Eddy-covariance measurements (at 82 m and at 3 m height)

82 m system

Measurements of the vertical flux of CO2 and H2O by eddy covariance 82 m above the ground began in April 1997. This was facilitated by mounting to the 4.4 m long instrument arm an ultrasonic anemometer (Gill Solent Enhanced), an aspirated thermocouple (50 cm from the anemometer) for fast-response temperature measurements, and an additional air sampling tube for CO2 and H2O measurements.

A schematic diagram of the air sampling system used to determine the CO2 and H2O mixing ratios for flux calculations.

Measurements of CO2 and H2O are made at 4 Hz using a fast response IRGA (Li-Cor model LI-6262). Air is pumped (dual diaphragm GAST model DAA V175 ED MFG Corp.) through the sampling tube and analyzer at about 15 l/min, producing a pressure drop of approximately 45 kPa. Pressure fluctuations generated by the pump are damped by means of a 6 l buffer volume. Flow in the sampling tube is turbulent. Immediately behind the sample cell of the analyzer we measure pressure (MKS Instruments model 122A barotron), temperature and relative humidity (Vaisala HMD20YB). The pressure and temperature data are used to correct the instrument response for variations in these parameters (see below). The humidity data are used to determine the calibration function for water vapor measurements by the Li-Cor 6262.

The CO2 analyzer runs in relative mode. Dry, synthetic air with a CO2 mixing ratio of 330-340 ppm is used as a reference gas (Messer Hungarogáz) and flows at a rate of 5-10 cm3/min through the reference cell of the IRGA. The analog output of the analyzer for CO2 and H2O, as well as the signals of the pressure and temperature/humidity sensors are digitalized by the common A/D-RS232 converter.

A separate data acquisition computer is used to read data from the fast-response instruments (sonic anemometer, thermocouple, and IRGA). The computer communicates with the instruments through two standard serial ports: COM1 receives the data from the sonic anemometer while the two A/D-RS232 devices are controlled via COM2. The data acquisition cycle is triggered by the signal from the sonic anemometer. After reception of a data package (horizontal and vertical wind speed, wind direction and error code) the computer requests data from the aspirated thermocouple, the CO2/H2O analyzer and its accessory sensors (pressure, temperature/humidity), and the CO2 profile analyzer. The position of the multiport valve of the profile system is also determined. As the time of the change of the position of the multiport valve is recorded on the computers of both the profile and eddy correlation systems, the parallel records allow the synchronization of the two independent data acquisition computers. The data acquisition software is written in Turbo Pascal language and runs under DOS 5.0. The eddy correlation system produces data at a rate of about 600 MByte/month, and the data are stored on a CD-R without compression.

The response of the IRGA is calibrated by comparison to ambient CO2 and H2O measurements from the slow-response sensors. A calibrated CO2 measurement is typically obtained every 8 minutes at the 82 m level. Exact synchronization of the signals requires accounting for the time for air to pass through the sampling tubes of the profile and eddy flux systems. The lag time of the profile system is calculated from the measured flow rates. For the eddy correlation system two different lag time values are calculated for H2O and CO2 using a spectral method.

For calibration of the H2O and CO2 response of the IRGA we use the following function:

f(V*p0/p)*(T/T0) + Cr = C

where V is the voltage signal of the analyzer, p0 is the reference pressure (1000 hPa), p is the pressure in the measuring cell, Cr is the mole fraction of CO2 or H2O in the reference gas (zero for H2O, 330-340 mmol/mol for CO2), T0 is the reference temperature (273.15 K), T is the temperature in the measuring cell, and C is the mole fraction of H2O or CO2 measured by the calibrated (profile) instruments. The analyzer is calibrated in terms of H2O and CO2 mole fraction. Linear regression is carried out between (C-Cr)*T0/T and V*p0/p to determine the slope of the response function, f. Occasional outliers are removed interactively, and the resulting linear fits typically show very high correlation. The manufacturer provides a fifth order polynomial calibration curve for the instrument, but the polynomial differs only slightly from linear in the range of interest (360-500 ppm). Calibration values are determined for each 24 hours of measurement. The calibration factors are quite stable in time. Since the fluctuations of H2O and CO2 measured by the Li-Cor 6262 are expressed in terms of mole fraction relative to dry air it is not necessary to perform the corrections for variations of the density of air.

Three dimensional wind vector rotation is applied to the sonic anemometer data. Rotation angles and measured average vertical wind speed are calculated and stored for each hourly period.

A linear trend is removed from each 60 min interval of all data used for eddy flux calculations (wind components, temperature, H2O and CO2). Data values outside ± 4st.dev. are removed. Turbulent fluxes are calculated from covariances of the detrended time series taking into account the delay time of signals determined during the calibration procedure outlined above.

Spectral corrections are applied in order to account for the damping of fluctuations caused by the long air sample tubes, limited sensor response time, sensor line averaging and sensor separation, and sampling. Average losses of the eddy-correlation system calculated from the theoretical considerations are about 10% for CO2 and 9% for H2O. It appears that the water vapour signal suffers from excess spectral degradation compared to the theoretical spectral damping because of the long tubing of the system which causes extra loss of water vapour flux.

