2001—2022年西藏地区旱柳物候期变化及其影响因素

Phenophase variations and influencing factors of Salix matsudana in Xizang from 2001 to 2022

为了揭示西藏高原温带植物旱柳的物候对气候变化响应的时空格局,利用2001—2022年泽当、日喀则和林芝3个观测站旱柳物候期数据(叶芽膨大期(LED)、展叶始期(LUD)、开花始期(FFD)、叶变色末期(LCD)和落叶末期(LFD)),以及逐日平均气温(T_(m))、最高气温(T_(max))、最低气温(T_(min))、气温日较差(TDR)、相对湿度(RH)、降水量(P_(r))、日照时数(S)和≥0℃积温(∑T_(0))等资料,采用线性回归、Pearson相关系数和逐步回归等方法进行了分析。结果表明:(1)空间上,西藏地区旱柳春季物候期(LED、LUD、FFD)林芝最早、日喀则最迟,秋季物候期(LCD、LFD)相反;生长季长度(LOS)为238~268 d,以林芝最大、日喀则最小。(2)时间尺度上,2001—2022年泽当、日喀则旱柳各物候期均呈推迟趋势,林芝旱柳秋季物候期趋于推迟;林芝和日喀则旱柳LOS显著延长,分别为11.90和6.12 d·10 a^(-1),泽当旱柳LOS缩短(-4.50 d·10 a^(-1))。(3)旱柳生长季内T_(m)呈上升趋势,T_(max)、T_(min)趋势各不相同;泽当和日喀则TDR变大、P_(r)减少、S增加,林芝正好相反;林芝和日喀则∑T0显著增加,泽当∑T0明显减少。(4)对春季物候期影响最大的气象因子以气温为主,降水仅对泽当、林芝旱柳LCD以及泽当LFD的影响最大;季风指数(MI)只与泽当站旱柳LED、FFD呈显著相关,其他2站物候期相关性不显著;7月MI与T_(m)、P_(r)、S均存在显著的相关性,8月MI与S也有显著的负相关关系,这些气象因子虽不是影响物候期的主要因子,也会起到间接作用。

To reveal the spatial-temporal pattern of Salis matsudana phenology in response to climate change on Tibetan Plateau,both the phenophase data and daily meteorological data at Tsedang,Xigazê,and Nyingchi from 2001 to 2022 were analyzed to identify the change trend of phenophase of S.matsudana and its influencing factors.The phenophase data included leaf bud enlargement date(LED),first leaf unfolding date(LUD),first flowering date(FFD),end of leaf coloring date(LCD),and end of leaf fall date(LFD).The daily meteorological factors included the daily average temperature(T_(m)),maximum temperature(T_(max)),minimum temperature(T_(min)),temperature daily range(TDR),relative humidity(RH),precipitation(P_(r)),sunshine duration(S),and accumulated temperature above 0℃(∑T_(0)).These data were analyzed with linear regression,Pearson correlation coefficient,and stepwise regression methods.The results showed that:(1)Spatially,the earliest spring phenophase(LED,LUD,and FFD)of S.matsudana occurred in Nyingchi while the latest occurred in Xigazê.However,this phenomenon was exactly opposite for the autumn phenophase(LCD and LFD).The length of growth season(LOS)ranged from 238 to 268 days,with the largest value in Nyingchi and the smallest in Xigazê.(2)On the temporal scale,all phenophases of S.matsudana delayed in Xigazêand Tsedang,but only the autumn phenophases of S.matsudana delayed in Nyingchi in the past 22 years.The LOS of S.matsudana was prolonged in Xigazêand Nyingchi at growth rates of 11.90 and 6.12 d·10 a^(-1),respectively.In contrast,the LOS of S.matsudana was shortened at a rate of 4.50 d·10 a^(-1) in Tsedang.(3)Increasing trends of T_(m) during the growing season of S.matsudana were found in Xizang,but the increasing trends were different for T_(max) and T_(min).The increased TDR and S and decreased P_(r) occurred in Tsedang and Xigazê,but this situation was opposite in Nyingchi.∑T0 increased significantly in Xigazêand Nyingchi,but decreased significantly in Tsedang.(4)Air temperature was the leading meteorological factor influencing spring phenophases.In contrast,precipitation only exerted the greatest influence on the LCD of S.matsudana at Tsedang and Nyingchi,and the LFD at Tsedang.The monsoon index(MI)was only significantly correlated with the LED and FFD of S.matsudana at Tsedang station,but did not correlate with the phenophases at other two stations.Significant correlations between MI and T_(m),P_(r) and S occurred ubiquitously at all stations over the Tibetan Plateau in July,while MI showed a significant negative correlation with S in August.Although these meteorological factors were not the main factors affecting phenophases,they would also play an indirect role.