the DTR. Thewarming climate cause the number of the ice days and frostdays to decrease significantly and the number of growingseason length to increase significantly. These results gener-ally agree with what has been observed in the world duringthe second half of the 20th century [Alexander et al., 2006;Frich et al., 2002]. For temperature extremes, the annualhighest/lowest minimum temperature and maximum tem-perature also has statistically significant increasing trend,the magnitude in lowest of minimum temperature showsgreater change. These temperature indices show spatiallyuniform patterns, even though the climate varies across the

Table 6. The Correlation Coefficients of Precipitation Indices (n = 45, when r = ±0.29, P = 0.05)

CDD CWD PRCPTOT R10mm R95 R99 RX1day RX5day SDII

CDD 1.00CWD 0.03 1.00PRCPTOT �0.26 0.19 1.00R10mm �0.23 0.13 0.94 1.00R95 �0.11 0.01 0.78 0.74 1.00R99 �0.15 0.04 0.65 0.62 0.85 1.00RX1day �0.04 �0.12 0.50 0.39 0.75 0.78 1.00RX5day �0.11 0.17 0.53 0.48 0.68 0.70 0.70 1.00SDII �0.10 �0.05 0.65 0.76 0.78 0.66 0.52 0.56 1.00

Figure 13. Regional series (a) for the ratio between the index of precipitation falling on very wet days(R95) and total precipitation and (b) for the ratio between the index of precipitation falling on extremelywet days (R99) and total precipitation. The smoother line is as in Figure 6.


15 of 17

D07101

region. Stations in the northwestern, southwestern andsoutheastern TP have larger trend magnitudes, in accor-dance with the average warming in the regions.[33] The TP has experienced statistically significant

warming since the mid-1950s, and the linear rate of theannual mean temperature during the period 1955–1996 isabout 0.16�C/decade [X Liu and Chen, 2000]. Most prob-ably, there are two main reasons accounting for the surfacewarming in the TP. One is that the increasing anthropogenicgreenhouse gases emissions contribute to the recent warm-ing in the TP [Duan et al., 2006], and a model study alsotestifies that enhanced climatic warming in the TP due todoubling carbon dioxide [Chen et al., 2003]. The other isthe change of cloud amount. The low-level cloud amount inthe TP exhibits a significant increasing trend during thenighttimes, leading to the strong nocturnal surface warming,and both the total and low-level cloud amounts duringdaytime display decreasing trends, resulting in surfacewarming [Duan and Wu, 2006]. In the context of unprec-edented global warming, temperature extremes show re-gional trends that agree with the average warming in theregion, and changes in temperature extremes can be used forclimate change. Analyzing the characteristics of the regionaltime series in the TP, it can be found that in the mid-1980s,the TP experiences a climatic jump [Niu et al., 2004], whichalso reflects in some temperature extremes. It is found thatgrain production in Qinghai Province exhibits strong corre-lations with the temperature, and the tree growth in SichuanProvince is closely related to the change in temperature [ X.Liu et al., 2006]. However, the influences of temperatureextremes on the ecosystems are not discussed. Furtherworks should be done to assess the aspects.[34] Compared with change in temperature, there is no

agreement yet for precipitation change in the TP whichmainly contribute to the complex terrain and sparsemeteorological stations [Du and Ma, 2004; Li and Kang,2006; Wu et al., 2007]. Some experts divide the TP intonine subregions in terms of precipitation variation regimesand find that some subregions became drier but otherswetter [Lin and Zhao, 1996]. Precipitation in the TPmostly happens in the summer monsoon season, and thesummer precipitation in the TP is closely associated withthe North Atlantic Oscillation (NAO). During the summerof low NAO index values, summer precipitation is usuallyabove normal in the southern TP but below normal in thenorthern TP, and vice versa [X. Liu and Yin, 2001]. Duringthe summer monsoon season, precipitation in the southernTP is influenced by the monsoon strength, while in thenorthern of TP, air masses from the Atlantic Ocean bringmoisture to the TP, the dividing line which separates theregions influenced by different air masses is locatedaround 34–35�N in the central TP [Yeh and Gao, 1979],which is in accordance with the results derived from theisotope of precipitation [Tian et al., 2007; Yeh and Gao,1979].[35] Changes in precipitation extremes could be detected

for the variation of precipitation, although a small fractionof station trends are statistically significant for indices.There is a consistent pattern of trends in precipitationindices and the majority of precipitation indices are corre-lated with the annual total precipitation. Most precipitationindices exhibit increasing trends in the southern and north-

ern TP and show decreasing trends in the central TP, whichis located around 34–35�N. These suggest that the changeof precipitation indices is connected with the summermonsoon and westerly, the latter is associated with theNAO. Despite the spatial and temporal variations ofprecipitation indices in the eastern and central TP havebeen examined, much work remains to be done in the future.

