Soil Respiration of the Alpine Meadow in Northwestern Sichuan, China: Seasonal Dynamics and Grazing effects

除植被冠层的光合作用之外，土壤的呼吸作用是陆地生态系统碳收支中最大的通量。土壤呼吸即使发生较小的变化也能显著地减缓或加剧大气中CO2浓度的增加，从而明显影响到全球气候变化。土壤呼吸速率变化与否以及变化的方向可以反映生态系统对环境变化的敏感程度和响应模式。尽管如此，土壤呼吸仍是一个为人们了解不多的生态系统过程。 草地生态系统是陆地生态系统的一个重要组成部分。针对草地土壤呼吸进行野外实验研究和相应方法论的探讨将对区域乃至全球碳源汇性质的准确估算具有重要的科学意义。然而，近几年来关于草地土壤呼吸的主要研究工作都集中在温带草原和部分热带草原，而针对高寒草甸生态系统土壤呼吸的研究报道还很少。 2008年4月至2009年4月期间，我分别在2008年6、8、10、12月和2009年2月和4月分6次对川西北的典型高寒草甸群落的土壤呼吸进行观测，分析了不同类型高寒草甸群落土壤呼吸的季节变化特征以及环境因子和放牧模式对其影响。主要研究结果如下： 1）该地区高寒草甸生态系统在生长季（6月～8月）土壤呼吸速率较大（6.07～9.30μmolCO2¡m-2¡s-1 ） ， 在非生长季（ 12 月～ 2 月） 较小（ 0.16 ～0.49μmolCO2¡m-2¡s-1 ） 。土壤CO2 年累积最大释放量为3963 ～ 5730gCO2¡m-2¡yr-1，其中，生长季土壤CO2的释放量占年总释放量的85%～90%。非生长季占10%～15%。非生长季所占比例略小于冬季积雪覆盖地区的冬季土壤呼吸占年土壤呼吸量的比例（14%～30%）。温度，尤其地温，是影响该地区高寒草甸生态系统土壤呼吸速率的最主要环境因子。土壤呼吸速率与地上生物量和土壤水分之间没有显著相关性，但是土壤含水量过大会导致土壤呼吸速率下降。 2）在观测期内，草丘区的土壤呼吸显著高于对照区的土壤呼吸，其最大土壤呼吸速率为16.77μmolCO2¡m-2¡s-1，土壤CO2 年累积最大释放量为8145gCO2¡m-2¡yr-1，是对照区的近2 倍。由于草丘在高寒草甸中占有较大的面积比例（近30%），因此，它将对高寒草甸生态系统的碳循环起着重要的作用。 3）放牧模式不仅可以影响高寒草甸群落的土壤CO2 排放，而且还可以改变土壤呼吸的温度敏感性（Q10）。本研究表明，在生长季有长期放牧活动干扰时将会增加土壤向大气中释放二氧化碳的速度，促使土壤碳库中碳的流失。禁牧样地的土壤呼吸速率在刚禁牧时先迅速增大，随着禁牧时间的延长土壤呼吸速率将会下降。此外，与其它放牧模式相比，冬季放牧将高寒草甸群落土壤呼吸速率在生长季达到最大值的时间明显向后推迟。不同放牧模式下高寒草甸群落土壤呼吸的Q10 值大小顺序为：禁牧一年群落>冬季放牧群落>禁牧三年群落>夏季放牧群落>自由放牧群落。 4）基于呼吸室技术的观测方法中，测量前的剪草处理可以明显改变该地区高寒草甸群落的土壤温度和土壤呼吸速率。在生长季，剪草处理将使土壤呼吸速率的瞬时响应增加90%左右。由于剪草处理明显增加了剪草样方白天的土壤温度，而土壤温度与土壤呼吸之间存在着极显著的指数相关关系，因而剪草处理导致土壤呼吸速率迅速增加。因此，在高寒地区基于呼吸室技术观测的土壤呼吸应当进行校正。 综上所述，川西北高寒草甸生态系统土壤呼吸速率在生长季较高，而在非生长季较低。土壤温度是影响该地区土壤呼吸的最主要环境因子。在实验观测期，草丘区土壤呼吸速率显著高于对照区的，是对照区土壤呼吸速率的近2倍。由于测量前的剪草处理可以明显改变待测点的土壤呼吸速率，因此，应对在高寒地区基于呼吸室技术观测的土壤呼吸进行校正。 Soil respiration is the second largest component (less than plant phtotosynthesis) of carbon dioxide flux between terrestrial ecosystems and the atmosphere. A minor change in soil respiration rate can significantly slow down or accelerate the increase of atmospheric CO2 concentration that is closely related to global climatic change. In turn, the change in the flux direction and rate of soil respiration may indicate the elasticity and stability of ecosystems to global changes and human disturbance. However, soil respiration is still an ecosystem process that has been poorly understood. Grassland ecosystem is an important component of the terrestrial ecosystem. Accurately estimating the CO2 flux from soil to atmosphere in situ is the key to evaluating the carbon resource and sink regionally or globally. Despite of extensive studies on the temperate and tropic grasslands, the soil respiration of alpine meadows has not substantially been measured. In the current study, soil respiration was measured for an annual cycle from April, 2008 to April, 2009 for the alpine meadow in northwestern Sichuan Province of China to determine the seasonal variation of soil respiration for the typical plant communities. The results are shown as follows: 1) Large seasonal variation of soil respiration was observed in