Canopy transpiration of Larix principis-rupprechtii plantation and its impact factors in diffe-rent slope locations at the south side of Liupan Mountains, China.

Abstract

Based on a continuous field observation in the Larix principis-rupprechtii plantation plots, located at the upper (P1), middle-upper (P2), middle (P3), middle-lower (P4), and lower (P5) positions, in a southeast-facing slope of the Xiangshuihe watershed of Liupan Mountains, China, the stem sap flow was observed with the thermal diffusivity probe method. The soil water potential and meteorological factors were monitored from May to October, 2014. We found significant differences among slope positions in the daily forest transpiration (Tr, mm·d-1), with an order of P2 (0.975)＞P4 (0.876)＞P3 (0.726)＞P1 (0.653)＞P5 (0.628). Tr was significantly positively correlated with the daily maximum temperature (Tmax), daily mean solar radiation (SR), daily mean saturated vapor pressure deficit (VPD), potential evapotranspiration (PET), and daily mean soil water potential (Ψ), but negatively correlated with the daily mean air relative humidity (RH), daily precipitation (P), and daily minimum temperature (Tmin). According to the upper boundary line ana-lysis, significant differences were found in the degree of Tr responding to each single environmental factor among slope positions. The degree of its responses gradually decreased for average daily air temperature (T), RH, VPD, PET and Ψ, whereas increased for the SR and daily average volumetric soil water content (VSM) from the upper position to the lower. Results from regression and partial correlation analysis showed that variation of Tr was mainly controlled by VPD, PET and RH in different slope positions. Tr was also strongly affected by Ψ and T in the upper-slope positions and by SR, Tmin and VSM at the lower-slope positions. Generally, the Tr difference among slope positions was a consequence of joint contributions of soil water and meteorological factors. It is necessary to consider the changes of soil water and meteorological factors in different positions along the slope when examining slope-scale or watershed-scale forest transpiration with sap flow estimated from xylem sap flux density measurements of a particular plot.