[36] Acknowledgments. This study is supported by the ‘‘TalentProject’’ of the Chinese Academy of Sciences, the National Natural ScienceFoundation of China (40771187, 40401054), the National Basic ResearchProgram of China (2005CB422004), and the Sixth Framework ProgramPriority (036952). The authors thank the National Climate Center, ChinaMeteorological Administration, for providing the historical climate data forthis study. We are very grateful to the two anonymous reviewers for theirconstructive comments and suggestions.

ReferencesAguilar, E., et al. (2005), Changes in precipitation and temperatureextremes in Central America and northern South America, 1961–2003,J. Geophys. Res., 110, D23107, doi:10.1029/2005JD006119.

Alexander, L. V., et al. (2006), Global observed changes in daily climateextremes of temperature and precipitation, J. Geophys. Res., 111,D05109, doi:10.1029/2005JD006290.

Chen, B., et al. (2003), Enhanced climatic warming in the Tibetan Plateaudue to doubling CO2: A model study, Clim. Dyn., 20, 401–413.

Du, J. (2001), Change of temperature in Tibetan Plateau from 1961–2000(in Chinese), Acta Geogr. Sin., 56, 690–698.

Du, J., and Y. C. Ma (2004), Climatic trend of rainfall over Tibetan Plateaufrom 1971–2000 (Chinese with English abstract), Acta Geogr. Sin., 59,375–382.

Duan, A. M., and G. X. Wu (2005), Role of the Tibetan Plateau thermalforcing in the summer climate patterns over subtropical Asia, Clim. Dyn.,24, 793–807.

Duan, A. M., and G. X. Wu (2006), Change of cloud amount and theclimate warming on the Tibetan Plateau, Geophys. Res. Lett., 33,L22704, doi:10.1029/2006GL027946.

Duan, A. M., et al. (2006), New proofs of the recent climate warming overthe Tibetan Plateau as a result of the increasing greenhouse gases emis-sions, Chin. Sci. Bull., 51, 1396–1400.

Easterling, D. R., et al. (2000), Climate extremes: Observations, modeling,and impacts, Science, 289, 2068–2074.

Frich, P., et al. (2002), Observed coherent changes in climatic extremesduring the second half of the twentieth century, Clim. Res., 19, 193–212.

Griffiths, G. M., et al. (2005), Change in mean temperature as a predictor ofextreme temperature change in the Asia-Pacific region, Int. J. Climatol.,25, 1301–1330.

Haylock, M. R., et al. (2006), Trends in total and extreme South Americanrainfall in 1960–2000 and links with sea surface temperature, J. Clim.,19, 1490–1512.

Intergovernmental Panel on Climate Change (2007), Summary for Policy-makers of Climate Change 2007: The Physical Science Basis. Contribu-tion of Working Group I to the Fourth Assessment Report of theIntergovernmental Panel on Climate Change, Cambridge Univ. Press,New York.

Kang, S., et al. (2007), Recent temperature increase recorded in anice core in the source region of Yangtze River, Chin. Sci. Bull., 52,825–831.

Katz, R. W., and B. G. Brown (1992), Extreme events in a changingclimate—Variability is more important than averages, Clim. Change,21, 289–302.

Kendall, M. R. (1955), Rank Correlation Methods, Charles Griffin, London.Klein Tank, A. M. G., et al. (2006), Changes in daily temperature andprecipitation extremes in central and south Asia, J. Geophys. Res., 111,D16105, doi:10.1029/2005JD006316.

Kunkel, K. E., et al. (1999), Temporal fluctuations in weather and climateextremes that cause economic and human health impacts: A review, Bull.Am. Meteorol. Soc., 80, 1077–1098.

Li, C. L., and S. C. Kang (2006), Review of the studies on climate changesince the last inter-glacial period on the Tibetan Plateau, J. Geogr. Sci.,16, 337–345.

Lin, Z. Y., and X. Y. Zhao (1996), Spatial characteristics of changes intemperature and precipitation of the Qinghai-Xizang (Tibet) plateau, Sci.China Ser. D, 39, 442–448.

Liu, B., M. Xu, M. Henderson, and Y. Qi (2005), Observed trends ofprecipitation amount, frequency, and intensity in China, 1960–2000,J. Geophys. Res., 110, D08103, doi:10.1029/2004JD004864.


16 of 17

D07101

Liu, X. D., and B. D. Chen (2000), Climatic warming in the Tibetan Plateauduring recent decades, Int. J. Climatol., 20, 1729–1742.

Liu, X. D., and Z. Y. Yin (2001), Spatial and temporal variation of summerprecipitation over the eastern Tibetan Plateau and the North Atlanticoscillation, J. Clim., 14, 2896–2909.

