Welcome to Chinese Journal of Applied Ecology! Today is Share:

Chinese Journal of Applied Ecology ›› 2017, Vol. 28 ›› Issue (3): 983-991.doi: 10.13287/j.1001-9332.201703.023

• CONTENTS • Previous Articles     Next Articles

Assessing the surface heating effect of the open-path analyzer of an eddy covariance system with fine-wire thermocouples

WANG Xing-chang1, WANG Chuan-kuan1*, LIU Fan1, ZHEN Xiao-jie2   

  1. 1Center for Ecological Research, Northeast Forestry University, Harbin 150040, China
    2Beijing Techno Solutions Limited, Beijing 100089, China
  • Received:2016-08-23 Published:2017-03-18
  • Contact: *E-mail: wangck-cf@nefu.edu.cn
  • Supported by:
    This work was supported by the National Natural Science Foundation of China (41503071), the Fundamental Research Funds for the Central Universities (2572016BA03), and the Program for Changjiang Scholars and Innovative Research Team of Ministry of Education (IRT_15R09)

Abstract: The open-path CO2/H2O analyzer that is currently widely used in the eddy covariance (EC) system possesses a surface heating effect, which may influence the accuracy of the eddy-flux measurements. Taking the open-path EC system in the temperate deciduous forest of the Maoershan Ecosystem Research Station as a case, we assessed the heating effect of the open-path CO2/H2O ana-lyzer (Li-7500) with fine-wire thermocouples and tested the applicability of Burba’s equations. The results indicated that between April 23 and May 28 in 2016, the daytime air temperature in the mid-optical-path of the Li-7500 was on average 0.2 ℃ higher than the ambient air temperature, while the nocturnal temperature was almost equal, with the peak of the heating often occurring in the morning transition period and at noon or in the early afternoon. The diurnal variation in the heating effect on the sensible heat flux (HS,HE) estimated from the simple linear-fitting models in the fourth method by Burba (BurbaLF) was overall a rectangle-like wave, whereas that from the multiple regression models (BurbaMR) showed a weak mono-peak shape. Compared with direct mea-surements (K079) with the K-type fine-wire thermocouples (with a 0.079 mm diameter of single bare wire) and the HS,HE modeled with two fine-wire thermocouples (K079Model), the estimated HS,HE with the BurbaLF and BurbaMR varied little in the daytime with a lower but wider daytime-peak (20 W·m-2), a sharp transition between daytime and nighttime, and a typical nocturnal va-lue of < 5 W·m-2. The peak values of HS,HE estimated from the K079 and K079Model could be higher than 40 W·m-2, and the values at night fluctuated around zero. The sensible heat flux in the optical-path of the Li-7500 was increased by 13.6% during the daytime. The value of the daytime heating effect on the CO2 turbulent flux (Fc,HC) estimated from the K079 and K079Model was averaged about 0.5 mg CO2·m-2·s-1, twice of the value estimated previously. Compared with the direct measurements, the Burba equations underestimated the Fc,HC around the noon due to the inadequate HS,HE, but overestimated the Fc,HC in the morning, the late afternoon and the night because of over-corrections of HS,HE, resulting in an overall overestimate of Fc,HC. These findings validated that both K079 and K079Model methods can be used to estimate the Fc,HC.