Changes in outdoor thermal sensation and sensitivity to climate factors in Beijing from 1960 to 2014
LIShuangshuang1,2, YANGSaini1,2, LIUXianfeng1,3, LIUYanxu4 1. State Key Laboratory of Earth Surface Processes and Resource Ecology,Beijing Normal University,Beijing 100875,China2. Academy of Disaster Reduction and Emergency Management,Beijing Normal University,Beijing 100875,China3. College of Resource Science and Technology,Beijing Normal University,Beijing 100875,China4. College of Urban and Environment Sciences,Peking University,Beijing 100871,China 收稿日期:2015-06-10 修回日期:2015-11-5 网络出版日期:2016-01-25 版权声明:2016《资源科学》编辑部《资源科学》编辑部 基金资助:基金项目:地表过程模型与模拟创新研究群体科学基金项目(413221001)国家重点基础研究发展计划项目(2012CB955402)北京市科委课题(Z151100002115040) 作者简介: -->作者简介:李双双,男,陕西潼关人,博士生,研究方向为全球变化与区域灾害防治。E-mail:lss40609010@126.com
关键词:气候变化;UTCI;户外感知温度;敏感系数;北京 Abstract :Using the Universal Thermal Climate Index(UTCI),we analysed climatic charac-teristics of human-perceived temperature in Beijing from 1960 to 2014. Dominant climatic factors driving variation in UTCI were quantitatively analysed using sensitivity analysis methods. We found that the average annual UTCI was 9.2℃ in Beijing over the last 55 years and that people felt comfortable on the whole. Regarding decadal change,the UTCI in Beijing underwent a decrease (1960-1970),increase (1971-1990)and high stable period (1991-2014). Regarding seasonal change,people feel comfortable in spring and autumn. Moderate heat stress and moderate cold stress were detected in summer and winter,respectively. A significant increase in the UTCI was detected beginning in the mid- to late 1980s; warming in spring and winter is more obvious than summer and autumn. The decrease in strong cold stress days was obvious in Beijing (-5.8d/decade)and variation in moderate cold stress days has increased over time (2.6d/decade). Linear trends in moderate heat stress and comfortable days were 1.0d/decade and 1.6d/decade,respectively. UTCI sensitivity to temperature seems to be strong in Beijing at annual and seasonal scales. However,major factors and contributions of climate factors showed considerable differences across different seasons. For example,changes in UTCI were related to increasing temperature and decreasing wind speed in winter,whereas increasing temperature and relative humidity were major driving forces in summer. These results and sensitivity analysis were validated under given conditions based on modification of the real data set.
