关键词: 多样性, T-RFLP, ppk1, 水库沉积物, 群落结构

Abstract: To accurately estimate whole slope-scale or watershedscale forest transpiration from sap flow measurements at plot scale, especially in semi-humid areas with significant spatial variation in environmental factors, it is necessary to examine differences in sap flow at different slope locations and the driving factors. Based on a field study in five larch (Larix principis-rupprechtii) plantation plots established respectively at the upper (P₁), middle upper (P₂), middle (P₃), middle lower (P₄) and lower (P₅) positions in a southeastfacing slope in Xiangshuihe watershed in the semi-humid region of Liupan Mountains of Ningxia, the sap flow velocity in trees as well as the soil water potential and meteorological factors were simultaneously monitored. During the study period, significant differences among slope positions were found in the daily average sap flow velocity (J_s, mL·cm^-2·min^-1), with an order of P₂ (0.0397)>P₄ (0.0368)>P₃ (0.0325)>P₅ (0.0311)>P₁ (0.0278). The J_s was significantly correlated with the maximum air temperature (T_max), solar radiation intensity (SR), water vapor pressure deficit (VPD), potential evapotranspiration (PET), soil water potential (Ψ), volumetric soil water content (VSM), but was negatively correlated with minimum air temperature (T_min), air relative humidity (RH) and precipitation (P). The correlation coefficients were significantly different among different slope positions. There were significant differences in the degree of J_s responses to environmental factors among slope positions. Decreasing trends in the response of J_s to the daily average air temperature (T), RH, VPD, PET and Ψ  were found by the envelope line analysis with the decreasing slope position. In contrast, increasing trends in the response of J_s to the SR and daily average volumetric soil water content (VSM) were observed. Results from regression analysis and partial correlation analysis showed that variations of J_s in different slope positions were mainly controlled by PETand VPD. The J_s at the upper slope positions was strongly influenced by Ψ and T_max, while that at the lower slope positions was largely affected by SR and VSM. In general, the slopedifference in J_s was caused by soil moisture and meteorological factors. Therefore, it is needed to consider jointly the changes of soil moisture and meteorological factors among the slope positions when we estimate the stand transpiration at the slope scale or watershed scale by upscaling the J_s.

Key words: community structure., T-RFLP, reservoir sediment, diversity, ppk1