Performance evaluation of an integrated open-path eddy covariance system in a cold desert environment

Size: px

Start display at page:

Download "Performance evaluation of an integrated open-path eddy covariance system in a cold desert environment"

Norah Holland
5 years ago
Views:

Performance evaluation of an integrated open-path eddy covariance system in a cold desert environment Wei Wang, Jiaping Xu, Yunqiu Gao, Ivan Bogoev, Jian Cui, Lichen Deng, Cheng Hu, Cheng Liu,

1 Performance evaluation of an integrated open-path eddy covariance system in a cold desert environment Wei Wang, Jiaping Xu, Yunqiu Gao, Ivan Bogoev, Jian Cui, Lichen Deng, Cheng Hu, Cheng Liu, Shoudong Liu, Jing Shen, Xiaomin Sun, Liming Wang, Wei Xiao, Guofu Yuan, Xuhui Lee wangwnuist@163.com Joint Conference of AsiaFlux Workshop 2017 and the 15 th Anniversary Celebration of ChinaFLUX August, 16-19, 2017 Beijing, China 1

2 Outlines 2 Unreasonable CO 2 uptake by open-path eddy covariance (EC) system. Self-heating, spectroscopic effect, bias in CO 2 density. Lessons learnt from desert IRGASON experiment. Wind statistics, heat fluxes, CO 2 flux. Meta-analysis across 64 FLUXNET sites. Results

3 Typical open-path EC system 3 LI-7500/7500A+Gill LI-7500/7500A+CSAT3 EC150 IRGASON

4 4 Physiologically unreasonable CO 2 uptake observed with open-path EC Wang et al., 2016, JTECH

5 5 Three possible explanations for unreasonable CO 2 uptake Self-heating: additional heat generated by the instrument electronics or by the solar loading (Burba et al., 2008), such as LI Spectroscopic effects: attenuation of temperature at high frequencies and spectroscopic cross-sensitivity (Detto et al., 2011; McDermitt et al., 2011; Bogoev et al., 2014), such as EC150 and IRGASON. Errors propagation through density correction procedure by bias in CO 2 density (Serrano-Ortiz et al., 2008; Fratini et al., 2014), all IRGAs.

6 Self-heating of open-path EC 6 First interval: low wind, U <0, U w <0, then w >0, surface warming, air expansion, CO 2 <0, then w CO 2 <0, artificial CO 2 uptake. Second interval: strong wind, U >0, U w <0, then w <0, less surface warming, smaller air expansion, CO 2 0, no artificial CO 2 uptake. Burba and Anderson, 2010

7 7 Theoretical consideration of selfheating WPL density correction algorithm (Webb et al., 1980) F c = w ρ c + ρ c തTC p ρ a 1 + ρ vm a ρ a M v H + ρ cm a ρ a M v λe F c = w ρ c + ρ c തTC p ρ a 1 + ρ vm a ρ a M v H + δh + ρ cm a ρ a M v λe b = F c = F c + bh ρ c തTC p ρ a 1 + ρ vm a ρ a M v δh H δh H = 0.14 (Burba et al., 2008), b is μmol m -2 s -1 per W m -2. Wang, Lee, Lin, et al., in review

8 8 Spectroscopic effect due to highfrequency temperature attenuation A 2 N SiiCL 1 exp i i d i,, S f T P f T P 1 i 2 i PT, 1/2 P T0 0 P0 T Jamieson et al F F H c c Bogoev et al., 2015

9 ҧ 9 Theoretical consideration of spectroscopic effect α = α + α, α = α o P P o T o T 1/2 α = 1 2 α T T A = ሚA α α = ሚA 1 2 T T, A = ሚA (1 + 1 T 2 T ) C = ሚC ( T T ) F c = F c + bh b = 1 1 C 2 ρc p തT μmol m 2 s 1 b μmol m 2 s 1 F c F c ρc p ҧ C തT H Wang et al., 2016, JTECH

10 10 Universal negative linear relationship between F c and H Helbig, et al., AFM, 2016 Significant and negative linear relationship existed at 37 sites across 64 FLUXNET sites. Wang, Lee, Lin, et al., in review

