2004年和2005年作物生长季,分别在三江平原水田和旱田,利用涡度相关法(EC)和暗箱-气相色谱法(DC—GC)开展了夜间生态系统呼吸比对实验.结果表明,湍流交换充分时2种方法观测数据没有显著性差异,并在选取更高摩擦风速下EC数据后,二者数据一致性得到进一步提高;EC观测的夜间呼吸与箱法模型的结果差异显著(P〈0.001),并因为EC数据本身巨大的变异性而与箱法模型结果的线性相关很弱(水稻田:R^2=0.06;大豆地:R^2=0.23),但在较长时间域内合计数据则可以明显地提高两者的线性相关(水稻田:R^2=0.16;大豆地:R^2=0.75).无论是与DC-GC观测数据比较,还是与其模型结果比较,EC观测的夜间呼吸都比DC—GC法低4%~30%,并且低估程度随季节变化而变化,在生长初期和晚期低估程度最小,而在生长旺期低估程度最大.
Measurements of nocturnal ecosystem respiration were conducted in a rice paddy of Sanjiang from July to September in 2004 and in a soybean field from June to September in 2005, using eddy eovarianee (EC) and dark chamber-gas chromatography (DC-GC). The differences of simultaneous data measured by two methods were not significant when night turbulent mixing was well The correlations between two methods can be improved under more large friction velocity. EC measurements were poorly correlated to night ecosystem respiration estimated by chamber models for the variability inherent to EC measurements( R^2 = 0.06, 0.23 for rice and soybean, respectively). However, this correlation can be improved by aggregating measurements over longer time period (R^2 = 0.16, 0.75 for rice and soybean field, respectively). In general, whatever comparison with simultaneous chamber measurements or modeled results, eddy eovarianee measurements were consistently lower 4% - 30% than chambers, and this bias varied with crop growing stage, which were maximum in middle and minimum in early and late growing stage.