ArticlePDF Available

Atmospheric measurement of point source fossil CO2 emissions

  • University of Auckland and Motu

Abstract and Figures

We use the Kapuni Gas Treatment Plant to examine methodologies for atmospheric monitoring of point source fossil fuel CO2 (CO2ff) emissions. The Kapuni plant, located in rural New Zealand, removes CO2 from locally extracted natural gas and vents that CO2 to the atmosphere, at a rate of ~0.1 Tg carbon per year. The plant is located in a rural dairy farming area, with no other significant CO2ff sources nearby, but large, diurnally varying, biospheric CO2 fluxes from the surrounding highly productive agricultural grassland. We made flask measurements of CO2 and 14CO2 (from which we derive the CO2ff component) and in situ measurements of CO2 downwind of the Kapuni plant, using a Helikite to sample transects across the emission plume from the surface up to 100 m a.g.l. We also determined the surface CO2ff content averaged over several weeks from the 14CO2 content of grass samples collected from the surrounding area. We use the WindTrax plume dispersion model to compare the atmospheric observations with the emissions reported by the Kapuni plant, and to determine how well atmospheric measurements can constrain the emissions. The model has difficulty accurately capturing the fluctuations and short-term variability in the Helikite samples, but does quite well in representing the observed CO2ff in 15 min averaged surface flask samples and in ~1 week integrated CO2ff averages from grass samples. In this pilot study, we found that using grass samples, the modeled and observed CO2ff emissions averaged over one week agreed to within 30%. The results imply that greater verification accuracy may be achieved by including more detailed meteorological observations and refining 14CO2 sampling strategies.
Content may be subject to copyright.
Atmos. Chem. Phys., 14, 5001–5014, 2014
© Author(s) 2014. CC Attribution 3.0 License.
Atmospheric measurement of point source fossil CO2emissions
J. C. Turnbull1,2, E. D. Keller1, T. Baisden1, G. Brailsford3, T. Bromley3, M. Norris1, and A. Zondervan1
1National Isotope Centre, GNS Science, Lower Hutt, New Zealand
2CIRES, University of Colorado at Boulder, Boulder, CO, USA
3NIWA, Wellington, New Zealand
Correspondence to: J. C. Turnbull (
Received: 16 October 2013 – Published in Atmos. Chem. Phys. Discuss.: 7 November 2013
Revised: 21 January 2014 – Accepted: 20 March 2014 – Published: 21 May 2014
Abstract. We use the Kapuni Gas Treatment Plant to ex-
amine methodologies for atmospheric monitoring of point
source fossil fuel CO2(CO2ff) emissions. The Kapuni plant,
located in rural New Zealand, removes CO2from locally ex-
tracted natural gas and vents that CO2to the atmosphere, at
a rate of 0.1Tg carbon per year. The plant is located in
a rural dairy farming area, with no other significant CO2ff
sources nearby, but large, diurnally varying, biospheric CO2
fluxes from the surrounding highly productive agricultural
grassland. We made flask measurements of CO2and 14CO2
(from which we derive the CO2ff component) and in situ
measurements of CO2downwind of the Kapuni plant, using a
Helikite to sample transects across the emission plume from
the surface up to 100m above ground level. We also deter-
mined the surface CO2ff content averaged over several weeks
from the 14C content of grass samples collected from the sur-
rounding area. We use the WindTrax plume dispersion model
to compare the atmospheric observations with the emissions
reported by the Kapuni plant, and to determine how well at-
mospheric measurements can constrain the emissions. The
model has difficulty accurately capturing the fluctuations and
short-term variability in the Helikite samples, but does quite
well in representing the observed CO2ff in 15min averaged
surface flask samples and in one week integrated CO2ff
averages from grass samples. In this pilot study, we found
that using grass samples, the modeled and observed CO2ff
emissions averaged over one week agreed to within 30%.
The results imply that greater verification accuracy may be
achieved by including more detailed meteorological obser-
vations and refining 14C sampling strategies.
