Malaysian Palm Oil Board (MPOB), Persiaran Institusi, Bandar Baru Bangi, Kajang, Selangor,
1
Malaysia.
Environmental Technology, School of Industrial Technology, Universiti Sains Malaysia, USM,
2
Abstract
A study was conducted on carbon dioxide (CO2) and energy fluxes (i.e. latent (L) and sensible heat (H)) above the
canopy of mature oil palms planted on the inland mineral soil in Keratong, Pahang, Peninsular Malaysia. The
measurement was conducted over an 18-month period from September 2013 to August 2014, using the eddy
covariance method. There was a significant seasonal variation in the monthly averaged CO 2 fluxes over the
measurement period. The monthly averaged CO2 flux values ranged between -2 to -6 μmol m–2 s–1, with an
average value of about -3.5 μmol CO2 m–2 s–1. This could be due to the irregular cumulative monthly precipitation
and net radiation during the observation period. Relatively low average monthly CO2 flux (or high uptake of CO2)
also corresponds with the lowest monthly average LE and rainfall in months February 2013 and 2014. The
negative CO2 flux value shows that the mature oil palm ecosystem from an inland mineral soil area in Peninsular
Malaysia was a sink for CO2. Analysis of energy balance closure shows that the slope between latent and sensible
heat fluxes and total incoming energy was about 0.69 with an r2 value of 0.86. The slope value obtained in this
study suggests that there was a surplus of available energy compared to the measured energy fluxes. Energy
balance ratio was about 0.81 and comparable to other agricultural surfaces. This means that 81% of the available
energy was accounted through the surface flux measurements.
* Corresponding Author: Haniff MH mhaniff@mpob.govmy
Meteorological data were collected by sensors The data processing steps carried out started from data
installed at the top of the tower. Solar radiations were despiking, statistical screening, sonic anemometer axis
measured by pyranometer energy sensor LI-200SL rotating, block averaging, compensating for time lag and
(LICOR, Lincoln, USA), quantum sensor LI-190 water vapour density, estimation of flux footprint and
(Kipp & Zonen, Netherlands). The quantum sensors A block averaging method was applied by averaging
were installed at 2 m, 10 m and 30.65 m above the 30 min blocks of data to obtain the CO2 flux (μmol m–
ground for measuring photosynthetic photon flux 2 s–1), latent heat flux, LE (W m–2) and sensible heat
density (PPFD). flux, H (W m–2) values. In addition, quality flags (QC)
were calculated for all the flux variables (i.e. CO2, H
Air temperature and relative humidity were measured and LE fluxes). Data points in the time series were
by the HMP155G probes (Vaisala, Finland) installed flagged according to results from two tests, i.e., steady
state test and the test on developed turbulent
at the same heights as the quantum sensors to acquire
conditions (Foken et al., 2004).
profile measurements. Rainfall was measured by a
tipping bucket rain gauge TR-525USW (Texas The two flags were then combined into a final flag
Electronics, USA) installed on top of the tower at based on the scheme after the Spoleto agreement in
30.65 m. Wind velocity and wind direction were 2004 for Carbo Europe-IP (Mauder and Foken,
2004). The data were flagged according to different
measured by the Wind Monitor 05305 (RM Young,
categories: QC0 denotes the best quality; QC1, good
USA) installed on top of the tower.
quality; and QC2, bad quality. All QC2 data points
were removed from the final analysed dataset. The
Three self-calibrating soil heat flux plates HFP01SC amount of flux data removed using this quality
(Huk-seflux, Netherlands) were buried 1 cm deep in scheme was 31% for CO2, 32% for LE and 30% for H.
the soil at 1-, 2- and 4-m distances from a palm close Subsequent data analyses were performed using R
version 3.2.1 (R Core Team, 2015). Footprint
to the tower. Soil temperature sensor LI-7900-180
estimation was done by evaluating the probability of
(LICOR, Lincoln, USA) and moisture probe EC5
the CO2 concentrations that contribute to the CO2 flux
(Decagon Devices, USA) were also installed at a soil at the sensor location were emitted from the surface
depth of 5 cm near each location of the soil heat flux of the study site. It estimates the relative importance
supply. The flux and supplementary data were atmospheric stability, and surface roughness (Kljun et
al., 2004). In this study, the footprint estimation
continuously recorded by the Biomet data logger and
method will use a simple footprint parameterization
downloaded at weekly intervals to a portable computer.
as described by Kljun et al. (2004) and Kormann and
Meixner (2001).
Flux data processing
Energy balance closure
The 10 Hz raw data from the sonic anemometer and
The energy balance closure (EBC) can be used as a
CO2 open path sensor was analyzed using the Eddy
quality check of eddy covariance flux measurements.
Pro software (version 5.1.1, LICOR, USA) to produce
It can determine whether the sensible heat flux (H)
the half-hourly averages of CO2 and energy fluxes. It and latent heat flux (LE) measured by an eddy
also processes all-time series data from the Biomet covariance system are adequate to explain all of the
data logger to produce mean values at the same ‘available energy’, indicated by net radiation (R n) and
averaging period of 30 min as the flux data. soil heat flux (G), i.e. Rn – G – H – LE = closure error.