the alpine meadow. The rate of soil respiration was the greatest (6.07～9.30μmolCO2¡m-2¡s-1) in June and the smallest (0.16 ～ 0.49μmolCO2¡m-2¡s-1) occurred from December to February in the non-growing season. The total emission of soil CO2 was estimated as 3963～5730 gCO2¡m-2¡yr-1, 85%~90% of which was released during the growing season, and 10%~15% during the non-growing season which was slightly less than the ratio of winter and annual CO2 flux from soil. Temperature, particularly the soil temperature, was the major environmental factor regulating the soil respiration. Significant and positive relationships were not found between soil respiration and soil moisture and between soil respiration and plant above-ground biomass, but excessive soil water content would decrease in the rate of soil respiration. 2) The rate of soil respiration in grass hummock communities was up to 16.77μmolCO2¡m-2¡s-1, which was about twice as great as in the controls (communities located in low and even sites). Considering the large proportion (about 30% on average) of hummock area in the meadow, it can be concluded that the hummocks played an important role in the carbon cycling of the study ecosystem. 3) Grazing patterns affected the flux of CO2 emission and the temperature sensitivity of soil respiration (Q10) in the alpine meadow. Grazing during growing season increased the rate of soil respiration. The rate of soil respiration increased significantly immediately after the alpine meadow being fenced, but thereafter decreased. In addition, grazing in winter delayed the peak respiration rate relative to the non-grazing mode. The Q10 value was the largest in the non-grazed area for one year, and next came the area with grazing in winter, followed by the non-grazed area for three years, the area with grazing in summer, and the non-limited grazed area. 4) In the chamber-based techniques, clipping manipulation before each measurement increased the transient rate of soil respiration by about 90% in the summer of the alpine meadow. As increase in soil temperature at daytime in the clipped plots by clipping and the exponential relationship between soil respiration and temperature, clipping manipulation led to increase in the rate of soil respiration. This suggested that a correction should be done for the techniques if employed in alpine and cold regions. In summary, the rate of soil respiration in the alpine meadow was the greatest in June and the smallest occurred from ecember to February in the non-growing season. Soil temperature was the major environmental factor regulating the soil respiration. The rate of soil respiration in grass hummock communities was up to 16.77μmolCO2¡m-2¡s-1, which was about twice as great as in the controls. A correction should be done for the techniques if employed in alpine and cold regions, because of the effect of clipping manipulation on soil temperature and respiration