1 Introduction
Emissions of fossil fuel carbon dioxide (CO2ff) are the main
driver of the post-industrial increase in atmosphere CO2mole
fraction (IPCC, 2007; Tans et al., 1990). Knowledge of the
sources and magnitude of CO2ff emissions is critical to im-
proving our understanding of Earth’s carbon cycle and cli-
mate system. Large point sources (electricity generation and
large-scale industry) make up roughly one third of all CO2ff
emissions (IPCC, 2007). These point sources are the first
CO2ff emissions sector to be regulated under various national
and international carbon tax and cap and trade schemes (e.g.
Australian Government, 2013; Government of India, 2010).
Point sources are also the most likely candidates for emis-
sions reduction by carbon capture and sequestration (IPCC,
The success of regulatory schemes depends on the ability
to demonstrate that emissions targets are actually achieved.
Regulating emissions without monitoring “is like dieting
without weighing oneself” (Nisbet and Weiss, 2010). Cur-
rently, point source emissions are determined using “bottom-
up” estimates from self-reported inventory data. Emissions
estimates are typically obtained from the volume of fossil
fuel (coal, oil or natural gas) consumed and carbon content
of that fuel (Andres et al., 2012; Gurney et al., 2009). Uncer-
tainties in the calculated emissions arise from uncertainties
in the amount of fuel used, which may include transcription
errors and errors in collating the data, and from uncertainties
in the carbon content of the fuels themselves. In some coun-
tries, smokestack CO2emissions are directly measured (e.g.
CEMS in the US), with a likely uncertainty of 20% (Ack-
erman and Sundquist, 2008). In upcoming regulatory envi-
ronments, it is also possible that deliberate falsification of
Published by Copernicus Publications on behalf of the European Geosciences Union.
5002 J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions
reported emissions will occur. Thus there is a need for inde-
pendent, objective measurements of these emissions both to
improve the accuracy of the reported emissions, and to pro-
vide independent monitoring as we move into a regulatory
Atmospheric measurements of recently added fossil CO2
mole fraction can be combined with knowledge of atmo-
spheric transport in a “top-down” approach to infer the CO2ff
emission flux, providing an emission estimate with quan-
tifiable uncertainties, that is independent from the bottom-
up approaches. In the top-down approach, two key compo-
nents are needed: measurements of CO2ff mole fraction and
a model of the atmospheric transport.
CO2ff cannot be directly measured in the atmosphere,
since CO2ff is but one component of the total CO2mole
fraction. The CO2background mole fraction is 400 parts
per million (ppm) and increasing by 1–2ppmyr1primarily
due to global CO2ff emissions (IPCC, 2007; Conway et al.,
2011). Large diurnal and seasonal cycles are superimposed
on this, mainly due to the seasonally and diurnally vary-
ing exchange with the terrestrial biosphere by photosynthe-
sis and respiration as well as biomass burning (IPCC, 2007).
Ocean exchange of carbon, although having a gross flux of
similar magnitude, is of lesser importance over the land ar-
eas where most CO2ff emissions occur. The CO2mole frac-
tion at a given site will also vary with meteorology as dif-
ferent air masses are advected to the location and as vertical
mixing varies through time. When the CO2ff mole fraction
added by a point source is large relative to the variability in
the CO2background, CO2measurements alone may be suf-
ficient to determine added CO2ff mole fraction. However, in
many cases, variability in CO2background is large relative to
the added CO2ff mole fraction. This is particularly important
when there is a strong biospheric carbon flux nearby. Loh et
al. (2009) used the WindTrax Lagrangian particle dispersion
model to evaluate atmospheric measurements of point source
CO2and methane emissions at a local scale. They showed
that the method could be useful for methane, where the emis-
sions were large relative to the methane background variabil-
ity, but was more difficult for CO2, where background vari-
ability was a dominant source of uncertainty.
