【作者】 陈亮中；

【导师】 谢宝元； 肖文发；

【作者基本信息】 北京林业大学 ， 生态环境工程， 2007， 博士

【摘要】 本文选择长江三峡库区马尾松针叶林、栎类混交林等11种主要森林植被类型土壤有机碳(SOC)为研究对象，采用土壤类型法和植被类型法分别对三峡库区土壤类型背景碳库及主要森林植被类型下土壤总有机碳库进行了估算，并对三峡库区主要森林植被类型SOC的组成、分配特征及其影响因子进行了分析，探讨库区主要森林植被类型下SOC的分配规律及其与各影响因子间的关系。主要结论如下：(1)采用土壤类型法估算的三峡库区0～20 cm和0～100 cm土层平均SOC密度分别为3.20 kg／m~2和12.59 kg／m~2；SOC总贮量分别为190.48×10~6 t和750.51×10~6 t。各土壤类型中，以黄壤、黄棕壤、暗棕壤等土壤类型碳密度较高，紫色土和粗骨土类型碳密度较低。(2)将三峡库区森林植被按主要优势树种和分布面积划分为马尾松针叶林、栎类混交林等11种主要森林植被类型。研究表明，不同森林植被类型下SOC含量和SOC密度均存在较大差异，二者总体上都随土层加深而降低。11种主要森林植被类型土壤平均厚度在56.3～98.5 cm间，其中杉木针叶林土壤最厚，达98.5 cm，灌木林土壤最薄，平均厚度仅56.3 cm。库区11种主要森林植被类型土壤总SOC贮量为366.36×10~6t，其中0～10 cm、10～20 cm、20～40 cm和＞40 cm土层分别占22.90％、18.36％、28.33％和30.41％。与库区森林植被覆盖率总体上东部高于西部的趋势相反，库区各主要森林植被类型下SOC密度的空间分布从东至西呈现增加的趋势。(3)三峡库区主要森林植被类型下凋落物总有机碳贮量为1150.31×10~4t。11种主要森林植被类型中，以杉木林和温性松林下凋落物生物量最大，柑桔林凋落物生物量最小。各森林植被类型凋落物含碳率为379.70～514.91 g／kg。研究发现，三峡库区各主要森林植被类型下SOC与凋落物生物量之间的相关关系并不显著。(4)对采用环刀渗透法测定土壤水溶性有机碳(WSOC)的方法研究结果表明，采用环刀渗透法代替常规的浸提法测定土壤WSOC的方法切实可行。本研究采用该方法测定的各主要森林类型下WSOC总贮量为6.53×10~6 t。11种森林植被WSOC库分别为：马尾松林2.43×10~6 t，栎类混交林1.57×10~6 t，灌木林1.22×10~6 t，柏木林0.46×10~6 t，杉木林0.60×10~6 t，柑桔林0.38×10~6 t，其它软阔林0.22×10~6 t，针阔混交林0.22×10~6 t，竹林0.16×10~6 t，其它硬阔林0.20×10~6 t，温性松林0.06×10~6t。(5)三峡库区11种主要森林植被类型下活性有机碳(ASOC)、缓性有机碳(SSOC)和惰性有机碳(RSOC)贮量分别占土壤总有机碳的2.2％～6.0％，46.6％～81.3％和16.1％～50.0％。各森林植被类型下ASOC、SSOC、RSOC和WSOC均与土壤总有机碳之间存在很好的回归关系。不同森林植被类型土壤上层SSOC和RSOC含量大于下层的SSOC和RSOC含量，但ASOC在土层间的变化规律不明显。(6)三峡库区宜昌、万州、涪陵、重庆4个站点5 cm、15 cm、20 cm土层的地温与SOC含量均存在一定的负相关关系。由于整个库区地理跨度较小，本研究所选择4个站点的30年平均温度差别不大等原因，地温与SOC之间的这种相关关系并不显著。(7)三峡库区各主要森林植被类型下SOC含量与土壤质地、容重之间均存在很好的负相关关系。11种森林植被类型下SOC与土壤质地总的回归方程为：Y=27.69-3.88X(R~2=0.146)；SOC与土壤容重间的关系则可用方程：Y=56.221-29.506X(R~2=0.361)进行表达。研究还表明，土壤水分含量、pH值和三要素养分含量均对SOC有十分重要的影响。其中，pH值和土壤水分烘干系数与SOC呈现负相关关系，土壤质量含水量和养分含量与SOC呈显著正相关关系。(8)海拔、坡向、坡度等环境因子与SOC含量存在较明显的相关关系，各森林植被类型下，海拔和坡度对SOC有正效应，而坡向因子对SOC有负效应。回归分析还表明，不同森林植被类型和不同土层，海拔、坡向、坡度等环境因子对SOC变异的可解释程度不同。(9)与农耕地相比，退耕还林实施5年后，大部分退耕模式下0～30 cm的土层SOC含量和碳密度都有不同程度的提高，但在整个0～60cm土层，SOC含量和碳密度没有显著变化(个别模式甚至还有一定程度的下降)。分析认为，5年的退耕还林经营对0～60 cm土层SOC的影响有限。更多还原

