黄土高原昼夜不对称性增温及其对植被NDVI的影响

本文选取1982—2015年黄土高原及周边地区102个气象站点的气象观测资料,结合归一化植被指数（NDVI）和植被类型数据,采用Mann-Kendall（M-K）非参数检验法和二阶偏相关分析法,分析了黄土高原植被生长季日最高气温（T_max）和夜间最低气温（T_min）的时间序列变化趋势和空间格局、并分植被类型探讨昼夜不对称性增温对植被的影响。研究表明：①1982—2015年间黄土高原植被生长季T_max及T_min都存在极为显著的上升趋势（P<0.01）,植被生长季T_min的增温速率约为T_max增温速率的1.6倍,昼夜增温呈不对称性;②昼夜不对称性增温对植被动态的影响呈现明显的分异特征,植被生长季植被NDVI与T_min表现出比T_max更显著的相关性,即相对于白天增温,夜间增温的变化对黄土高原植被的影响程度更加显著;③不同植被类型的NDVI与昼夜增温的偏相关分析可知,除草原、农作物外其他植被类型均与T_max表现为显著偏正相关（P<0.05）;不同植被类型与T_min的偏相关关系差异明显,农作物与T_min表现为显著偏负相关（P<0.01）,其他植被类型与T_min均呈偏正相关性（P<0.05）。

Diurnal asymmetry of temperature and its effect on NDVI in Loess Plateau

In this paper, the meteorological observation data of 102 meteorological stations in the Loess Plateau and its surrounding areas in 1982-2015 were selected. The Normalized Difference Vegetation Index (NDVI) and vegetation type data were used. And the Mann-kendall (m-k) Nonparametric test method and the second-order partial correlation analysis were adopted. The time series change trend and spatial pattern in the growing seasonal minimum temperature (T_min) and maximum temperature (T_max) over the Loess Plateau between 1982 and 2015 were analyzed, and the effects of asymmetric warming on different vegetation type were discussed. The results show that: ① We found a persistent increase (P<0.01) in the growing seasonal T_min and T_max over the Loess Plateau between 1982-2015, whereas the rate of increase of T_min was 1.6 times that of T_max. ②The influence of diurnal asymmetry on vegetation dynamics showed an obvious difference, and NDVI and T_min showed a more significant correlation than NDVI and T_max in vegetative growth season. The effect of nighttime warming on NDVI in the Loess plateau is more significant than that of daytime warming. ③We also found diverse responses of vegetation type to daytime and night-time warming across the Loess Plateau. We found that the partial correlation between NDVI and T_max was positive (P<0.05) except grassland and crop. The partial correlation between T_min and NDVI of crop was negative (P<0.01), and the partial correlation between T_min and NDVI of other vegetation was positive (P<0.05). Our results provide a demonstration for studying regional responses of vegetation to climate extremes under global climate change.