CO2derived from fossil sources is entirely free of the iso-
tope 14C, which is removed by radioactive decay with a half-
life of 5730 yr (Karlen et al., 1968). All other sources of CO2
contain 14C at levels close to that of the current atmosphere
(Randerson et al., 2002; Turnbull et al., 2009). Thus mea-
surements of the radiocarbon content of CO2(114CO2)can
be used to quantify the CO2ff mole fraction (Suess, 1955;
Tans et al., 1979). In the current atmosphere, 1 ppm of added
CO2ff decreases 114CO2by about 2.6 ‰ (Turnbull et al.,
114CO2can be determined directly from measurements of
14C in CO2extracted from flask samples of air (e.g. Turnbull
et al., 2007; Graven et al., 2007). The 14C content of CO2is
also maintained in carbon assimilated by plants so that the
average 114C of the assimilated CO2, and hence the overly-
ing atmosphere at the time of uptake, can be determined from
the 14C content of plant materials (e.g. Hsueh et al., 2007;
Palstra et al., 2008). CO2assimilation rates vary with lo-
cal climatic weather conditions, plant type, and plant growth
phase, meaning that a complex weighting function may be
needed to describe the averaging period (Bozhinova et al.,
2013). CO2absorption by an alkaline solution (sodium hy-
droxide, NaOH) is another commonly used method to obtain
time-integrated average 114C in the atmosphere (e.g. Levin
et al., 2010; Currie et al., 2009; van der Laan et al., 2010).
A number of studies have used correlate tracers to esti-
mate CO2ff. In this method, a trace gas that is co-emitted
with CO2ff, such as carbon monoxide (CO), is monitored
in the atmosphere. If the emission ratio of CO:CO2ff is
known, then CO2ff can easily be determined (Levin and
Karstens, 2007). CO is much more readily measured than
114CO2, so this method can obtain CO2ff mole fractions at
higher spatial and temporal resolution (Vogel et al., 2010;
Turnbull et al., 2011a). Unfortunately, the emission ratio
CO:CO2ff is imperfectly known, variable by combustion ef-
ficiency and method, and large power plants typically emit
little or no CO (USEPA, 2012). Other correlate tracers have
been considered, including sulphur hexafluoride (Turnbull et
al., 2006), perchloroethylene (Miller et al., 2012), and acety-
lene (LaFranchi et al., 2013), but most of these tracers are
only indirectly associated with CO2ff combustion sources, so
are likely not appropriate for monitoring of individual point
source emissions. Further, the amount of correlate trace gases
emitted directly from the point source may vary widely de-
pending on the fuel used, combustion process, and “scrub-
bing” of pollutant gases before they are emitted into the at-
mosphere. Hence 114CO2remains the most robust method
for quantifying CO2ff across a range of environments. Once
the CO2ff mole fraction has been determined, the emission
flux can be modeled or estimated by using a description
of atmospheric transport from the emission source to the
measurement location. This has been performed at various
scales using techniques ranging from a simple mass balance
model for urban scale emissions (Turnbull et al., 2011a), to
a Lagrangian particle dispersion model for the regional scale
(Turnbull et al., 2011b), to tracer:tracer flux estimates using
radon (Levin et al., 2003; Van der Laan et al, 2010). Mod-
eling studies have demonstrated that long-term trends in re-
gional CO2ff emissions could be determined from a com-
bination of 114C observations and regional or global model-
ing (Levin and Rödenbeck, 2007). Monitoring of CO2ff from
point sources has not previously been attempted, but other
species emitted by point sources have been monitored in the
atmosphere. Mass balance modeling has been successfully
used to monitor ozone from large power plants (Trainer et
al., 1995; Ryerson et al., 2001). This method uses aircraft
sampling at high temporal resolution across transects down-
wind of the point source, and a simple description of plume
dispersion to quantify emissions. It can estimate emissions
Atmos. Chem. Phys., 14, 5001–5014, 2014
J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions 5003
Figure 1. Map showing location of Kapuni processing plant and kite sampling locations.
to within 50% under consistent wind conditions when
emissions are large relative to the background mole fraction.
More commonly, Lagrangian atmospheric transport model-
ing is used to both identify emission sources and to quantify
those emissions. Point source emissions of numerous pollu-
tant species have been evaluated using this method, including
SO2, NOxand particulates (e.g. Dresser and Huizer, 2011;
Ghannam and El-Fadel, 2013). These studies are focused on
air quality impacts, and there is little detailed information
about the quality of total flux estimates in the models.
