【摘要】 【目的】研究中亚热带常绿阔叶林（天然林）改造成人工林后土壤碳排放量的变化及主要影响因子,为评估森林类型转换对土壤碳排放的影响提供科学依据。【方法】在福建农林大学西芹教学林场的常绿阔叶林及由其改造而来的38年生闽楠人工林与35年生杉木人工林中分别设置4块20 m×20 m样地,利用Li-8100土壤碳通量观测系统于2014年9月—2016年9月进行定点观测,并同期观测土壤温度、含水量、有机碳含量（SOC）、微生物生物量碳含量（MBC）、可溶性有机碳含量（DOC）、0～20 cm土层细根生物量和年凋落物量及凋落物碳氮比（C/N）。【结果】常绿阔叶林改造成闽楠（38年后）和杉木人工林（35年后）,年均土壤碳排放通量由16. 22显著降为12. 71和4. 83 tC·hm^-2a^-1,分别减少21. 60%和70. 20%;各林分类型的土壤呼吸温度敏感性Q¹⁰值表现为常绿阔叶林（1. 97）<闽楠人工林（2. 03）<杉木人工林（2. 91）,转换为杉木人工林后,Q¹⁰值显著升高（P<0. 05）;土壤温度能分别解释常绿阔叶林、闽楠人工林与杉木人工林土壤呼吸速率变化的89. 70%、88. 50%和87. 90%,土壤呼吸速率和土壤含水量相关不显著（P>0. 05）;土壤呼吸速率和SOC、MBC、DOC、年凋落物量及0～20 cm土层细根生物量均极显著正相关（P<0. 01）;土壤呼吸温度敏感性指数Q¹⁰值和凋落物C/N极显著正相关（P<0. 01）,而与年均土壤呼吸速率及MBC极显著负相关（P<0. 01）;进一步分析发现土壤MBC和SOC含量是影响土壤呼吸速率的2个最重要因子,而凋落物C/N在影响土壤呼吸温度敏感性中的贡献最大。【结论】中亚热带地区常绿阔叶林改造成闽楠（38年）或杉木（35年）人工林后,土壤碳排放通量显著降低。林分类型转换后树种组成和林分结构发生改变,凋落物数量、质量及细根生物量显著降低,土壤SOC和MBC含量显著下降可共同导致土壤呼吸通量的下降。土壤温度是3种林分类型土壤呼吸季节变化的主导因素,而土壤总有机碳库和土壤微生物量碳库的差异是不同林分之间土壤呼吸差异的主导因素,凋落物C/N对土壤呼吸的Q¹⁰影响最大。为提高模型预测森林类型转换影响土壤碳排放的精度,应综合考虑土壤有机碳库、易变性有机碳库及底物质量的变化。更多还原

【Abstract】 【Objective】The aim of this study was to explore the changes in soil carbon（C） fluxes after conversion from evergreen broad-leaved forest（natural forest） to plantations for 35 and 38 years in central-subtropical areas,the results will provide a theoretical basis for assessing soil C emissions.【Method 】An evergreen broad-leaved forest and two plantations converted from the natural forest,i.e.,a Phoebe bournei stand and a Cunninghamia lanceolata stand in Xiqin Forest Farm of Fujian Agriculture and Forestry University were monitored. Four plots（20 m×20 m） were established in each stand and soil carbon dioxide fluxes were measured using an Automated Soil CO₂ Flux System（Li-8100） over 25 months（from September 2014 to September 2016）. Soil temperature,soil water content,organic C content,microbial biomass C content,soluble C content,annual litter biomass,fine root biomass（0-20 cm soil layer） and litter C/N were measured.【Results】Results showed that stand conversion significantly reduced soil C fluxes. The annual soil C fluxes decreased from 16. 22 tC·hm-2 a-1 in natural forest to 12. 71 and 4. 83 tC·hm^-2a^-1 in a Phoebe bournei stand and a Cunninghamia lanceolata stand,reduced by 21. 6% and 70. 2% respectively. Stand conversion resulted in increased temperature sensitivity of soil respiration. The Q¹⁰ values were ordered as follows: evergreen broad-leaved forest（1.97） <P. bournei stand（2.03） <C. lanceolata stand（2.91）. The Q¹⁰ value of C.lanceolata stand was significantly higher than that of the evergreen broad-leaved forest（P < 0. 05）. Soil temperature explained 89. 70%,88. 50%,and 87. 90% of the variation in soil respiration rate in the evergreen broad-leaved forest, P. bournei stand and C. lanceolata stand,respectively. There were no significant relationships between soil water content and soil respiration rate（P > 0. 05）.Correlations showed highly significant positive relationships between soil respiration rate and soil organic C content,dissolved organic C,microbial biomass C,annual litter biomass,fine root biomass（0-20 cm soil layer）（P < 0. 01）.Accordingly,the Q¹⁰ values were significantly positively correlated with litter C/N ratios, whereas it significantly negatively correlated with annual C fluxes and microbial biomass C（P<0.01）. Further analysis showed that soil organic C and microbial biomass C were two important factors for determining the changes in soil respiration rates. However,litter C/N ratios were the most important regulators of Q¹⁰ values.【Conclusion】In central-subtropical area,stands conversion from evergreen broad-leaved forest to plantations for 35 and 38 years resulting in significant decreases in soil C fluxes,which were associated by shifts in tree species composition and structure,decreased quantity and quality of litters,fine root biomass,soil organic C and microbial biomass C. Soil temperature drove the seasonal changes in soil respiration,soil organic C and soil microbial biomass C were the key factors determine soil respiration rates,while litter C/N ratio was the most important factor of Q¹⁰ values. Our study highlights that soil organic C,soluble C and substrate quality should be incorporated into future models when predicting stand conversion effect on accuracy of soil C emissions.更多还原