Net ecosystem exchange (NEE, the sum of the eddy flux at 82 m, the rate of change of CO2 storage below 82 m) is calculated from the eddy correlation data using the profile data for storage calculations using surface layer similarity theory.

The NEE measurements are representative to a larger region around the tall tower. The tower is surrounded by agricultural fields (mostly crops and fodder of annually changing types) and forest patches. The distribution of vegetation types (60% arable land, 30% forest and woodland, 10% other) within 10 km of the tower is not greatly different from the average for the Western Hungarian Landscape Unit or the whole country (85% of the area is cultivated, 77% of which is agricultural and 23% is forest).


3 m system

The second eddy-covariance system owned by AIST (the former NIRE, Japan) is operated in the garden of the TV tower, above semi-natural grass (Barcza et al., 2003). In the beginning the measuring system was operational during the period of 1999-2000. Due to malfunction of the anemometer the measurements were ceased in 2001. In September 2006 the measurements were resumed when the ultrasonic anemometer was replaced with a new model (Solent Research R3-50, Gill Instruments Ltd., Lymington, United Kingdom).

The sonic anemometer and the inlet tube of the IRGA are mounted on a mast at 3 m elevation above the grass covered surface. The inlet tube is mounted at the elevation of the active center of the ultrasonic anemometer, 25 cm away from it horizontally. Raw voltage data generated by the fast response sensors were collected and digitized by means of a TEAC data logger between 1999 and 2000, while since 2006 the digital signals of the sensors have been recorded by a PC also at 5 Hz frequency.

The area of the site was used as arable land previously and was turned into a grassland around 1990. The dominant species of the grassland are Arrhenatherum elatius, Taraxacum officinale, Poa pratensis, Agropyron repens, Anthoxanthum odoratum, Dactylis glomerata, Holcus lanatus, Briza media and Festuca pratensis. The grass is mowed two times a year, and the mowed grass is taken away from the site and utilized as fodder. Fore up-to-date results please see Nagy et al. (2010).

For more details please see:

Nagy, Z., Barcza, Z., Horváth, L., Balogh, J., Hagyó, A., Káposztás, N., Grosz, B., Machon, A., Pintér, K., 2010. Measurements and estimations of biosphere-atmosphere exchange of greenhouse gases -- Grasslands. In: Atmospheric Greenhouse Gases: The Hungarian Perspective (Ed.: Haszpra, L.). Springer, Dordrecht - Heidelberg - London - New York, pp. 91-119. ISBN 978-90-481-9949-5, e-ISBN 978-90-481-9950-1, doi: 10.1007/978-90-481-9950-1

Tóth, E., Barcza, Z., Birkás, M., Gelybó, Gy., Zsembeli, J., Bottlik, L., Davis, K. J., Haszpra, L., Kern, A., Kljun, N., Koós, S., Kovács, Gy., Stingli, A., Farkas, Cs., 2010: Measurements and estimations of biosphere-atmosphere exchange of greenhouse gases -- Arable lands. In: Atmospheric Greenhouse Gases: The Hungarian Perspective (Ed.: Haszpra, L.). Springer, Dordrecht - Heidelberg - London - New York, pp. 157-197. ISBN 978-90-481-9949-5, e-ISBN 978-90-481-9950-1, doi: 10.1007/978-90-481-9950-1

Barcza, Z., Kern, A., Haszpra, L., Kljun, N., 2009. Spatial representativeness of tall tower eddy covariance measurements using remote sensing and footprint analysis. Agricultural and Forest Meteorology 149, 795-807. doi: 10.1016/j.agrformet.2008.10.021

Haszpra, L., Barcza, Z., Davis, K. J., Tarczay, K., 2005. Long-term tall tower carbon dioxide flux monitoring over an area of mixed vegetation. Agricultural and Forest Meteorology, 132, 58-77.

Bakwin, P. S., Davis, K. J.,Yi, C., Wofsy, S. C., Munger, J. W., Haszpra, L., Barcza, Z., 2004. Regional carbon dioxide fluxes from mixing ratio data. Tellus B, 56 (4), 301-311.

Barcza, Z., Haszpra, L., Kondo, H., Saigusa, N., Yamamoto, S., Bartholy, J., 2003. Carbon exchange of grass in Hungary. Tellus B, 55 (2), 187-196.

Haszpra, L., Barcza, Z., Bakwin, P. S., Berger, B. W., Davis, K. J., Weidinger, T. 2001: Measuring system for the long-term monitoring of biosphere/atmosphere exchange of carbon dioxide. Journal of Geophysical Research. Vol. 106D, 3057-3070.

Haszpra, L., 1999: On the representativeness of carbon dioxide measurements, Journal of Geophysical Research, Vol. 104D, 26953-26960.

Barcza, Z., 2001: Long term atmosphere/biosphere exchange of CO2 in Hungary. Ph.D. Dissertation, Eötvös Loránd University, Department of Meteorology, Budapest, 2001.

<< back to the previous page