陇中地区典型玉米农田生态系统呼吸作用对温湿度的响应研究

The Response of Respiration of Typical Maize Farmland Ecosystem to Temperature and Humidity in Longzhong Area

农田生态系统碳通量变化主要由作物光合作用和农田的呼吸作用引起。在陆地生态系统中,准确评估玉米农田生态系统的呼吸作用产生的碳排放量对全球碳通量的研究意义重大。使用甘肃省会宁县大山川村(属陇中半干旱区)玉米农田生态系统在2020年11月至2021年10月的涡度和土壤水分及温度监测数据,分析了玉米农田生态系统呼吸速率(R_(eco))的日平均变化和呼吸敏感性指数(Q_(10))在生长季内的月平均变化,探讨R_(eco)对温度的响应,并用不同模型进行拟合评估。陇中半干旱区玉米农田生态系统呼吸速率在年内呈单峰变化,日均最大呼吸速率11.49μmol/(m^(2)·s),相比呼吸速率对空气温度(Ta)与不同深度土壤温度(Ts)的响应,深度20 cm处的土壤温度(Ts_(20))对该地区R_(eco)全年拟合效果最佳。Q_(10)指数在生长季内表现出单峰型变化,最大值出现在8月(5.52μmol/(m^(2)·s)。通过各个模型拟合得出陇中地区玉米农田生态系统年呼吸碳排放量介于863.71~990.18 g C/m^(2),选择拟合效果最好的模型(Q_(10)模型),在考虑参数时变性对Q_(10)模型优化后,Q_(10)模型拟合精度得到大幅度提升,年碳排放量估计值由868.57 g C/m^(2)提高至1094.48 g C/m^(2),更加接近观测值(1110.28 g C/m^(2))。R_(eco)在土壤水分低于0.16 m^(3)/m^(3)时(出现在12-2月)受到抑制。

Carbon flux changes in farmland ecosystems are mainly caused by crop photosynthesis and respiration in farmland.Accurate assessment of carbon emissions generated from the respiration of maize farmland ecosystems is significant for studying global carbon fluxes in terrestrial ecosystems.This study collected eddy covariance,soil moisture,and air temperature measurements from November 2020 to October 2021 in the maize farmland ecosystem,which is located in Dashanchuan Village,Huining County,Gansu Province.We analyzed the variation of daily ecosystem respiration rate(R_(eco))and monthly respiratory sensitivity index,explored the response of R_(eco)to temperature,and evaluated different models for R_(eco)simulation.The results showed that the carbon emission rate of maize farmland ecosystem in the semi-arid region was unimodal,and the average daily maximum respiration rate was 11.49μmol m^(-2)·s^(-1).Compared with the response of respiration rate to air temperature(Ta)and soil temperature(Ts)at different depths,soil temperature(Ts_(20))at 20cm depth has the best yearround fitting effect on R_(eco)in this region.Q_(10)showed unimodal changes during the growing season,with the maximum value occurring in August(5.52μmol·m^(-2)·s^(-1)).By fitting each model,the annual respiratory carbon emission of maize farmland ecosystem in the Longzhong region ranged from 863.71 to 990.18 g C m^(-2).The model that exhibited the most effective fitting(Q_(10)model)was chosen.After the Q_(10)model was optimized considering the parameter time-variability,the fitting accuracy of the Q_(10)model was greatly improved,and the estimated annual carbon emission value increased from 868.57 g C·m^(-2)to 1094.48 g C·m^(-2),which was closer to the observed value(1110.28 g C·m^(-2)).R_(eco)was inhibited when soil moisture was lower than 0.16 m^(3)/m^(3)(during December to February).