1960-2014年北京居民户外感知温度年均值为9.2 ℃,人体感觉舒适。在季节上,春季和秋季UTCI分别为9.8℃和10.1℃,人体感觉亦为舒适;夏季UTCI值相对较高为26.8℃,人体感知为中度热胁迫;冬季UTCI值相对较低为-10.2℃,人体感知为中度冷胁迫。在变化趋势上,四季UTCI均呈现增加趋势,其变化速率分别为:夏季(0.09℃/10a)<秋季(0.41℃/10a)<全年(0.52℃/10a)<春季(0.60℃/10a)<冬季(1.08℃/10a),除夏季变化趋势不显著外,其他季节均通过0.01显著水平检验。由全年和四季变化趋势可知,近55年北京户外人体感知温度增加主要以冬春显著增加为主,尤其是20世纪80年代中期后,春季和冬季正距平年份比重明显增加(表2)。 Table 2 表2 表21960-2014年北京UTCI变化描述统计特征 Table 2The statistical characteristic of UTCI change in Beijing from 1960 to 2014
指标
春季
夏季
秋季
冬季
全年
年平均值/℃
9.8
26.8
10.1
-10.2
9.2
胁迫类型
舒适
中度热胁迫
舒适
中度冷胁迫
舒适
变化趋势/(℃/10a)
0.60
0.09
0.41℃/10a
1.08
0.52
显著水平R2
0.27
0.02
0.20
0.43
0.37
p值
p<0.01
不显著
p<0.01
p<0.01
p<0.01
变异系数
0.20
0.03
0.15
0.25
0.15
1985年之前正距平比重/%
23.1
38.5
42.3
15.4
23.1
1985年之后正距平比重/%
79.3
55.2
65.5
89.7
93.1
新窗口打开 1960-2014年北京人体感知温度表现出阶段性(图2)。从10年滑动平均曲线看:①春季,20世纪60-70年代UTCI先减少后增加,在70年代形成相对偏冷期,80年代后除各别年份为负距平外,UTCI整体处于相对偏暖期;②夏季,20世纪60-70年代中期UTCI为波动减少期,60年代中期-90年代初UTCI多处于0值以下,UTCI为相对偏冷期,90年代后UTCI呈增加趋势,正距平年相对70-80年代明显增加;③秋季,20世纪60年代偏暖期要短于夏季,60年代中期-80年中期为相对偏冷期,80年代中期-21世纪初UTCI呈现正负交替变化,2005年UTCI大幅增加,形成一个相对高值峰期;④冬季,20世纪60年代经历了短暂的偏暖期后大幅下降,在60年代-80年代中期形成一个相对偏冷期,80年代中期后UTCI快速上升后维持正距平波动,整个时期UTCI明显偏暖。 显示原图|下载原图ZIP|生成PPT 图21960-2014年北京全年和四季UTCI变化趋势 -->Figure 2The change trends of annual and seasonal UTCI in Beijing from 1960 to 2014 -->
3.2 不同等级UTCI变化特征
全球变暖背景下,北 京不同等级UTCI趋势变化具有差异性(图3),主要表现在以下几个方面:①1960-2014年极端热胁迫日数整体下降趋势并不显著(图3a),其变化可分为3个阶段:1960-1977年波动减少期,1978-1995年低位波动期,1996-2014年快速上升期;②1960-2014年中度热胁迫日数整体上升趋势虽不显著(p>0.05)(图3b),但是在20世纪70年代中期后快速增加,并在2002年后维持稳定波动;③就舒适日数而言(图3c),其变化可亦分为3个阶段,1960-1989年持续增加期,1990-2000年波动下降期,2000-2014年稳定波动期;④轻微冷胁迫变化虽有所增加但变化相对稳定(图3d),除20世纪60-70年代呈现先减少后增加趋势外,其余阶段均表现为震荡变化;⑤中度冷胁迫日数则表现出明显的增加趋势(图3e),上升速率为0.26d/10a,其阶段性变化与极端冷胁迫日数巧好相反,以1986年为界,前期多为负距平,后期多为正距平;⑥在冷胁迫日数方面,极端冷胁迫日数呈下降趋势(图3f),下降速率为-0.58d/10a,且以1986年为界,前期多为正距平,后期多为负距平。 显示原图|下载原图ZIP|生成PPT 图31960-2014年北京不同等级UTCI日数变化特征 -->Figure 3The characteristics of UTCI change on different levels in Beijing from 1960 to 2014 -->
Fröhlich等对卡塔尔多哈地区UTCI对气候因子敏感性进行定量分析[5],其方法以调节气候因子变幅为主,这与本文敏感性分析方法略有差异。在此,借鉴其研究方法,绘制4个关键气候因子绝对变化分布图,每一个“豆荚”宽度代表气温频率,黑色实线代表一个“豆荚”分布的均值(图4)。从图中可以看出,气温和风速调整对UTCI变化影响最为明显,相对湿度和辐射温度变化对UTCI变化的影响则相对较弱,这从另一个角度佐证了UTCI对气候因子敏感性结论。赵娜等通过北京12个台站气象观测资料,分析了快速城市化对北京区域气候的影响[28]。研究发现,近30年北京城区增温要快于周边地区,风速下降亦快于周边地区。快速城市化背景下,城市热岛效应和下垫面性质变化,直接影响气温和风速变化,可能是北京UTCI呈现快速上升的重要原因之一。 显示原图|下载原图ZIP|生成PPT Figure 4Distribution of UTCI calculated with modified climatic factors in Beijing(注:“——”为UTCI均值;图4d中shade指在阴凉处人体感知温度,Tmrt=Ta;sun指在阳光下人体感知温度;50%shade指50%辐射温度下人体感知温度,Tmrt′=0.5Ta+0.5Tmrt。) --> -->
