Chin J Plant Ecol ›› 2018, Vol. 42 ›› Issue (12): 1179-1191.

• Research Articles •

### Impact of environmental factors on the decoupling coefficient and the estimation of canopy stomatal conductance for ever-green broad-leaved tree species

ZHANG Zhen-Zhen1,*(),ZHAO Ping2,ZHAO Xiu-Hua2,ZHANG Jin-Xiu1,ZHU Li-Wei2,OUYANG Lei2,ZHANG Xiao-Yan1

1. 1 School of Geography and Environmental Sciences, Zhejiang Normal University, Jinhua, Zhejiang 231004, China
2 South China Botanical Garden, Chinese Academy of Sciences, Guangzhou 510650, China
• Received:2018-07-30 Revised:2018-10-23 Online:2019-04-04 Published:2018-12-20
• Contact: Zhen-Zhen ZHANG E-mail:zhangzhen@zjnu.cn
• Supported by:
Supported by the National Natural Science Foundation of China(41630752);Supported by the National Natural Science Foundation of China(41701226);Supported by the National Natural Science Foundation of China(41030638)

Abstract:

Aims Accurate simulation of canopy stomatal conductance (GS) is quite important for the assessment of regional evapotranspiration.

Methods In this study, two planted broad-leaved tree species, Eucalyptus urophylla (exotic species) and Schima superba (native species), were chosen to estimate their GS with two different methods of K?stner (GS1) and inversed Penman-Monteith equation (GS2). The effect of environmental factors on canopy decoupling coefficient (Ω) was evaluated before they were adopted to assess the reasonability of GSsimulated by the two methods.

Important findings Results showed that the GS of the two tree species was well coupled with meteorological conditions (Ω = 0.10 ± 0.03 for E. urophylla and 0.17 ± 0.03 for S. superba). Principal component analysis showed that photosynthetically active radiation (PAR) and vapor pressure deficit (D) significantly dominated the variations of Ω, while the effect of wind speed (u) was very weak. Multivariate correlation analysis also found weak relations between those environmental factors and Ω. Boundary line analysis revealed that the increase of D and PAR would eventually force Ω approaching a constant value as determined by tree species (S. superba ≈ 0.20, E. urophylla ≈ 0.05), while Ω decreases exponentially with the increase of u. Compared with S. superba, E. urophylla has higher GS. The annual averages GS2 of E. urophylla and S. superba were (33.42 ± 9.37) mmol·m-2·s-1 and (23.40 ± 2.03) mmol·m-2·s-1, respectively. Linear fitting showed that the GS2/GS1 ratio of E. urophylla and S. superba was 0.92 (R2 ≈ 0.70) and 0.98 (R2 ≈ 0.76), respectively, implying the overestimated canopy stomatal conductance for GS1 (p < 0.01). In addition, the ratio of the sensitivity of canopy stomatal conductance to vapor pressure deficit to stomatal conductance at D = 1 kPa (GSiref) for GS1 and GS2 is closely related to Ω. Based on the estimation, GS1 was relatively reliable when Ω = 0.05-0.15 (83.1% of all the data) and 0.10-0.20 (47.8% of all the data) for E. urophylla and S. superba.