Abstract:

Objective: As a key component of forest evapotranspiration, tree transpiration is primarily regulated by environmental factors such as meteorological conditions and soil moisture. However, its response patterns to these factors exhibit significant differences across temporal scales. A thorough understanding of these dynamics can provide support for the scaling-up estimation of field measurements and the precise regulation of forest-water relationships. Method: A continuous monitoring of stem sap flow dynamics was conducted in a Larix gmelinii var. principis-rupprechtii plantation in the Xiangshuihe small watershed of the Liupan Mountains, Ningxia, during the 2023 growing season (May to October). Concurrently, environmental factors and canopy leaf area index (LAI) were observed. The responses of tree transpiration to the main influencing factors were analyzed at hourly, daily and monthly scales. A multi-factor transpiration model was established, and the relative contribution rates of each factor to tree transpiration were further quantified. Result: 1) Both daily and hourly tree transpiration rate showed a saturated exponential relationship with solar radiation (R_s) and vapor pressure deficit (VPD). However, the threshold values of R_s and VPD at which daily transpiration began to slow were approximately one-third of those at the hourly scale. Additionally, daily tree transpiration exhibited a saturated exponential relationship with relative extractable soil water (REW) and an “S”-shaped logistic curve in response to LAI. 2) By coupling these response functions and calibrating them with measured data, the hourly scale transpiration model was able to predict hourly transpiration of trees by meteorological factors (R²=0.80), and the daily transpiration model was able to predict daily transpiration of trees by meteorological factors, REW and LAI (R²=0.83). 3) Analysis of transpiration contribution rates from hourly and daily forest transpiration models revealed distinct patterns: On sunny days, VPD was the dominant factor affecting hourly transpiration variations. During cloudy and overcast conditions, VPD was the primary factor affecting hourly transpiration variations at night and morning, while R_s became the dominant factor affecting hourly transpiration variations from noon to afternoon. On rainy days, VPD was the main contributor at night and R_s during daytime. At the monthly scale, LAI was the dominant factor affecting transpiration changes at the beginning (May), end (October), and peak growth season (July-August) of the growing season, whereas meteorological factors took the lead in June and September. The relative contribution rate of REW remained relatively low throughout all months. Conclusion: This study establishes a multi-factor coupling model integrating hourly tree transpiration responses (R_s and VPD) with daily transpiration responses (R_s, VPD, REW, and LAI). The model effectively predicts hourly and daily transpiration patterns, revealing that the combined effects of the same factors on transpiration do not simply add up across time scales. These findings contribute to a deeper understanding of the temporal scale differences in the response of forest transpiration to dominant factors and provide a theoretical basis for precise forest-water management. For example, in small-scale plantation tending practices, the control of different dominant environmental factors should be considered based on the target temporal scale.

Key words: Larix gmelinii var. principis-rupprechtii, biotic environmental factors, forest transpiration, temporal scale, coupled model