| 摘要: |
| 基于1960—2017年黄土高原47个气象站逐日降水量数据，对黄土高原极端降水量、极端降水日数和极端降水强度的时空演化特征进行了分析。结果显示：（1）黄土高原极端降水时空分布存在明显异质性。极端降水事件在全年发生频率不高（13.2%—46.1%），但对年降水量贡献突出（50.4%—91.4%）。（2）1960—2017年黄土高原极端降水量和降水日数出现下降，但极端降水强度上升。在黄土高原干旱化的同时，极端降水在全年降水中的占比有升高趋势。作为黄土高原土壤侵蚀的主要驱动力，愈发增强的极端降水将会给该地区水土保持和地质灾害防治工作带来严峻挑战。（3）极端降水特征EOF分解第一模态反映了黄土高原极端降水的一致性变化，其空间分布特征可能受到了当地水汽条件及大气层结稳定度等大气动力效应的影响。第二模态反映了黄土高原极端降水的差异性变化，极端降水量和极端降水日数均大致以临夏—太原一线呈反相位变化，这种分布模式可能和地形因素密切相关。（4）黄土高原极端降水特征的年际变化受厄尔尼诺-南方涛动（ENSO）活动的显著影响，厄尔尼诺年极端降水量偏低、日数偏少，拉尼娜年反之。 |
| 关键词: 黄土高原 极端降水 EOF分解 ENSO |
| DOI：10.7515/JEE222066 |
| CSTR：32259.14.JEE222066 |
| 分类号: |
| 基金项目:国家杰出青年科学基金（42025304）；中国科学院战略性先导科技专项（XDB40000000） |
| 英文基金项目:National Science Fund for Distinguished Young Scholars (42025304); Strategic Priority Research Program of the Chinese Academy of Sciences (XDB40000000) |
|
| Spatio-temporal evolution of extreme precipitation in Loess Plateau during 1960—2017 and its response to atmospheric circulation changes |
|
WANG Guozhen, YAN Hong, LIU Chengcheng, HAN Tao, ZHAO Nanyu, YANG Haotian
|
1. State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, Chinese Academy of Sciences, Xi’an 710061, China
2. CAS Center for Excellence in Quaternary Science and Global Change, Xi’an 710061, China
3. Institute of Global Environmental Change, Xi’an Jiaotong University, Xi’an 710049, China
4. Xi’an Institute for Innovative Earth Environment Research, Xi’an 710061, China
5. University of Chinese Academy of Sciences, Beijing 100049, China
|
| Abstract: |
| Background, aim, and scope In addition to the variation in the mean climate state, extreme weather events are also important characteristics of the climate system. These are hard to predict and can have serious impacts on socio-economic conditions. IPCC6 demonstrated that the frequency and intensity of heavy precipitation events have increased in many areas since 1950s, under a global warming patterns. Eco-fragile areas with severe soil and water erosion in the world, such as the Chinese Loess Plateau (LP), are one such situation. Here, we investigated the spatio-temporal evolution of extreme precipitation events in the LP and discuss the main atmospheric circulation factors. Materials and methods Based on the daily precipitation data from 47 meteorological stations during 1960—2017 in the LP, the threshold values of extreme precipitation event were defined. We employ correlation analysis, inverse distance weighted interpolation (IDW), radial basic function interpolation (RBF), empirical orthogonal functions (EOF) and spectral analysis. Results The distribution of extreme precipitation amount and intensity are similar. Seasonally, the amount and intensity are high in the main monsoon season and low in other seasons. Spatially, both of these values are high in the southeast and low in the northwest. However, the distribution of extreme precipitation days across the LP seems to be spatially homogeneous in seasons except winter. The contribution of each season to annual extreme precipitation amount and days has a series with summer＞autumn＞spring＞winter. Annual extreme precipitation amount (−1.6 mm·(10a)−1) and days (−0.12 d·(10a)−1) gently decreased during 1960—2017. Simultaneously, the annual extreme precipitation intensity increased (0.05 mm·d−1·(10a)−1). None of these trends pass the 0.05 significance test. In the LP, 51.1% of the stations show a decreasing trend while 48.9% of the stations show an increasing trend in annual extreme precipitation amount. Decrease centers are located in Lintao, Huashan, and Yangcheng, with a rate of −21 mm·(10a)−1. Increase centers are located in Xining and Yulin, with a rate of 11 mm·(10a)−1. In most stations of the LP (about 60%), annual extreme precipitation days decreased. The decrease center is located in Lintao and Huashan, with a rate of about −0.6 d·(10a)−1. Notably, about 60% of the stations show an increasing trend in annual extreme precipitation intensity. Increase centers are located in Zhengzhou, Changwu and Yuanping, with a rate of about 0.5 mm·d−1·(10a)−1. The results of EOF indicate that the first EOFs of extreme precipitation elements (amount and days) have consistent variations in the LP, which may be dominated by the water vapor content and atmospheric dynamical effects (such as summer stratification). The second EOFs represent the regional difference of extreme precipitation, in which the tendency rates of extreme precipitation elements are reversibly distributed in the LP. This probably results from terrain factors. All corresponding PCs (principal component) of extreme precipitation have no significant trend at P=0.05 level. Discussion Extreme precipitation variations in the LP are closely related to regional moisture sources and atmospheric circulation. In monsoon season, the southeastern LP is affected by both the South Asian monsoon and the East Asian monsoon because of a strong Pacific high pressure. The warm moist flow from the Bay of Bengal and the South China Sea provides sufficient moisture for rainstorms. However, moisture on the LP is mainly transported by the prevailing westerlies in non-monsoon season. The different of moisture sources and transmission distance may lead to a gradient of extreme precipitation from the southeast to the northwest LP. The variation of extreme precipitation in the LP may be significantly affected by ENSO. Extreme precipitation in the LP is reduced in El Niño years. The SST of tropical Western Pacific is colder in El Niño years, which may cause a weakening in both the Walker cell and local Pacific Hadley cell. And then the weakening of radial atmospheric convection may lead to a mean south-westerly position of the Western Pacific subtropical high. The main rain and wind belt is thus southerly and it is not conducive to the occurrence and development of extreme precipitation on the LP. Conclusion (1) The spatial and temporal distributions of the extreme precipitation are almost heterogeneous. Although extreme precipitation events do not occur frequently throughout the year (13.2%—46.1%), but contribute significantly to annual precipitation (50.4%—91.4%). (2) Extreme precipitation amount and days decreased, but the intensity increased during 1960—2017 in the LP. The proportion of extreme precipitation in the annual precipitation increases significantly with aridity of the LP. (3) The EOF results of the extreme precipitation elements indicated that the first EOFs represent consistent variations in the LP, which may be influenced by the water vapor content and atmospheric dynamical effects (such as summer stratification). The second EOFs represent the regional difference of extreme precipitation, in which the tendency rates of extreme precipitation elements are reversibly distributed in the LP. These distribution could be a result of terrain factors. (4) Extreme precipitation on the LP is significantly affected by ENSO activities on an interannual scale. Recommendations and Perspectives Although multiple of studies have focused on the extreme precipitation on the LP, our understanding of its controlling factors is wanting. Obtaining extreme precipitation records under different climatic backgrounds is crucial for predicting the variation of extreme precipitation in the LP under an expected global warming. It is necessary to derive extreme precipitation information from paleoclimate and paleoweather archives to enhance understanding of future climate trajectories. |
| Key words: the Loess Plateau extreme precipitation EOF analysis ENSO |