在研究方法上,由于无法获得更多区域辐射观测数据,这就限制了从时空角度再认识北京地区UTCI变化特征。笔者尝试利用Penman-Monteith模型对北京地面总辐射进行估算,将实测数据与其相减,绘制两者偏差直方图,并对比模拟总辐射与实测总辐射计算UTCI结果的差异(图5)。从图中可以看到,实际总辐射与模拟总辐射差值主要分布于0.0~6.5MJ/(m2·d)(图5a),相应的UTCI差值变化范围0.0~1.2℃(图5b)。也就是说,模拟总辐射低于实际总辐射,会导致人体感知温度被低估。因此,如何准确模拟区域总辐射数据,是未来从“格局-过程-机制”角度认识感知温度时空变化的关键。 显示原图|下载原图ZIP|生成PPT 图5模拟总辐射和实际总辐射下日总辐射差值和UTCI差值分布特征 -->Figure 5Distribution of global radiation and UTCI difference between station and formula in Beijing -->
大量研究表明:不仅在气候平均态,而且在单次极端高温事件过程中,城市热岛对北京最低气温(夜间)影响要高于最高气温(白天),城市化增温作用具有季节和昼夜“非对称性”特征[29,30]。本文以平均气温作为环境变量,评估人体感知温度的趋势变化,并未区分昼夜感知温度差异,势必会低估气候变化对居民生活和健康的风险。Nicholls等分析了1979-2001年墨尔本地区高温与65岁以上老人死亡率的关系发现,当最低温度(夜间)超过24.0°C,老年人日死亡率增加19.0%~21.0%[31]。未来需要从昼夜异常角度,分析北京地区感知温度响应差异,定量气候变化和城市化对人体昼夜感知温度的相对作用。同时,细化UTCI时间分辨率,以日内(上班高峰期和上班时间、白昼和夜晚)、周内(工作日和周末)、月内(月初、中旬、月末)、特殊节假日(春节和国庆长假)为视角,探讨城市化对人体感知温度的影响。 The authors have declared that no competing interests exist.
[YANYechao,YUEShuping,LIUXuehua,et al.Advance in assessment of bioclimatic comfort conditions at home and abroad [J]. ,2013,28(10):1119-1125.] [本文引用: 1]
[2]
BlazejczykK,JendritzkyG,BrodeP,et al.An introduction to the universal thermal climate index (UTCI) [J]. ,2013,86(1):5-10. [本文引用: 1]
[3]
BlazejczykK,EpsteinY,JendritzkyG,et al.Comparison of UTCI to selected thermal indices [J]. ,2012,56(3):515-535. [本文引用: 1]
[4]
JendritzkyG,Dear RD,HavenithG.UTCI-Why another thermal index [J]. ,2012,56(3):421-428. [本文引用: 1]
[5]
FröhlichD,MatzarakisA.A quantitative sensitivity analysis on the behaviour of common thermal indices under hot and windy conditions in Doha,Qatar [J]. ,2015,doi:10.1007/s00704-015-1410-5. [本文引用: 3]
[6]
BrödeP,Krüger EL,Rossi FA,et al.Predicting urban outdoor thermal comfort by the Universal Thermal Climate Index UTCI-a case study in Southern Brazil [J]. ,2012,56(3):471-480. [本文引用: 1]
[7]
BletaA,Nastos PT,MatzarakisA.Assessment of bioclimatic conditions on Crete Island,Greece [J]. ,2014,14(5):1967-1981. [本文引用: 1]
[8]
PappenbergerF,JendritzkyG,StaigerH,et al.Global forecasting of thermal health hazards:The skill of probabilistic predictions of the Universal Thermal Climate Index (UTCI) [J]. ,2015,59(3):311-323. [本文引用: 1]
[9]
HöppeP.Different aspects of assessing indoor and outdoor thermal comfort [J]. ,2002,34(6):661-665. [本文引用: 1]
[10]
ParkS,Tuller SE,JoM.Application of Universal Thermal Climate Index (UTCI)for microclimatic analysis in urban thermal environments [J]. ,2014,125(2):146-155.
[11]
RuttyM,ScottD.Bioclimatic comfort and the thermal perceptions and preferences of beach tourists [J]. ,2015,59(3):37-45. [本文引用: 1]
[MALijun,SUNGennian,LILingfen,et al.Correlative analysis of climate comfort and monthly variation of tourists in Haikou city [J]. ,2008,30(11):1754-1759.] [本文引用: 1]
[MALijun,SUNGennian,XIEYuefa,et al.A study on variations of the tourism climate comfort degree in five typical cities in eastern China during the last 50 years [J]. ,2010,32(10):1963-1970.] [本文引用: 1]
[CAOWeihong,HEYuanqing,LIZongxing,et al.A correlation analysis between climate comfort degree and monthly variation of tourists in Lijiang [J]. ,2012,32(12):1459-1464.]
[WANG Shen. TIANHong,XIEWusan,et al.A study on variations and comprehensive zoning of climate comfort degree in Anhui province in the past 50 years [J]. ,2012,31(1):40-45.]
[ZHAOZijian,CHENJingyi,ZHONGXiuwen,et al.Nanjing thermal comfort status analysis based on standard effective temperature and discomfort index [J]. ,2013,36(4):16-21.]
[YUZhikang,SUNGennian,LUOZhengwen,et al.An analysis of climate comfort degree and tourism potential power of cities in northern China in summer to the north of 40ºN [J]. ,2015,30(2):327-339.]
[CHENHui,YANYechao,YUEShuping,et al.Types of summer comfortable climate in China and their temporal and spatial distribution [J]. ,2015,34(2):175-284.] [本文引用: 2]
[22]
FialaD,HavenithG,BrödeP,et al.UTCI-Fiala multi-node model of human heat transfer and temperature regulation [J]. ,2012,56(3):429-441.
[23]
HavenithG,FialaD,BłazejczykK,et al.The UTCI-clothing model [J]. ,2012,56(3):461-470. [本文引用: 1]
[24]
BrödeP,FialaD,BłażejczykK,et al.Deriving the operational procedure for the Universal Thermal Climate Index (UTCI) [J]. ,2012,56(3):481-494. [本文引用: 1]
[25]
McCuen RH. A sensitivity and error analysis of procedures used for estimating evaporation [J]. ,1974,10(3):486-498. [本文引用: 1]
[26]
Yin YH,Wu SH,Dai EF.Determining factors in potential evapotranspiration changes over China in the period 1971-2008 [J]. ,2010,55(29):3329-3337. [本文引用: 1]
[LIUXianfeng,PANYaozhong,ZHANGJinshui,et al.Spatiotemporal variation patterns of potential evapotranspiration in five provinces of north-west China in 1960-2011 [J]. ,2013,24(9):2564-2570.] [本文引用: 1]
[ZHOULi,JIANGZhihong,LIZhaoxin,et al.Numerical simulation of urbanization climate effects in regions of east China [J]. ,2015,39(3):596-610.] [本文引用: 1]
[31]
NichollsN,SkinnerC,LoughnanM,et al.A simple heat alert system for Melbourne,Australia [J]. ,2008,52(5):375-384. [本文引用: 1]