【Abstract】 The soil organic carbon (SOC) under 11 forest types such as pinus massoniana forest, rubour mixed forest and shrubs of the Three Gorges reservoir area (TGRA) were studied. Method of soil type was chosed to calculate the SOC pool of the main soil types as the background pool of the SOC under 11 forest types, and method of vegetation type was chosed to calculate the SOC pool under 11 forest types. Fraction and distribution of SOC under main forest types, and the control factors of SOC in this paper were studied too. Major conclusions were summarized as follows:(1) The SOC density of the TGRA of 0～20 cm was 3.20 kg/m~2, the total SOC storage was 190.48×10~6 t; SOC density of 0～100 cm was 12.59 kg/m~2, the total SOC storage was 750.51×106 t. The SOC densities of yellow soil, yellow-brown soil and dark brown soil were higher than those of other soil types, and the SOC densities of purple soil and skeleton soil were the lowest.(2) Based on the data of forest inventory, the forest types of the TGRA were divided into 11 types of vegetation by its dominant tree species and acreages, such as Pinus massoniana forest, mixed forest of Quercus, and Shrubs, etc. The concentrations and densities of SOC under 11 types of vegetation were significantly different and generally decreased with depth. The average soil thickness of 11 types of vegetation was 56.3～98.5 cm, the soil thickness of forest of Cunnighamia Lanceolata was the largest (98.5 cm), and that of shrubs was the smallest (56.3 cm). The total storage of SOC under 11 main types of forest types in the TGRA was 366.36 ×10~6 t, with 22.90 %, 18.36 %, 28.33 % and 30.41 % in 0～10, 10～20, 20～40 and >40 cm soil depth interval, respectively. The densities of SOC under every vegetation type in the west of the TGRA were higher than those in the east of the TGRA. It changed counter to the trend of cover rate of forest vegetation in the TGRA.(3) The total carbon storage of litters under main forest types of the TGRA was 1150.31×10~4 t. The biomass of litter under cunnighamia lanceolata forest was the most of 11 forest types, and that of forest of Citrus reticulate, Citrus aurantium and Citrus grandis was the least. Content of litters under 11 forest types was 379.70～514.91g/kg. The study results show that the biomass was related with SOC, but the relationship was not significant.(4) The study on methods of determing of WSOC indicated that the method of ring sampler infiltration was practicable. The total storage of WSOC of the main forest types of the TGRA detrmined by ring sampler infiltration method in this paper was 6.528×10~6 t. The storages of WSOC under 11 forest types were as follows: Pinus massoniana forest 2.43 ×10~6 t , robur mixed forest 1.57 ×10~6 t, shrubs 1.22×10~6 t, Cupressus funebris forest 0.46×10~6 t, Cunninghamia lanceolata forest 0.60×10~6 t, Citrus reticulate plantation 0.38×10~6 t, other soft broad leaved forest 0.22×10~6 t, the broad leaved and coniferous mixed forest 0.22×10~6 t, Neosinocalamus affinis 0.16×10~6 t, other hard broad leaved forest 0.20 ×10~6 t, temperate Pine forest 0.06×10~6 t.(5) Contents of ASOC, SSOC and RSOC under 11 forest types occupied SOC with 2.2 %～6.0%, 46.6 %～81.3 % and 16.1 %～50.0 % , respectively. ASOC, SSOC, RSOC and WSOC under every main forest type in the TGRA were close related to SOC. The contents of SSOC and RSOC in upper soil layer werer higher than those in underlayer, but content of ASOC did not change with depth.(6) The contents of SOC were negative associated with soil temperature in 5 cm, 15 cm and 20 cm soil layer at 4 stations: YiChang, WanZhou, FuLing and ChongQing. The relationship of SOC and soil temperature were not significant, because of the small span in geography of the TGRA and the little difference among the averages of 30 years of soil temperature at 4 stations.(7) The contents of SOC were negative associated with soil texture and soil bulk density significantly. The regression equation of SOC and soil texture: Y = 27.69 - 3.88 X(R~2 = 0.146); the relationship between SOC and soil bulk density could be described by equation: Y = 56.221 - 29.506X(R~2 = 0.361) . The results of this paper also indicated that soil pH and content of soil water and soil nutrients had a very important influence on SOC. Soil pH and the soil coefficient of drying were negative associated with SOC, while soil nutrients and soil quality content of water were positively associated with SOC significantly.(8) Gradient and elevation had a positive influence on SOC under serval main forest types of the TGRA, while the factor of aspect had a negative influence on SOC. Analyse of linear regression indicated that the description veracity of regression equations changed with forest types and soil layers.(9) In five years after the beginning of the project of Grain for Green, the concentration and inventory of SOC in the 0～30 cm soil layer under most reafforestation patterns were higher than those under cropland conditions, while the concentration and inventory of SOC in the 0～60 cm soil layer did not change or even decreased in some cases. The concentration and inventory of SOC and the decreasing magnitude of SOC with the soil profile depth under all the studied reafforestation patterns were lower than those under shrubbery land conditions, which received the least impact of land use change. It is thus suggested that the impact of the five-years reafforestation patterns on the SOC in the 0～60 cm soil layer is insignificant.更多还原