Liu, X., Z.-Y. Yin, X. Shao, and N. Qin (2006), Temporal trends andvariability of daily maximum and minimum, extreme temperatureevents, and growing season length over the eastern and central TibetanPlateau during 1961–2003, J. Geophys. Res., 111, D19109, doi:10.1029/2005JD006915.

Manton, M. J., et al. (2001), Trends in extreme daily rainfall and tempera-ture in Southeast Asia and the South Pacific: 1961–1998, Int. J. Clima-tol., 21, 269–284.

New, M., et al. (2006), Evidence of trends in daily climate extremes oversouthern and west Africa, J. Geophys. Res., 111, D14102, doi:10.1029/2005JD006289.

Niu, T., et al. (2004), The characteristics of climate change over the TibetanPlateau in the last 40 years and the detection of climatic jumps, Adv.Atmos. Sci., 21, 193–203.

Peterson, T. C., et al. (2002), Recent changes in climate extremes in theCaribbean region, J. Geophys. Res., 107(D21), 4601, doi:10.1029/2002JD002251.

Sen, P. K. (1968), Estimates of regression coefficient based on Kendall’stau, J. Am. Stat. Assoc., 63, 1379–1389.

Tian, L., T. Yao, K. MacClune, J. W. C. White, A. Schilla, B. Vaughn,R. Vachon, and K. Ichiyanagi (2007), Stable isotopic variations inwest China: A consideration of moisture sources, J. Geophys. Res.,112, D10112, doi:10.1029/2006JD007718.

Vincent, L. A., et al. (2005), Observed trends in indices of daily temperatureextremes in South America 1960–2000, J. Clim., 18, 5011–5023.

Vose, R. S., D. R. Easterling, and B. Gleason (2005), Maximum and mini-mum temperature trends for the globe: An update through 2004, Geo-phys. Res. Lett., 32, L23822, doi:10.1029/2005GL024379.

Wang, X. L. (2003), Comments on ‘‘Detection of undocumented change-points: A revision of the two-phase regression model,’’ J. Clim., 16,3383–3385.

Wang, X. L. L., and V. R. Swail (2001), Changes of extreme wave heightsin Northern Hemisphere oceans and related atmospheric circulationregimes, J. Clim., 14, 2204–2221.

Wang, Y. Q., and L. Zhou (2005), Observed trends in extreme precipita-tion events in China during 1961–2001 and the associated changes inlarge-scale circulation, Geophys. Res. Lett., 32, L09707, doi:10.1029/2005GL022574.

Wu, S. H., et al. (2007), Climatic trends over the Tibetan Plateau during1971–2000, J. Geogr. Sci., 17, 141–151.

Yanai, M. H., et al. (1992), Seasonal heating of the Tibetan Plateau and itseffects on the evolution of the Asian summer monsoon, J. Meteorol. Soc.Jpn., 70, 319–351.

Yeh, T. C., and Y. X. Gao (1979), Meteorology of the Qinghai-Xizang(Tibet) Plateau (in Chinese), Science Press, Beijing.

You, Q.-L., S. Kang, N. Pepin, and Y. Yan (2008), Relationship betweentrends in temperature extremes and elevation in the eastern and centralTibetan Plateau, 1961 – 2005, Geophys. Res. Lett., 35, L04704,doi:10.1029/2007GL032669.

Zhai, P. M., et al. (1999), Chances of climate extremes in China, Clim.Change, 42, 203–218.

Zhang, X. B., et al. (2000), Temperature and precipitation trends in Canadaduring the 20th century, Atmos. Ocean, 38, 395–429.

Zhang, X. B., et al. (2005), Trends in Middle East climate extreme indicesfrom 1950 to 2003, J. Geophys. Res., 110, D22104, doi:10.1029/2005JD006181.

Zhang, X. L., et al. (2004), Avoiding inhomogeneities in percentile-basedindices of temperature extremes, J. Clim., 18, 1641–1651.

Zhang, Y. L., et al. (2002), A discussion on the boundary and area of theTibetan Plateau in China (Chinese with English abstract), Geogr. Res., 21,1–8.

��E. Aguilar, Climate Change Research Group, Geography Department,

Universitat Rovira i Virgili de Tarragona, Plaza Imperial Tarraco, Tarragona143005, Spain.S. Kang and Q. You, Institute of Tibetan Plateau Research, Chinese

Academy of Sciences, Beijing 100085, China. ([email protected])Y. Yan, National Climate Center, Beijing 100081, China.


17 of 17

D07101

Changes in daily climate extremes in the eastern and ... · Tank et al., 2006]. Global changes in...

Documents

Transcript of Changes in daily climate extremes in the eastern and ... · Tank et al., 2006]. Global changes in...