黄土高原地区柠条人工林土壤呼吸

Soil Respiration Characters in a Caragana Plantation in Loess Plateau Region

2005—2008年用红外气体分析法测定柠条人工林地的土壤呼吸。结果表明:柠条林地土壤呼吸具有明显的日变化特征,最大值出现在14:00左右,最低值出现在凌晨;柠条林地3—12月的土壤呼吸总量为814.9～1224.7gC·m-2,表现出明显的季节变化和年际变化;土壤呼吸与土壤温度呈显著指数正相关,方程的决定系数R2为0.31～0.67,由拟合的指数方程系数计算出柠条林2005,2006,2007和2008年的Q10值分别为2.02,1.70,1.76和1.75,生长季和非生长季的Q10值分别为0.64和2.11;2005和2006年土壤呼吸速率与土壤水分呈极显著线性相关(P<0.01),方程的决定系数R2在0.30左右,而2007和2008年则不显著(P>0.05);生长季(5—9月)土壤水分起主要作用,可以解释土壤呼吸季节变化的55%,非生长季土壤温度起主要作用,可以解释土壤呼吸季节变化的37%;4个双因子模型可以解释土壤呼吸季节变化的51%～83%。

By using an infrared gas analyzer（ IRGA） with a 6400-09 chamber,the diurnal,seasonal and interannual dynamics of soil CO2 efflux rate was measured in a Caragana plantation located in Taiyuan basin over four years from January 2005 to December 2008. The soil temperature （T） at 10 cm depth and the soil water content （W） over top 10 cm were also measured at the same time. The results showed that soil respiration rate（R） of the Caragana plantation had an asymmetric diurnal pattern,with the minimum appearing about 6： 00 and the maximum around 14： 00. The CO2 efflux exhibited pronounced seasonal variations that clearly followed the seasonal change in T,with high values appearing in August,and low value in early spring and winter. Cumulative annual respiration amount from March to December was 1 145. 7,1 224. 7,814. 9 and 904. 7 g C·m -2 a -1 ,respectively,in 2005,2006,2007 and 2008,and there was an obvious difference between years. The relationships between R and T at 10 cm depth could be described by exponential equation （R = aebT ）,with a determination coefficient （ R2 ） of 0. 41,0. 31,0. 67 and 0. 54,respectively,for 2005, 2006,2007 and 2008,and calculated Q10 （Q10 = e10b） varied from 1. 70 to 2. 02 over four years and was 0. 64 and 2. 11 for growing season and non-growing season,respectively. The relationships between R and W over top 10 cm depth had significant correlation （P 0. 01） in 2005 and 2006,with a determination coefficient （ R2 ） about 0. 30,however the relationship was not significant （P 0. 05） in 2007 and 2008. The four two-variable models of linear and non-linear were able to predict relationships of R with both T and W,with the R2 ranging from 0. 51 to 0. 83.