Energy balance closure is based on the fundamental The energy balance closure can be evaluated by
theory of the first law of thermodynamics, where the determining the slope and intercepts of a regression
sum of the measured latent heat flux (LE) and line calculated by using the ordinary least squares
sensible heat flux (H) is required to be equal to all method between the sum of LE and H (i.e. LE + H)
other energy sources and sinks (Wilson et al. 2002). and the available energy (i.e. Rn – G – S) for half-
hourly data throughout the measurement period.
Closure of the energy balance can be calculated using
Similarly, it can be checked by the energy balance
the following equation:
ratio (EBR), which is the ratio of the summation of
LE + H = Rn – G - S
(LE + H) and the summation of (Rn – G – S) as
described by Wilson et al. (2002).
Where, LE is latent heat flux (W m–2), H is sensible
heat flux (W m–2), Rn is net radiation (W m–2), G is Results and discussion
soil heat flux (W m–2) and S is heat storage (W m–2). Meteorological data
LE and H are both measured directly by the eddy Table 1 shows the daily average net radiation, air
covariance system, while Rn and G are measured temperature, relative humidity, soil water content and
soil temperature measured at the study site over a
using slower response sensors recorded by the Biomet
period of 18 months (September 2013 to February
data logger. Heat storage in terms of latent heat and
2015). The daily average for net radiation was 135.0
sensible heat were calculated from the measurements
W m–2 , air temperature 26.6˚C, relative humidity
taken by the air temperature and relative humidity
85.3%, soil water content 0.177 (m3 water m–3 soil)
sensors installed at different heights on the tower.
and soil temperature 26.1°C.
and subsequently, an unclosed energy balance. For of the Malaysian Palm Oil Board (MPOB) for
permission to publish this paper. We also greatly
example, due to the high prod-uctivity of oil palms, the
appreciate the collaboration given by the MPOB
conversion of photosynthetic active radiation (PAR) to
Keratong station management and valuable assist-
chemically stored energy in its biomass can be a ance from the staff of the Tropical Peat Research
significant energy sink. However, the amount of Institute (TROPI) Unit.
energy absorbed by plants is difficult to measure
References
accurately even when photosynthetic photon flux
Aoki M, Yabuki K, Koyama H. 1975. Microme-
density (PPFD) is measured. This is because the
teorology and assessment of primary production of a
calculation of the total amount of photosynthetic tropical rain forest in West Malaysia. Journal of
photons absorbed by plants and the energy contained Agricultural Meteorology 31(3), 115-124.
within the photosynthetic photons are required.
Asari N, Suratman MN, Jaafar J, Khalid MM.
Similarly, the efficiency in converting PAR into
2013. Estimation of aboveground biomass for oil palm
chemical energy, the amount of energy utilized by the
plantations using allometric equations. IPCBEE 2013:
plants, the loss of energy to the surroundings and the Proceedings of the 4th International Conference on
amount of energy that is chemically stored within the Biology, Environment and Chemistry pp.110-114.
plant biomass have to be identified.
Aubinet M, Vesala T, Papale D. 2012. Eddy
Covariance: A Practical Guide to Measurement and
Conclusion
Data Analysis (Springer Atmospheric Sciences).
This paper presents the first assessment of CO2 flux Dordrecht; New York: Springer.
measured over a mature palm oil plantation from an
inland mineral soil area in Peninsular Malaysia, using Baldocchi D, Falge E, Gu L, Olson R, Hollinger
D, Running S. 2001. FLUXNET: a new tool to study
the eddy covariance method, over a continuous period
the temporal and spatial variability of ecosystem-
of 18 months. The monthly averaged CO2 flux values
scale carbon dioxide, water vapor, and energy flux
over the 18-month period show a significant densities. Bulletin of the American Meteorological
relationship with seasonal variation, such as rainfall Society 82(11), 2415-2434.
and net radiation. The CO2 flux values ranged
Baldocchi D. 2013. Ecosystem Services of Energy
between -2 to -6 μmol m–2 s–1, with an average value
Exchange and Regulation. Climate Vulnerability:
of about -3.5 μmol CO2 m–2 s–1. The negative CO2 flux
Understanding and Addressing Threats to Essential
value indicates that the mature oil palm ecosystem
Resources. Elsevier Inc., Academic Press 81-92 pp.
from an inland mineral soil area in Peninsular
Malaysia was a sink for CO2. Energy balance closure Burba G. 2013. Eddy Covariance Method for
analysis had a slope value of 0.69 between latent and Scientific, Industrial, Agricultural and Regulatory
Applications: a Field Book on Measuring Ecosystem
sensible heat fluxes and total incoming energy, and an
Gas Exchange and Areal Emission Rates. Lincoln,
energy balance ratio of 0.81, which was comparable to
USA: LI-COR Biosciences.
other agricultural surfaces. Further long-term
research and measurements are needed to collect Clark DA. 2004. Sources or sinks? The responses of
continuous and consistent eddy covariance datasets, tropical forests to current and future climate and
which will help us to understand the oil palm atmospheric composition. Philosophical Transactions of
ecosystem dynamics, environment and climate the Royal Society B: Biological Sciences 359(1443),
477-491.
change.
