주간에 두 타워로부터 관측된 에디 공분산 자료의 확률 오차의 추정
Estimation of the Random Error of Eddy Covariance Data from Two Towers during Daytime
1
조회
0
다운로드
4
인용
0
참고문헌
키워드
초록
We have examined the random error of eddy covariance (EC) measurements on the basis of two-tower approach during daytime. Two EC towers were placed on the grassland with different vegetation density near Gumi-weir. We calculated the random error using three different methods. The first method (M1) is two-tower method suggested by Hollinger and Richardson (2005) where random error is based on differences between simultaneous flux measurements from two towers in very similar environmental conditions. The second one (M2) is suggested by Kessomkiat et al. (2013), which is extended procedure to estimate random error of EC data for two towers in more heterogeneous environmental conditions. They removed systematic flux difference due to the energy balance deficit and evaporative fraction difference between two sites before determining the random error of fluxes using M1 method. Here, we introduce the third method (M3) where we additionally removed systematic flux difference due to available energy difference between two sites. Compared to M1 and M2 methods, application of M3 method results in more symmetric random error distribution. The magnitude of estimated random error is smallest when using M3 method because application of M3 method results in the least systematic flux difference between two sites among three methods. An empirical formula of random error is developed as a function of flux magnitude, wind speed and measurement height for use in single tower sites near Nakdong River. This study suggests that correcting available energy difference between two sites is also required for calculating the random error of EC data from two towers at heterogeneous site where vegetation density is low.
Abstract
We have examined the random error of eddy covariance (EC) measurements on the basis of two-tower approach during daytime. Two EC towers were placed on the grassland with different vegetation density near Gumi-weir. We calculated the random error using three different methods. The first method (M1) is two-tower method suggested by Hollinger and Richardson (2005) where random error is based on differences between simultaneous flux measurements from two towers in very similar environmental conditions. The second one (M2) is suggested by Kessomkiat et al. (2013), which is extended procedure to estimate random error of EC data for two towers in more heterogeneous environmental conditions. They removed systematic flux difference due to the energy balance deficit and evaporative fraction difference between two sites before determining the random error of fluxes using M1 method. Here, we introduce the third method (M3) where we additionally removed systematic flux difference due to available energy difference between two sites. Compared to M1 and M2 methods, application of M3 method results in more symmetric random error distribution. The magnitude of estimated random error is smallest when using M3 method because application of M3 method results in the least systematic flux difference between two sites among three methods. An empirical formula of random error is developed as a function of flux magnitude, wind speed and measurement height for use in single tower sites near Nakdong River. This study suggests that correcting available energy difference between two sites is also required for calculating the random error of EC data from two towers at heterogeneous site where vegetation density is low.
참고문헌 (23)
- Aubinet, M., "Eddy covari ance. A practical guide to measurement and data analysis", pp.438~ (2012), Springer
- Seon-Young Choi, "Analysis of Local Wind Induced by Surface Heterogeneity and Sloping Terrain Near Nakdong River", 한국기상학회지, 51(3), pp.249~257 (2015), 한국기상학회
- Finkelstein, P. L., "Sampling error in eddy correlation flux measurements", J. Gephys. Res, 106, pp.3503~3509 (2001)
- Foken, T., "The energy balance closure problem: an overview", Ecol. Appl, 18, pp.1351~1367 (2008)
- Hollinger, D. Y., "Uncertainty in eddy covariance measurements and its application to physiological models", Tree Physio, 25, pp.873~885 (2005)
- Hu, Q., "Quality control for USDA NRCS SM-ST Network soil temperatures: A method and a dataset", J. Appl. Meteor, 41, pp.607~619 (2002)
- 강금아, "낙동강 강정고령보에서 관측된 플럭스의 지표면에 따른 특성 비교", 한국지구과학회지, 34(6), pp.561~574 (2013), 한국지구과학회
- Kessomkiat, W., "Estimating random errors of eddy covariance data: An extended two-tower approach", Agric. Forest Meteorol, 171, pp.203~219 (2013)
- 이영희, "낙동강 유역에서 관측된 에디 공분산 자료의 품질 관리 및 플럭스 특성", 대기, 23(3), pp.307~320 (2013), 한국기상학회[DOI]
- Lenschow, D. H., "How long is long enough when measuring fluxes and other turbulence statistics?", J. Atmos. Oceanic. Technol, 11, pp.661~673 (1994)
- Mahrt, L., "Flux sampling errors from aircraft and towers", J. Atmos. Oceanic Technol, 11, pp.661~673 (1998)
- Mann, J., "Errors in airborne flux measurements", J. Geophys. Res, 99, pp.14519~14526 (1994)
- Massman, W. J., "Eddy covariance flux corrections and uncertainties in long term studies of carbon and energy", Agric. Forest Meteorol., 113, pp.121~144 (2002)
- Mauder, M., "A strategy for quality and uncertainty assessment of long-term eddy-covariance measurements", Agric. Forest Meteorol, 169, pp.122~135 (2013)
- Hydrometeorology research for the test-bed region (I), "National Institute of Meteorological Research", National Institute of Meteorological Research, 625~ (2011)
- Hydrometeorology research for the test-bed region (II), "National Institute of Meteorological Research", National Institute of Meteorological Research, 690~ (2012)
- Technical Description of the Community Land Model (CLM), "Oleson, K. W.", NCAR, 173~, pp.NCAR (2004), TN-461+STR
- Richardson, A. D., "A multi-site analysis of random error in tower-based measurements of carbon and energy fluxes", Agric. Forest Meteorol, 136, pp.1~18 (2006)
- Schmid, H. P., "Source areas for scalar and scalar fluxes", Bound.-Layer Meteor, 67, pp.293~318 (1994)
- Taylor, J. R., "An introduction to error analysis", pp.327~ (1997), University Science Books
- Twine, T. E., "Correcting eddy-covariance flux underestimates over a glassland", Agric. Forest Meteorol, 103, pp.279~300 (2000)
- Webb, E. K., "Correction of flux measurements for density effects due to heat and water vapor transfer", Quart. J. Roy. Meteor. Soc, 106, pp.85~100 (1980)
- Wilks, D. S., "Statistical methods in the atmospheric sciences", pp.467~ (1995), Academic Press