Here we examine methodologies for atmospheric monitor-
ing of point source CO2ff emissions. Our experimental site
is a small, isolated industrial CO2ff emission source in ru-
ral New Zealand. Our focus is on CO2ff quantification from
114CO2measurements, examining two different sampling
methods: snapshot flask sampling in the atmosphere, and
time-integrated sampling from grass. We use a Lagrangian
plume dispersion model run forward in time to predict the
CO2ff mole fraction from the known emissions and meteo-
rological data. We then compare the predicted and observed
CO2ff mole fractions to examine the different methods.
Our goal is to evaluate the methods from both scientific
and application perspectives, considering:
Measurement cost and complexity. How easily can the
sampling method be deployed at field sites, and how dif-
ficult is the measurement?
What sampling methods are most compatible with the
strengths of the current generation of atmospheric trans-
port models? Models imperfectly simulate atmospheric
transport, and emissions detection will be more or less
robust depending on how the model is used.
What are the uncertainties in the estimate of the CO2ff
emission flux, and how could these uncertainties be re-
2 Methods
2.1 Sampling location and point source description
Our experimental site is the Kapuni Gas Treatment Plant,
located in rural New Zealand and run by Vector (Fig. 1).
The Kapuni plant processes natural gas extracted from
nearby onshore natural gas wells in the Taranaki Basin.
Natural gas from this field contains 40% CO2. At the
Kapuni plant, the CO2is stripped from the natural gas
and vented to the atmosphere at a rate of 0.1TgCyr1
(NZMED, 2010). This equates to average emissions of about
3300gCs1. The emissions are small relative to many in-
dustrial facilities and power plants around the world, for ex-
ample, the world’s largest power plant (Taichung, Taiwan)
emits 300000gCs1(Ummel, 2012; Wheeler and Um-
mel, 2008). We recognize that there will be differences in
applying the results of our study to larger emission sources.
The Kapuni plant is located in a rural dairy farming area,
with no other significant CO2ff sources nearby. The agricul-
tural urea plant located 500m west of the Kapuni plant does
emit a small amount of CO2, but this is approximately 1 % of
the Kapuni plant emissions (NZMED, 2010). We avoid sam-
pling close to local roads, and also note that traffic counts are
low (one vehicle every 10 min), so the overall contribution
of traffic CO2ff in our measurements is expected to be min-
imal. There is a small CO2ff source from residential heating
using natural gas and from farm vehicle exhaust, but farm and
residential power are typically from mains electrical supply
with no local CO2ff emissions. The farmland is highly pro-
ductive grassland, with large, diurnally varying, biospheric
CO2fluxes. The surrounding terrain is relatively flat, with
elevations within 2km of the Kapuni plant varying by about
10 m. There are some trees of 20m height to the south and
west of the plant, and a dip to lower elevation directly to the
east where a stream flows (Fig. 1). Atmos. Chem. Phys., 14, 5001–5014, 2014
5004 J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions
2.2 Sampling methods
2.2.1 Kite platform
A Helikite, a patented combination kite and helium balloon
(Allsopp Helikites Ltd, Hampshire, England) was used to
sample air from the surface up to 100m above ground level,
downwind of the Kapuni plant, on 26 October, 2012. The
Helikite was fitted with a GPS (Garmin 60CSx) to deter-
mine location at 1s time resolution. A tethersonde (Graham
Digital Design, Amberley, New Zealand) with an anemome-
ter was used to measure wind speed and direction, tempera-
ture, and pressure at 10s resolution. Transmitted data were
received at a ground station providing real-time height and
wind data. The anemometer cups tangled with the tether line
for short periods during the measurement campaign, identi-
fied as zero wind speeds; we exclude these periods from our
data set. 300m of 4mm OD polyethylene (Leda Extrusions,
New Zealand) tubing was attached to the kite tether close
to the tethersonde, bringing air from the kite to our mobile
lab. A diaphragm pump (KNF, model # N186.1.2KN.18) was
placed halfway along the inlet line on the ground to improve
flow rate.
