节点文献

北京大兴人工杨树林土壤呼吸动态与环境影响因素研究

The Dynamic of Soil Respiration and Its Dependence on Soil Temperature and Water Content in a Poplar Plantation in Daxing District of Beijing, China

【作者】 谭炯锐

【导师】 戴伟; 查同刚;

【作者基本信息】 北京林业大学 , 土壤学, 2010, 硕士

【摘要】 人工林土壤呼吸动态及其环境影响,是与人工林固碳效益密切相关的重要研究课题。本研究以位于北京大兴的杨树人工林(也是北京林业大学碳通量监测站点)为研究对象,采用外业调查、试验监测和室内分析相结合的方法,系统研究了林分土壤呼吸的时间动态及其环境影响因素,同时采用壕沟法和根生物量外推法研究了根系呼吸和非根系呼吸比例及动态规律。主要研究结果包括:(1)阐明了杨树人工林土壤呼吸的时间动态。杨树人工林土壤呼吸(Rs)的日动态呈单峰曲线,土壤CO2释放量在中午至下午时达到最大值,夜间随着土壤温度(T)的降低而降至最低值。季节尺度上,Rs夏季高冬季低,总的变化趋势为单峰曲线。全年Rs的最大值出现在7月下旬到8月上旬之间,分别为5.83(2007),5.07(2008),5.81(2009)μmolCO2·m-2·s-1,冬季Rs接近于零。2007,2008,2009年Rs的总值对应分别为54.43,49.06和48.79molCO2·m-2·a-1T和湿度(W)是控制Rs日动态和季节动态的主导因子,年际动态是植被生理、太阳辐射、土壤温湿度、叶面积指数和降雨量等因子综合作用的结果。(2)将典型降雨过程Rs与土壤水分特征值相结合进行分析,确定Rs的最适含水量为土壤萎蔫含水量到毛管断裂含水量(田间持水量的80%)之间。典型降雨过程中Rs与土壤湿度(W)的关系可以用三次方程(Rs=a+bW+cW2+dW3)来描述,R2变化范围为0.70-0.81。通过土壤水分特征曲线(W=0.2636ψ0.22。R2=0.963)拟合和环刀法测定的土壤田间持水量(FC) (10KPa)为15.65%±0.31%,萎蔫点(WP)(150KPa)为5.49%±0.52%,毛管断裂含水量(WRC) (40KPa)%11.12±0.47%。针对Rs和W关系观察到的土壤湿度临界值6%和11%,分别与WP 15.65%±0.31%和WRC 11.12±0.47%吻合。因此,Rs的最适含水量为WP到WRC(田间持水量的80%)之间;当W低于WP时,Rs随W的增大而升高;当W高于WRC(田间持水量的80%)时,Rs随W的增大而降低,W取代T成为影响Rs的主要因素。(3)揭示了土壤呼吸对土壤温度的响应Rs在一年中的变化的主要影响因素是T。全年Rs与T的关系可以用指数方程来描述,三年的拟合方程分别为Rs=0.758e-.044T, Rs=0.550e-.053T,和Rs=0.615e0-049T;T分别可以解释67%,78%和72%的土壤呼吸季节变化。2007,2008,2009年的Q1o平均值为1.63。春秋两季土壤呼吸速率的Q1o高于夏季和冬季,秋季Q10(1.57-1.86)最高。W在WP到WRC之间时Q10值(1.87-1.97)高于其他水分区间(1.26-1.86),也高于全年的Q10值(1.55-1.70)。(4)在验证常用经典模型的基础上,根据水分特征值进行了土壤呼吸与土壤温度和土壤水分的分段模拟。常用的9个Rs与正W双因素模型拟合结果表明,指数模型(Ln Rs=a+bT+cW +dTW2)对全年的Rs变化拟合效果较好,R2在0.75-0.81间。以WP和WRC作为临界点将数据分为三个区段后,拟合效果优于未分段。当W低于WP时,指数模型Ln RS=a+bT+cW+dTW的拟合效果最好,R2>0.86;当W在WP到WRC之间时,含指数形式的模型拟合效果最好,R2>0.80;当W大于WRC时,渐进模型Rs=aebTW/(W+c)拟合效果最好,R2>0.80。运用单一模型和分段模型对每一年的土壤呼吸速率进行模拟,获得的土壤呼吸速率与实测土壤呼吸速率(RsM)比较分析表明,使用分段模型模拟土壤呼吸数据比未分段模型模拟的数据更准确。分段模型中,各段选用不同模型的组合模型比分段后各段用统一模型的组合模型的土壤呼吸估算值与实测值更接近。(5)研究了根系呼吸(RRoot)和异养呼吸(RH)对土壤呼吸的贡献及其动态,并对根系呼吸和异养呼吸与土壤温、湿度关系模型进行了对比分析。用根生物量外推法和壕沟法两种方法测得的RRppt稍有不同。RRoot和RH都是夏季高冬季低,RRoot对Rs贡献率最高的时候是在八月,占59.32%,最小值出现在二月,只占21.47%,均值39.43%。RRoot和RH均与T呈显著的指数关系,RRoot的Q10(2.07)大于土壤总呼吸的Q10(1.63),大于RH的Q10(1.45)。土壤温湿度双因素经验模型中,RRoot比较适于用指数模型,指数-幂模型和渐进模型, R2>0.83;RH用渐进模型描述的R2最高,为0.90。

