内蒙古温带荒漠草原能量平衡特征及其驱动因子

基于内蒙古苏尼特左旗温带荒漠草原生态系统观测站2008年全年的涡度相关观测与相应的生物、环境观测资料,对生态系统能量平衡特征及其驱动因子分析表明:能量平衡各分量(净辐射,Rn;感热通量,H;潜热通量,LE;土壤热通量,G)呈单峰型日动态,白天大部分时间H/RnG/RnLE/Rn;夜间G/Rn占主导;全天LE/Rn相对较小,即使在植物生长盛期。较低的LE/Rn可能与荒漠草原气候干旱及植被分布稀少有关。日Rn受天气变化的影响,特别是在雨季,Rn日间差异较大,呈现锯齿状波动。能量平衡各分量季节变化明显,Rn、H、LE和G最大月分别为7、5、6月份和6月份。全年H是Rn的主要能量支出项(58%);LE其次(26%),年蒸散量(190.3mm)大于年降水量(136.3mm),与多年平均降水量接近(183.9mm),其中最大日蒸散率3.8mm/d;G所占比例较小(1%),全年基本保持平衡。然而G白天吸收能量,夜间释放能量;夏季储存能量,冬季释放能量的特点,在能量平衡中存在类似"能量缓存"的作用,不能被忽略。降水过程显著影响内蒙古温带荒漠草原水热交换。降水后较降水前LE峰值明显增大,而H峰值降低。日蒸散率峰值多数与降水事件有关。而且,生长季日蒸散率波动与降水引起的SWC变化趋势一致。生长季潜热分配(LE/Rn)主要受到土壤含水量(SWC)、饱和水汽压差(VPD)及叶面积指数(LAI)共同影响。LE/Rn随SWC增大呈增加趋势,LE/Rn随VPD增大而降低,LE/Rn随LAI增大呈二次曲线变化。其中LAI为0.2m2/m2是一个阈值,当LAI0.2m2/m2,SWC是LE/Rn主要驱动因子;当LAI0.2m2/m2,SWC和LAI共同驱动LE/Rn。应用退耦因子(Ω)评价了荒漠草原与大气之间水汽交换的耦合状况。与其他草原类型相比,本研究区退耦因子(Ω)相对较低(生长季平均0.15)。生长盛期Ω相对较高,Rn是LE的主导因子;而生长前期和后期Ω相对较低,VPD是LE的主要控制因子。

Characteristics and driving factors of energy budget over a temperate desert steppe in Inner Mongolia

The driving factors of water vapor and energy exchange over an Inner Mongolian temperate desert steppe (Sunitezuoqi Temperate Desert Steppe Ecosystem Research Station) were investigated during the year of 2008, using eddy covariance, biotic and environmental data. The results showed that diurnal variation of energy balance components (net radiation (Rn); sensible heat flux (H); latent heat flux (LE); soil heat flux (G)) followed the expected single-peak shape. During daylight hours, H/Rn > G/Rn > LE/Rn, and at night the ratio G/Rn dominated the energy balance. On a daily basis, LE/Rn was relatively lower than H/Rn, even during the peak growth period. This energy partitioning could be explained by the typical dry climate and low vegetation cover of the desert steppe. The values of daily Rn were affected by the weather variation, especially during the rainy season when high day-to-day variation was observed. Energy balance components showed different seasonal patterns. The maximum monthly totals of Rn, H, LE and G occurred in July, May, June and June, respectively. On an annual scale, H was 58% and LE was 26% of Rn. Annual evapotranspiration (190.3 mm) was higher than the annual precipitation (136.3 mm), and got close to the annual average precipitation (183.9 mm). The maximum daily evapotranspiration rate was 3.8 mm d-1. As expected, G was a minor component of the annual energy balance (1%). Nevertheless, G was highly relevant on different time scales, acting as an "energy buffer" in the energy balance, i.e., the soil stored energy in the daytime and in the summer and released energy at night and in the winter, respectively. Precipitation profoundly affected the water vapor and energy exchange over the steppe, mainly by changing the soil water content (SWC). After precipitation events, the midday values of LE increased considerably, while H showed the opposite. The variation of daily evapotranspiration rates followed closely the variation of SWC during the growing season (from May 1st to October 15th), showing a direct relationship with SWC. On the daily scale, LE/Rn was effected by SWC, saturation water vapor pressure deficit (VPD), and leaf area index (LAI) during the growing season. Linear increase of LE/Rn with SWC and linear decrease of LE/Rn with VPD were observed. Additionally, quadratic curve variation of LE/Rn with LAI was shown. It was also found a LAI threshold of 0.2 m2/m2. When the value of LAI was less than 0.2 m2/m2, LE/Rn was driven by SWC alone; otherwise if LAI was more than 0.2 m2/m2, LE/Rn was co-driven by SWC and LAI. Decoupling coefficient was applied to assess degree of the coupling between the desert steppe and the atmosphere. Compared with other grassland ecosystems, the decoupling coefficient of this steppe was relative low (0.15 in growing season). Nevertheless, results showed that LE was mainly controlled by Rn during the peak-growth stage, because of the relatively high decoupling coefficient values. On the other hand, VPD was the main controlling factor during the early and later growth stages, due to the relatively low decoupling coefficient values.