The inlet line ran to a cavity ring down spectrometer
(CRDS, Picarro model G1301) inside a mobile laboratory.
The CRDS provided real-time mole fractions for CO2in the
air arriving from the intake on the Helikite. Individual obser-
vations were made at 2s intervals. The measurement pre-
cision for CO2is better than 0.1ppm, determined from the
spread of repeat measurements of an air standard sampled us-
ing an experimental setup similar to that used for this exper-
iment. The CO2measurements are referenced to the World
Meteorological Organization WMO-X2007-CO2mole frac-
tion scale to within 0.05 ppm, and one-minute averages of
transfer gases have also been determined to a standard de-
viation on replicates of 0.05ppm. Methane (CH4)was si-
multaneously measured but is not discussed here since CH4
sources in the area are complex. The transit time from the
inlet to the mobile lab was determined from timed puffs of
(high CO2)human breath, and determined as 173 seconds.
The flask filling and CO2mole fraction measurements are ad-
justed for this time delay and matched to the GPS and mete-
orological measurements, which were operating on the same
time stamp.
Previously evacuated glass flasks (0.8–2L volume) were
filled by opening a valve directly upstream of the CRDS unit
without reducing sample flow to the CRDS. The air was dried
using magnesium perchlorate and then passed through a di-
aphragm pump to fill the flasks to a pressure of 2bar abso-
lute. Flask fill times varied from 2 to 6min, depending on
the flask volume. We determine the CO2mole fraction in the
flask sample as a weighted average of the CO2mole fraction
measured on the CRDS made during the flask filling time.
The weighting function for the flask fill was obtained by log-
ging the pressure increase in a flask, as a function of time,
for the flask sampling pump (KNF, model N814KNE) and
scaling the resulting function by flask size. The weighting
function was then approximated by fitting a polynomial to
the pressure change through time, which approximates the
fill rate well (r2=0.99).
2.2.2 Surface flasks
Five surface samplers were also deployed on 26 October
2012 at one location upwind of the Kapuni plant and four
locations downwind and beneath the Helikite track. For each
sampler, the air is drawn in through an inlet line (6mm OD,
polyethylene) from an intake 3m off the ground. A deflated
4L Tedlar bag is slowly filled at a designated flow rate from
a manifold operating at a preset overpressure. In this case,
three liters of air was collected over 15min. Each sampler
were pre-programmed to purge the sample lines for 1 min
and then collect a 15min sample once every 18 min. After
sampling was complete, a small aliquot was used to deter-
mine the CO2mole fraction on the CRDS, and then the air
sample was transferred into a pre-evacuated glass flask using
the flask pump and pressurization method described above.
A subset of these surface samples was selected for 114CO2
2.2.3 Grass samples
When plants photosynthesize CO2, the 14C/12C ratio of that
CO2is altered only by isotopic fractionation during pho-
tosynthesis (Suess et al., 1955). The 14C/12C fractionation
can be quantified from the 13C/12C fractionation (δ13C), and
114C for CO2and plant material is normalized to a δ13C
of 25‰ (Stuiver and Polach, 1977). Thus the 114C of the
plant material can be considered identical to the photosyn-
thesized CO2, integrated over the period of plant growth.
A number of studies have shown that plant material
records the broad spatial patterns of 114CO2in the modern
atmosphere, using corn leaves (Hsueh et al., 2007; Riley et
al., 2008), wine ethanol (Palstra et al., 2008), and rice grains
(Shibata et al., 2005). Several of these studies compared the
observations with model predictions, and achieved reason-
able agreement at the continental and regional scales, mostly
reflecting the spatial pattern of CO2ff emissions. However,
the exact 114C measured will depend on the growth period
of the plant, variations in photosynthetic uptake during the
growth period (e.g. weather conditions) and how the plant
allocates the photosynthesized carbon among different parts
of the plant (Bozhinova et al., 2013). The resulting sample
integrates over variable rates of photosynthesis but can gen-
erally be viewed as an integrator of the daytime photoperiod
We collected samples of grass from farmland around the
Kapuni plant on 15 August 2012 and 24 October 2012. The
grass species was not specifically identified for these sam-
ples, but the dominant species in South Taranaki is a ryegrass,
Atmos. Chem. Phys., 14, 5001–5014, 2014
J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions 5005
Lolium perenne (Roberts and Thomson, 1984). The farmland
in this region is divided into small paddocks (fenced fields)
and each paddock is grazed by the dairy cow herd for one
day every 18–25 days. The grass grows 20cm during the
regrowth period, and regrazing occurs before any flowering
has begun. We sampled grass from paddocks that had been
grazed one to two weeks previously, so our samples likely
represent an average over one to two weeks. We collected
samples of the 20cm regrowth, and radiocarbon measure-
ment was performed on part of an individual grass leaf from
each sample. As all growth is in the vegetative phase, allo-
cation of carbon to the leaves is likely consistent across the
growth period, but there will be some variability in uptake
with weather patterns, which we do not account for. We make
the simplifying assumption that the leaf samples represent
the daytime average for the one-week period preceding sam-
pling. Sample locations were determined using a handheld
GPS, and locations close to obstructions such as hedges and
buildings were avoided, as were sites close to roads.
2.3 14C measurement and CO2ff determination
CO2was cryogenically extracted from the flask samples by
slowly flowing the air over a Russian doll type liquid nitrogen
trap (Brenninkmeijer and Röckmann, 1996). In the case of
grass samples, pieces of grass were acid washed (0.5M HCl
at 85C for 30 min) to remove any adhering material, then
rinsed to neutral and freeze dried, prior to sealed tube com-
bustion with copper oxide and silver wire at 900C. The
resulting CO2was cryogenically purified. CO2from either
sample type was then reduced to graphite with hydrogen
over an iron catalyst, using methods adapted from Turnbull et
al. (2007). The 14C content was measured using accelerator
mass spectrometry at GNS Science (Baisden et al., 2013).
Measurement uncertainty in each sample was derived from
three sources: counting error of 14C atoms in the sample,
counting error in the standards used for calibration, and ad-
ditional variability amongst those standards. While counting
errors in the measurement process are governed by Poisson
statistics, we regard the variability in excess of counting er-
rors as being representative of an additional source of uncer-
tainty in the measurement and/or sample preparation. Since
these three error sources are assumed independent, they are
added in quadrature. All samples in each experimental data
set (two grass sampling experiments, and one flask experi-
ment) were measured in the same AMS measurement wheel.
Therefore, we do not include additional uncertainty due to
wheel-to-wheel scatter in secondary standards (Turnbull et
al., 2007; Graven et al., 2007). The grass samples were mea-
sured to 1000000 14C counts or until the graphite target
was exhausted, resulting in overall, single sample precision
of 1.1–1.5‰. Anticipating large CO2ff contributions in the
flask samples, they were counted to 650000 counts or un-
til the graphite target was exhausted. This, combined with
poorer AMS stability during the flask sample wheel mea-
surement (as derived from the scatter of the calibration stan-
dards), resulted in overall uncertainties of 2.0–2.5‰. Be-
tween 4 and 10 secondary standards were also measured
in each wheel, and the scatter of these secondary standards
within each wheel is, in all cases, consistent with their as-
signed uncertainties.
Results are reported as 114C, the deviation of the sample
14C content from that of the absolute radiocarbon standard,
and corrected for radioactive decay since time of collection
and normalized to a δ13C of 25 ‰ (Stuiver and Polach,
1977). CO2ff is determined from the 114C of the grass or
flask sample, taking advantage of the fact that CO2ff contains
zero 14C (114C= −1000 ‰), whereas all other CO2sources
have 114C values close to that of the atmosphere. The CO2ff
added relative to a clean air background measurement can be
determined using mass balance (Levin et al., 2003).
When the CO2content of the observed sample is known,
as for our flask samples, CO2ff is calculated from
CO2ff =CO2obs(1obs 1bg)
1ff 1bg
following equation 3 in Turnbull et al. (2009). CO2obs is the
CO2mole fraction in the observed sample, and 1obs and 1bg
are the 114C of the observed sample and background sample,
respectively. 1ff is the 114C of CO2ff, and is assigned to be
1000‰. 1bg for the flask samples was determined from
surface flasks collected upwind of the Kapuni plant (Fig. 1)
on the same day, at about the same time of day.