11 11 Theoretical consideration of bias CO 2 density F c = w ρ c + = F c + bh ρ c + δρ c തTC p ρ a 1 + ρ vm a ρ a M v H + ρ c + δρ c M a ρ a M v λe b = δρ c തTC p ρ a 1 + ρ vm a ρ a M v δρ c ρ c 0.05 δρ c ρ c Across 64 sites, δρ c ρ c m -2 s -1 per W m -2. = -5%, b is about μmol Wang, Lee, Lin, et al, in review

12 12 Hypothesis Integrating the infrared gas analyzer s sensing heads into the sensing volume of the sonic anemometer has negligible effects on dynamic flows of the IRGASON. Inadequate spectroscopic correction by slow response air temperature measurement partly contribute to ecologically unreasonable CO 2 uptake with IRGASON.

13 13 Site (cold arid desert) and Instrumentation

14 Period Date Populus trees Winter Dec.16, 2013- Jan.3, 2014, Spring Mar.12, 2014- Apr.

14 14 Period Date Populus trees Winter Dec.16, Jan.3, 2014, Spring Mar.12, Apr.13, 2014 dormant season flowering stage T a ( o C) ρ v (ppm) K (W m -2 ) Albedo

15 Wind statistics 15 Wang et al., 2016, JTECH

16 The Monin-Obukhov scaling relationships 16 u u w w 1/ Wang et al., 2016, JTECH Garratt 1992; Kaimal and Finnigan 1994

17 Spectral and cospectral analysis 17 Wang et al., 2016, JTECH

18 Sensible and latent heat flux 18 R n H λe Wang et al., 2016, JTECH

19 19 CO 2 flux time series F F H c c Winter mean F c was μmol m -2 s -1 and μmol m -2 s -1 for the separated EC system and the IRGASON, respectively. Wang et al., 2016, JTECH

20 Diurnal composition of CO 2 flux in winter 20 After applying correction for spectroscopic effect, the wintertime IRGASON CO 2 flux became physiologically reasonable (mean value μmol m -2 s -1 ). Wang et al., 2016, JTECH

21 21 Comparison among three gas analyzer types The slope parameter b (μmol m 2 s 1 per W m 2 ; gray bars) and the R 2 value (white bars) of the linear regression between wintertime F c and H. Error bars are ±1 standard error. Wang, Lee, Lin, et al, in review

22 22 Comparison among geographic regions Wang, Lee, Lin, et al, in review

23 Conclusions 23 Integrating an IRGA into measuring volume of the IRGASON sonic anemometer had negligible effects on its wind statistics. Both EC systems observed negative CO 2 fluxes (-1.6 μmol m -2 s -1 ) in the daytime during the winter experiment. Sensor selfheating was ruled out as the cause of the apparent uptake flux. After applying correction for spectroscopic effect, the wintertime IRGASON CO 2 flux became physiologically reasonable (mean value μmol m -2 s -1 ). The negative linear relationship between observed CO 2 flux and sensible heat flux is universal and was confirmed by a metaanalysis of open-path EC data from 64 FLUXNET sites.

24 Related papers 24 Bogoev, I., 2014: Improved eddy flux measurements by open-path gas analyzer and sonic anemometer co-location. Geophysical Research Abstracts, Vol. 16, Abstract EGU Wang, W., et al., 2016: Performance evaluation of an integrated open-path eddy covariance system in a cold desert environment. Journal of Atmospheric and Oceanic Technology, 33(11): Helbig, M., et al., 2016: Addressing a systematic bias in carbon dioxide flux measurements with the EC150 and the IRGASON open-path gas analyzers. Agricultural and Forest Meteorology, 228: Wang, L.M., et al., 2017: A meta-analysis of eddy covariance observations of apparent CO 2 uptake in cold conditions in the FLUXNET network. In review.

25 Data sharing 25 Thanks for your attention!

Temporal and spatial variations in radiation and energy fluxes across Lake Taihu

Temporal and spatial variations in radiation and energy fluxes across Lake Taihu Wang Wei YNCenter Video Conference May 10, 2012 Outline 1. Motivation 2. Hypothesis 3. Methodology 4. Preliminary results