随着气候变化的加剧,极端天气事件频率和强度均有所增加,其对生态系统结构和功能的影响远高于渐进的趋势变化.热带森林作为陆地生态系统的重要组成部分,其物候对于气候变化的响应始终为研究热点,但由于其高植物多样性与常绿性特征而无法统一科学发现.本研究以物候特征明显的纯林橡胶林为研究切入点,逐像元多曲线拟合橡胶林物候,通过机器学习等手段筛选了物候响应敏感的极端天气事件,揭示了2003-2018年期间的春季物候(Start of growing Season,SOS)、秋季物候(End of growing Season,EOS)和极端天气事件的时空分布模式;基于偏相关性分析,探究了极端气候指数对物候的影响.结果表明:(1)研究期间海南岛橡胶林SOS多以0.73 d·a-1的趋势提前,EOS多以0.60 d·a-1的趋势推迟;少数极冷事件呈逐年增加趋势,而极热事件则呈相反趋势;(2)极端昼夜温度是影响SOS、EOS的主要因素,冷夜日数(TN10p)、冷日日数(TX10p)与SOS呈正相关,而暖夜日数(TN90p)、暖日日数(TX90p)与SOS呈负相关;TN10p、TN90p、TX90p与EOS呈正相关,但TX10p与EOS呈负相关;(3)SOS、EOS对不同强度和频率的极端天气事件的响应在空间上存在明显的东西差异.本研究结果表明考虑极端天气因素能够增进对热带森林结构与功能响应气候变化的理解.
The impact of extreme weather on the phenology of rubber plantations in Hainan Island,China
With the intensification of climate change,the frequency and intensity of extreme weather events have increased,and their impacts on ecosystem structure and function far exceeds those of gradual trend changes.As an important component of terrestrial ecosystems,the phenological response of tropical forests to climate change has always been a research hotspot.However,due to their high plant diversity and evergreen characteristics,scientific findings have not been consolidated.This study focuses on monoculture rubber plantations with distinct phenological characteristics as an entry point.Extreme weather events sensitive to phenological responses were selected by fitting multi-curves pixel-by-pixel to rubber plantations phenology and employing machine learning techniques to reveal the spatiotemporal distribution patterns of spring phenology(Start of growing Season,SOS),autumn phenology(End of growing Season,EOS),and extreme weather events during 2003-2018.The impacts of climate extreme indices on phenology were analyzed based on partial correlation analysis.The results show that the SOS of rubber plantations in Hainan Island advanced by an average of 0.73 d·a-1,while the EOS was delayed by 0.60 d·a-1.A few extreme cold events showed an increasing trend,while extremely hot events showed the opposite trend.The days of extreme day-time and night-time temperatures were the main factors affecting SOS and EOS;the number of days with low day-time temperature(TN10p)and the number of days with low night-time temperature(TX10p)were positively correlated with SOS,while the number of days with warm day-time temperature(TN90p)and the number of days with warm night-time temperature(TX90p)were negatively correlated with SOS;TN10p,TN90p,and TX90p were positively correlated with EOS,but TX10p was negatively correlated with EOS.The responses of SOS and EOS to extreme weather events with different intensities and frequencies exhibited obvious east-west differences in space.All these findings showed that it can enhance our understanding of the response of tropical forest structure and function to climate change when extreme weather factors are considered.
NETL
NSTL
万方数据
