农田下垫面观测通量的变化特征及其气候学足迹分析

Analysis of the Characteristics of Turbulent Flux and Its Footprint Climatology at An Agricultural Site

利用2008—2010年馆陶站涡动相关仪和自动气象站观测资料,在保证数据质量的基础上,分析了通量交换特征,并采用算术平均和通量加权2种气候学足迹计算方法,详细探讨了不同时间尺度观测通量的源区分布特征。结果表明:①感热和潜热通量全年都有相同的日变化特征,CO2通量仅在作物生长季表现出与水热通量相反的日变化趋势。各通量受作物种类及其物候特征影响,季节变化明显。在生长季以潜热通量、CO2通量交换为主,其值在生长中期也明显高于生长始末期,且夏玉米的交换量强于冬小麦的。②受风、湍流条件、下垫面状况等共同影响,日尺度的观测通量源区差异最显著,季节尺度次之,年尺度最小。③不同时间尺度下,通量加权的气候学足迹能更合理地反映观测通量源区的平均状况,其源区大小普遍小于算术平均的结果。

Based on quality controlled data from eddy covariance system and automatic weather station collect ed at Guantao farmland site from 2008 to 2010, the characteristics of diurnal, seasonal and annual variations of tur- bulent flux were reported. The corresponding source areas of flux measurement at different temporal scales were an alyzed in detail, using arithmetic-averaged and flux-weighted footprint climatology calculation method, respectively. The main findings are as follows. Firstly, sensible heat and latent heat flux both show consistent diurnal variation throughout the year, while CO2 fluxes only have significant diurnal variation in growing season with an opposite trend. The seasonal variation of the turbulent flux is mainly affected by the crop type and its growth status in differ ent phenological periods. During growing season, latent heat flux and CO2 flux are the dominant flux exchange i tems whose value are significantly higher in their middle growth stage than other ones during which latent heat and CO2 flux exchange of the summer corn is stronger than winter wheat. Secondly, with combined effects of wind, tur bulence and surface condition, the source area of flux measurement change most significantly at daily scale, less obvious at seasonal scale and smallest at annual scale. Finally, compared with arithmetic-averaged footprint clima- tology method, flux-weighted footprint climatology is a more reasonable method to calculate the source areas of the flux measurement, in that they account for the time change of the actual turbulent flux. The arithmetic-averaged re sults are most likely to overestimate the size of source area during small observed flux due to its weak turbulent ex- change.