本文選題：亞熱帶 + 生物量��； 參考：《中國科學院研究生院(武漢植物園)》2016年博士論文

【摘要】：本文以中亞熱帶北部森林為研究對象,以獲取的解析木、湖南八大公山25公頃監(jiān)測樣地和湖北森林107個樣地調查數據為基礎,運用了層次貝葉斯、結構方程模型等技術手段,研究了群落尺度上的生物量估算模型構建、現狀、分布、驅動機制以及個體水平的碳分配策略和與物種多樣性的關系,目的在于探討亞熱帶森林的碳在個體、群落和生態(tài)系統(tǒng)內的分布和儲存機制。本文主要結論如下：1.通過147棵樹的地上部分和23個根部構建局域生物量估算模型,在原有模型基礎上,我們添加了木材密度和冠幅并檢驗其作用,發(fā)現木材密度能夠改進模型而冠幅卻不能。另外,我們用本實驗數據去驗證其他亞熱帶模型的預測能力,發(fā)現大多數模型對于本數據的預測質量都不好,但是模型中有樹高變量的比只有胸徑模型預測要穩(wěn)定些。基于此,我們建立一套中亞熱帶北部常綠落葉闊葉混交林地上和地下生物量模型。2.根據已經建好的模型,估算了湖南八大公山25 ha樣地生物量,發(fā)現20x20m樣方尺度上的生物量平均值是252.7±108.7 Mg/ha。通過空間插值的方法,繪制了森林生物量空間分布圖,發(fā)現生物量呈斑塊化分布,且與地形有較強的相關。同時,我們也發(fā)現大樹密度可以解釋71%的生物量變異,依此,我們應用方差分割發(fā)現,地形、空間和生物因子可以解釋約64.8%的生物量變異。通過選擇函數發(fā)現,地形因子中的凹凸度和地形濕度指數以及生物因子中的樹密度和個體密度為驅動因子,生物量與凹凸度、樹密度和木材密度正相關,與地形濕度指數負相關。影響大樹分布的環(huán)境因子與生物量相同,只是解釋度和相關關系有所差異。3.利用亞熱帶59個物種的樹結構數據,應用層次貝葉斯方法,檢驗樹結構(H-D, H-F, H-W1)在種內、種間和功能群的差異,發(fā)現群落水平上的樹結構表現出一定的收斂,大部分的物種沒有表現出種間差異。通過Kendall秩相關分析樹結構隨上限徑級和光的需求能力在個體生長階段的變化,發(fā)現高大物種有較纖細的樹干,在小徑級有較淺和窄的樹冠,在大徑級有較寬和深的樹冠；需光物種在在林冠上有較纖細的樹干,在中等徑級有較寬的樹冠。4.利用中亞熱帶107個森林樣地調查數據和土壤、地形、林齡數據,研究了亞熱帶物種多樣性與森林生態(tài)系統(tǒng)關系及其驅動機制。發(fā)現物種多樣性和森林生物量在亞熱帶是呈現積極的正相關,林齡和海拔與物種多樣性和生物量呈現正相關,土壤因子與多樣性負相關而與生物量正相關。當加入大樹密度后,物種多樣性與森林生物量關系從正相關變?yōu)樨撓嚓P,通過比較大樹的光需求值,發(fā)現這些大樹大都是需光物種。因而物種通過大樹來進行資源互補來影響森林生態(tài)系統(tǒng)功能,而這些大樹都是需光物種則表明有較強的選擇效應。
[Abstract]:Taking the forest in the northern part of the middle subtropics as the research object, taking the obtained analytic wood, 25 ha monitoring sample land in Badgongshan, Hunan Province and 107 sample plots of Hubei forest as the basis of the investigation data, this paper uses hierarchical Bayes, structural equation model and other technical means. In this paper, biomass estimation models on community scale, current situation, distribution, driving mechanism, individual level carbon allocation strategies and their relationship with species diversity are studied. The purpose of this study is to explore the individual carbon content of subtropical forests. Distribution and storage mechanisms in communities and ecosystems. The main conclusions of this paper are as follows: 1. Based on the local biomass estimation model of the aboveground parts and 23 roots of 147 trees, the wood density and crown width were added and tested. It was found that the wood density could improve the model, but the crown width could not. In addition, we use the experimental data to verify the prediction ability of other subtropical models. It is found that most of the models have poor prediction quality for this data, but the model with tree height variables is more stable than the DBH model. Based on this, we establish a model of aboveground and underground biomass of evergreen deciduous broad-leaved mixed forest in the north of central subtropics. Based on the established model, the biomass of 25 ha plot in Badagong Mountain, Hunan Province was estimated. It was found that the average biomass on the scale of 20x20m was 252.7 鹵108.7 mg / ha. The spatial distribution map of forest biomass was plotted by spatial interpolation. It was found that the biomass was patch distributed and had strong correlation with terrain. At the same time, we also found that tree density can account for 71% of the biomass variation. Based on this, we found that terrain, space and biological factors can account for about 64.8% of the biomass variation. Through the selection function, it is found that the concave convexity and topographic humidity index in topographic factors and the tree density and individual density in biological factors are driving factors, and the biomass is positively correlated with concavity, tree density and wood density. It is negatively correlated with topographic humidity index. The environmental factors affecting tree distribution were the same as biomass, but the degree of interpretation and correlation were different. 3. Using tree structure data of 59 subtropical species and using hierarchical Bayes method, the differences of tree structure within species, between species and functional groups were examined. It was found that the tree structure at community level showed a certain convergence. Most species do not show interspecific differences. By Kendall rank correlation analysis, it was found that tall species had slender trunks, shallower and narrower crowns, and wider and deeper crowns at large diameter levels. Light species have slender trunks on the canopy and a wider crown. 4 at the middle diameter level. The relationship between species diversity in subtropics and forest ecosystem and its driving mechanism were studied by using the survey data of 107 forest plots in the middle subtropics and the data of soil, topography and forest age. It was found that species diversity was positively correlated with forest biomass in subtropics, forest age and altitude were positively correlated with species diversity and biomass, soil factors were negatively correlated with diversity but positively correlated with biomass. When tree density was added, the relationship between species diversity and forest biomass changed from positive correlation to negative correlation. By comparing the light demand values of big trees, it was found that most of these trees were light-demanding species. As a result, species complement each other to influence forest ecosystem function, and these big trees are light-demanding species, which indicate that they have strong selective effect.
【學位授予單位】：中國科學院研究生院(武漢植物園)
【學位級別】：博士
【學位授予年份】：2016
【分類號】：S718.5
，

本文編號：1918221