树种水平地上生物量(每个树种的地上生物量)是衡量森林生态系统结构功能的重要指标.为揭示树种水平森林地上生物量变化机制及其与气候变化和干扰的关系,运用KNN(k-nearest neighbor distance)方法将森林调查数据和MODIS数据相结合,估算了黑龙江大兴安岭2000、2010和2015年树种水平的森林地上生物量,在此基础上运用典型对应分析和随机森林方法,分析了研究区树种水平地上生物量变化特征及其与气候和干扰因素的关系.研究结果表明:20002015年期间,研究区总的森林地上生物量增加了8.9%(0.41*10~8 t),其中20102015年期间地上生物量的增加速度要明显高于20002010年;地上生物量增加最多的树种为白桦(Betula platyphylla Suk.),与2000年相比生物量增加了0.40* 10~8 t,其次为樟子松(Pinus sylvestris var. mongolica Litv.)、山杨(Populus davidiana Dode)和蒙古栎(Quercus mongolica Fisch. ex Ledeb.),落叶松(Larix gmelinii(Rupr.) Kuzen)地上生物量下降了0.08*10~8 t,柳树(Chosenia arbutifolia(Pall.) A. Skv.)和云杉(Picea koraiensis Nakai)基本上没有变化;林火、采伐和造林等森林干扰均对树种水平地上生物量影响显著,林火对树种水平地上生物量的影响要高于造林和采伐;气候要素显示出了比干扰要素更为重要的作用,多年平均温度和降水解释了最多的树种水平地上生物量变异.年均温度与阔叶树种的生物量以及林火干扰有显著的正相关性,与总的森林地上生物量呈现出显著的负相关,与落叶松和白桦表现出微弱的负相关,预示着气候变暖将影响该区域的树种组成并降低该区域的森林生产力.
英文摘要:
Species-level biomass and its change are important indicators in the measurement of the reliability of forest ecosystem structure and function. To understand the changes in species-level biomass and its linkage to climate and forest disturbances, maps of changes in species-level biomass in the Great Xing' an Mountains between 2000 and 2015 were generated by integrating Moderate Resolution Imaging Spectroradiometer (MODIS) images with forest inventory data using k-nearest neighbor distance (kNN) methods. The distribution information regarding fires, logging, and afforestation in the Great Xing'an Mountains between 2000 and 2015 were also extracted from MODIS production. The relationships between species-level biomass changes, climatic factors, and forest disturbances were explored using canonical correspondence analysis (CCA) and the random forests method. The results showed that the total aboveground biomass in the study area increased by 8.9% (0.41*10~8 t) from 2000 to 2015. For the species-level biomass, white birch (Betula platyphylla Suk.) increased most significantly (0.40 * 10~8 t), followed by pine ( Pinus sylvestris var. mongolica Litv.), aspen (Populus davidiana Dode), and Mongolian oak ( Quercus mongolica Fisch. ex Ledeb.). The biomass of larch ( Larix gmelinii (Rupr.) Kuzen) declined the most, decreasing by 0.08* 10~8 t from 2000 to 2015. There were no obvious changes for the species-level biomass of willow ( Chosenia arbutifolia (Pall.) A. Skv.) and spruce ( Picea koraiensis Nakai) during these period. The areas of fire disturbance derived from MODIS products were consistent with the records in the yearbook for the Great Xing'an Mountains (R~2 = 0.91). Fire disturbance, logging, and afforestation all significantly influenced the changes in the species-level biomass ( P < 0.01). Among the three disturbance factors, fire disturbance exhibited the greatest explanatory ability for the variation in species-level biomass change based on the random forests method. However, climatic factors exhibited a relatively greater importance value than did forest disturbances. Annual temperature and precipitation explained the greatest variation in species-level biomass change among all the climatic factors and forest disturbances, and exhibited a strong positive correlation with broad-leaf species biomass and fire disturbance,but a negative correlation with total aboveground biomass (AGB),logging, and afforestation. A weak negative correlation between annual temperature and the biomass of white birch and larch was observed. Annual precipitation was positively correlated with the biomass of white birch and total AGB but negatively correlated with the biomass of larch and annual temperature. Our results indicated that climatic warming would increase the proportion of deciduous broad-leaved tree species and reduce forest ecosystem productivity in the Great Xing'an Mountains in northeast China.