【Abstract】 Soil respiration of plantation and its environmental impact is an important research subject which is closely related to plantation carbon fixation. Poplar plantations in Daxing, Beijing (the carbon flux monitoring site of Beijing Forestry University) were studied with the field investigation, trial, monitoring and laboratory analysis methods. We systematically studied soil respiration temporal dynamics and its influencing environmental factors. The dynamic and the contribution of the non-root respiration and root respiration were studied by using trench method and root biomass extrapolation method. Main results are as follows:(1) Soil respiration temporal dynamics of poplar plantation.The diurnal curve of poplar plantation soil respiration (Rs) showed a single peak. Soil CO2 efflux reached the maximum at noon or afternoon and reduced to minimum with the decline of soil temperature (T) at night. Rs is high in summer (late July to early August) and low in winter at the seasonal time scale, which showed as a single-peak curve. The maximum value of Rs in the year of 2007-2009 are 5.83,5.07 and 5.81μmolCO2·m-2·s-1, and the annual total amount of soil CO2 efflux are 54.43,49.06 and 48.79 molCO2·m-2·a-1 respectively. Rs was close to zero in winter. T and soil moisture (W) were the main variables accounting for the diurnal and seasonal dynamic of Rs. Annual dynamic is the combined result of the influence of plant physiology, solar radiation, T, W, leaf area index (LAI) and precipitation and so on.(2) Rs and soil moisture characteristics are associated and analyzed in an innovative way during typical precipitation process, thus finding out the optimum soil water content of Rs is between wilting point and fracture pore water content (80% of field capacity).The relationship between Rs and W during a typical rainfall can be described by a cubic model Rs= a +bW+cW2+dW3), R2 ranges from 70% to 81%. By fitting the soil water characteristic curve (W=0.2636ψ-0.22, R2=0.963) and cutting ring method, we found that field capacity (10KPa) was 15.65%±0.31%, wilting point (WP) (150KPa) was 5.49%±0.52% and the water content of rupture of capillary (WRC) (40KPa) was 11.12±0.47%. The soil moisture threshold 6% and 11%, which were observed from the relationship between Rs and W, met the WP (15.65%±0.31%) and WRC (11.12±0.47%). Therefore, the optimum soil water content of Rs was between WP and WRC (80% of field capacity); when W was below the WP, Rs increased with W increased; when W was higher than WRC (80% of field capacity), the Rs decreased with W increased, W, instead of T, became the main factor of Rs variation.(3) Revealed the soil respiration response to soil temperatureT was the primary factor accounting for the Rs variation during a year. The relationship between Rs and T can be described by exponential equations, three fitted equations were:Rs=0.758e0.044T, Rs= 0.550e0.0537T, and Rs= 0.615e0.0497T; Tcan explain 67%,78% and 72% of the seasonal variation of Rs. The Q10 values of 2007,2008 and 2009 were 1.55,1.70 and 1.63 respectively. Q10 in spring and autumn are higher than in winter and summer. And Q10 in autumn (1.57-1.86) were higher than in spring (1.42-1.79), also higher than the Q10of whole years (1.55-1.70).When W between WP to WRC, Q10 (1.87-1.97) were higher than other W condition and the Q10of whole years.(4) Based on the verification of the classical models, Rs has been segmented modeled with T and W according to soil water characteristic values.The fitting results of 9 T-W double-factor empirical models indicate that exponential model (Ln Rs=a+bT+cW+dTW2) was the best model for Rs of whole years, R2 ranged 0.75-0.81. R2 have been improved after the data was segmented by WP and WRC. When W was lower than WP, exponential model Ln Rs=a+bT+cW+d T Wfitted better than others, R2>0.86; when W between WP and WRC, all the models which have a exponential form fitted better than others, R2>0.80; When W waas higher than WRC, Asymptote model Rs=aehTW/(W+c) was the best model, R2>0.80.Compared with measured soil respiration (RSM), estimated soil respiration (RSE) calculated by models after segmentation were more accurate than models before segmentation. Among segmented models, the RSE of combination of different models in each segment were closer to RSM/than the combination of same models in each segment.(5) Quantified dynamic and contribution of root respiration (RRoot) and heterotrophic respiration (RH), compared the T-W models of RRoot and RH.RRoot, measured with root biomass extrapolation method and trenching method were slightly different. RRoot and RH are higher in summer and lower in winter. The contribution of RRoot, to total soil respiration reached the maximum (59.32%) in August, reduced to the minimum (21.47%) in February. RRoot, and RH both had a exponential relationship with T, Q10 of RRoot, (2.07) was higher than total soil respiration (1.63), and RH (1.45). Among all the double-factor empirical models, exponential models, exponential-power model and Asymptote model fitted better than others for RRoot, R2>0.83; Asymptote model fitted RH better than others for RH, R2=0.90.

节点文献中: