Diurnal variations of stand transpiration of Larix principis-rupprechtii forest and its response to environmental factors in Liupan Mountains of northwestern China

    Objective   This paper aims to examine the diurnal variations of stand transpiration and its response to environmental factors in a Larix principis-rupprechtii plantation, and to further explain the regulation mechanism of environmental conditions to the stand transpiration in short time scale.

    Method   A field experiment was conducted in a Larix principis-rupprechtii plantation stand located in the Xiangshuihe Watershed within the Natural Reserve of Liupan Mountains, Ningxia of northwestern China. The sap flow of sample trees was continuously monitored in the growth season (from May to October) in 2018. The meteorological and soil moisture conditions were continuously measured simultaneously. The response of hourly stand transpiration (T) to environmental factors was analyzed, and a T model coupling the effects of multiple influencing factors was established.

    Result   It was shown that: (1) the response of T to solar radiation (R_s) and vapor pressure deficit (VPD) followed a binomial equation. The T firstly increased with rising R_s and VPD till to peak at the thresholds of 666.7 W/m² and 1.86 kPa, and then gradually decreased afterwards. (2) The T response to relative extractable water (REW) of the 0−60 cm soil layer followed a saturated exponential growth function. The T increased firstly with rising REW, and then became stable when the REW was above 0.3. (3) The response functions of T each individual driving factors (R_s, VPD and REW) were determined using the upper boundary line method, and then were coupled to form the frame of T model. Thereafter, the observed data of odd-days and even-days were used for model fitting and validation. The fitted model is: T = (− 6.347 0 × 10^{− 5}\begindocumentR_\rms^2\enddocument− 0.637 0R_s − 208.734 8) × (− 0.003 2VPD² + 0.013 8VPD + 0.001 7) × (− 0.008 1 − 0.004 6(1 − exp(− 12.469 6REW))). The model was well calibrated (R² = 0.74, Nash coefficient (NSE) = 0.82) and validated (R² = 0.77, NSE = 0.84).

    Conclusion   The T variation can be well predict by the T model coupling the effects of solar radiation, vapor pressure deficit, and soil moisture conditions. This model can accurately predict the variation of T of Larix principis-rupprechtii plantation under changing environment. Meanwhile, the model establishment approach used here can be a reference for developing the stand transpiration model in other regions and for other tree species.