Foken T, Göockede M, Mauder M, Mahrt L, Kljun N, Calanca P, Rotach MW, Schmid HP.
Amiro B, Munger W. 2004. Post-field data quality 2004. A simple parameterisation for flux footprint
control. In: Lee X, Massman W, Law B, Eds. Handbook predictions. Boundary-Layer Meteorology 112(3),
of Micrometeorology: A Guide for Surface Flux 503-523.
Measurement and Analysis (Vol. 29, pp. 181-208,
Atmospheric and Oceanographic Sciences Library). Kormann R, Meixner F. 2001. An analytical footp-
USA: Springer Science + Business Media Inc. rint model for non-neutral stratification. Boundary-
Layer Meteorology 99(2), 207-224.
Foken T, Wimmer F, Mauder M, Thomas C,
Liebethal C. 2006. Some aspects of the energy Kosugi Y, Takanashi S, Ohkubo S, Matsuo N,
balance closure problem. Atmospheric Chemistry and Tani M, Mitani T. 2008. CO2 exchange of a tropical
Physics 6(12/2), 4395-4402. rainforest at Pasoh in Peninsular Malaysia. Agricul-
tural and Forest Meteorology 148(3), 439-452.
Foken T. 2008. Micrometeorology. Berlin, Germany:
Springer-Verlag.
Lamade E, Bouillet JP. 2005. Carbon storage and
global change: The role of oil palm. Oleaginaux Corps
Haniff MH, Henson IE, Md Noor MR. 2004.
Gras Lipides 12, 154-160.
Continuous measurement of canopy CO2 and H2O
fluxes of an oil palm plantation using eddy correlation
Malhi Y, Grace J. 2000. Tropical forests and atmo-
technique. In: 15th Malaysian Society of Plant
spheric carbon dioxide. Trends in Ecology and
Physiology Conference Advances in Plant Science,
Evolution 15(8), 332-337.
Port Dickson, Malaysia 14-16 September 2004.
Henson I E. 1993. Carbon assimilation, water use and Mauder M, Foken T. 2004. Documentation and
energy balance of an oil palm plantation assessed using Instruction Manual of the Eddy Covariance Software
micrometeorological techniques. In: Proceedings of 1993 Package TK3. Bayreuth: University of Bayreuth, Dep-
PORIM International Palm Oil Congress - Update and artment of Micrometeorology.
Vision (Agriculture), 1993: Palm Oil Research Institute
of Malaysia, Kuala Lumpur. pp. 137-158. MPOB. 2016. Malaysian Oil Palm Statistics 2015.
Bangi, Selangor: Malaysian Palm Oil Board.
Henson IE, Haniff MH. 2005. The influence of
climatic conditions on gas and energy exchanges Oliphant AJ, Grimmond CSB, Zutter HN,
above a young oil palm stand in North Kedah, Schmid HP, Su HB, Scott SL. 2004. Heat storage
Malaysia. Journal of Oil Palm Research 17, 73-91. and energy balance fluxes for a temperate deciduous
forest. Agricultural and Forest Meteorology 126(3-
Henson IE. 1999. Comparative ecophysiology of oil
4), 185-201.
palm and tropical rainforest. In: Gurmit Singh et al.,
Eds. Oil Palm & the Environment: A Malaysian
R Core Team. 2015. R: A Language and
Perspective. Malaysian Oil Palm Growers Council,
Environment for Statistical Computing. R Foundation
Kuala Lumpur pp. 9-39.
for Statistical Computing, Vienna, Austria. URL
http://www.R-project.org/.
Khalid H, Zin ZZ, Anderson JM. 1999. Quantif-
ication of oil palm biomass and nutrient value in a
Soepadmo E. 1993. Tropical rainforests as carbon
mature plantation. I. Above-ground biomass. Journal
sinks. Chemosphere 27(6), 1025-1039.
of Oil Palm Research 2, 23-32.
Sunaryathy PI, Suhasman, Kanniah, KD, Tan Yasuda Y, Ohtani Y, Watanabe T, Okano M,
KP. 2015. Estimating Aboveground Biomass of Oil Yokota T, Liang N. 2003. Measurement of CO2 flux
Palm Trees by Using the Destructive Method. World
above a tropical rain forest at Pasoh in Peninsular
Journal of Agricultural Research 3(1), 17-19.
Malaysia. Agricultural and Forest Meteorology 114(3-
Wilson K, Goldstein A, Falge E, Aubinet M, 4), 235-244.
Baldocchi D, Berbigier P. 2002. Energy balance
closure at FLUXNET sites. Agricultural and Forest
Meteorology 113(1-4), 223-243.