黑龙江省森林大气CO_(2)施肥效应的时空变化趋势

Spatiotemporal Variation Trend Analysis of Atmospheric CO_(2) Fertilization Effect of Forests in Heilongjiang Province

探讨大气CO_(2)体积分数上升是否对黑龙江省森林存在显著的施肥效应,并进一步分析这一效应的时空变化趋势。采用了2种方法进行分析,第1种方法是利用自然生长的树木生长轮数据,结合广义加性混合模型(GAMMs),分析树木的径向生长是否存在显著的大气CO_(2)施肥效应;第2种方法则是基于通量数据对光能利用率模型修正后所估算的总初级生产力数据集(EC-LUEGPP),结合机器学习方法,探讨植被生产力变化是否存在显著的大气CO_(2)施肥效应及其时空变化趋势。结果表明:1960—2015年,所有9个红松天然林样地中有7个样地存在显著的大气CO_(2)施肥效应。同时在EC-LUE GPP数据中发现有88.1%的区域存在显著的大气CO_(2)施肥效应,而且在1982—2015年,黑龙江省森林总体的大气CO_(2)施肥效应呈现出显著的下降趋势(P<0.01),平均每年下降速率约为1.1%。

To explore whether an increase in CO_(2)concentration in the atmosphere leads to an evident fertilization effect on forests in Heilongjiang Province, and to further analyze the temporal and spatial trend of this effect, we adoptedtwo methods. The first method is to use the annual ring data of trees that grow naturally, combined with generalized additive mixed models(GAMMs), so as to analyze if the radial growth of trees has a remarkable atmospheric CO_(2)fertilization effect. The second one is with the estimated total primary productivity data set(EC-LUE GPP) from the corrected efficiency of solar energy utilization model by flux data, combined with machine learning methods, so as to find out whether changes in vegetation productivity has a pronounced atmospheric CO_(2)fertilization effect and also its temporal and spatial trends. The results demonstrate that seven of the nine natural Pinus koraiensis samples has a significant atmospheric CO_(2)fertilization effect(P>0.05) from 1960 to 2015. In EC-LUE GPP data, 88.1% of the areas prove that there is a remarkable effect of atmospheric CO_(2)fertilization. And between 1982 and 2015, the overall atmospheric CO_(2)fertilization effect of forests in Heilongjiang Province shows a dramatically downward trend(P<0.01), with an average annual decline rate of about 1.1%.