βis a small correction term to account for the fact that
the 114C of CO2from other sources may be slightly differ-
ent from that of the atmosphere, and may include contribu-
tions from heterotrophic respiration, oceanic CO2sources,
and nuclear-industry-produced 14C. Here, we set βto zero,
and justify this choice for each possible contribution. The
background sample was collected close to our observational
sampling sites in both space and time, so that at our site a
few tens of kilometers from the ocean, it is likely that ocean
CO2exchange has altered 114C, but this alteration occurred
in both background and observed samples. There is no nu-
clear industry activity in New Zealand and only a handful
of reactors elsewhere in the Southern Hemisphere (Graven
and Gruber, 2011), so we assume there is no nuclear indus-
try bias in our samples. Of most importance is the effect
of heterotrophic respiration occurring throughout the land-
scape. This is expected to have equally impacted both back-
ground and observed 114C, and hence the heterotrophic res-
piration correction is implicitly included in the background.
We tested how important this assumption might be, using the
Biome-BGC model v4.2 (Thornton et al., 2002; Thornton et
al., 2005), calibrated to New Zealand pasture (Baisden and
Keller, 2013; Keller et al., 2014). The Biome-BGC model is
an ecosystem process model that simulates the biological and
physical processes controlling cycles of carbon, nitrogen and
water of vegetation and soil in terrestrial ecosystems. Impor-
tant inputs include weather conditions at a daily time step and Atmos. Chem. Phys., 14, 5001–5014, 2014
5006 J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions
site-specific information such as elevation, soil composition
and rooting depth. The model has a set of 43 ecological pa-
rameters that can be customized for a particular ecosystem.
In previous work, using pasture clipping data from several
sites distributed across New Zealand, we adjusted selected
model parameters to fit modeled pasture growth to the data to
obtain a national model of pasture production for both dairy
and sheep/beef pasture at a grid scale of 5km (Keller et
al., 2014). We ran the dairy model for the grid location that
includes the Kapuni processing plant to arrive at an estimate
for the respiration CO2flux and its 114C at the sampling
sites. We assume a boundary layer flushing time of one day
at our site, and using the Biome-BGC estimates, βdue to the
heterotrophic respiration flux could be 0.2–0.4ppm. This is
the maximum bias if the heterotrophic respiration flux occurs
at the observation site but not in the background, an unlikely
In the case of the grass samples, the CO2content of the
sampled air is unknown, so CO2ff was calculated using the
slightly different formulation reported as equation 6 in Turn-
bull et al. (2009), which requires that the CO2of the back-
ground air (CO2bg) be known.
CO2ff =CO2bg(1obs 1bg)
1ff 1obs
The grass 1bg values were from samples collected 20km
to the north of the Kapuni plant in similar dairy farmland, on
the same day as the observed samples were collected. The
background CO2mole fraction was estimated as 390 ppm,
from measured values at Baring Head, New Zealand at the
same time (Currie et al., 2009; data set extended to 2012). A
4ppm error in the choice of CO2observed (Eq. 1) or back-
ground (Eq. 2) value equates to a 1% error in the determined
CO2ff mole fraction, small relative to the measurement and
atmospheric transport uncertainties. β’ is also a bias correc-
tion, formulated slightly differently to β, but accounting for
the same biases. We also set this value to zero.
A further very small bias is induced by the δ13C normal-
ization in the calculation of 114C, since the δ13C of CO2ff
is different from that of the atmosphere (Vogel et al., 2013).
In our case, this is of minimal importance, since the CO2ff
from Kapuni is 13.8‰ (measured in our laboratory using
CO2supplied by the Vector Kapuni plant), quite close to that
of the atmosphere. This implies an overestimate of CO2ff of
1–2%, less than 0.1 ppm for most of our measurements. We
ignore this bias, as δ13C was not measured on these samples,
and in fact, the atmospheric δ13C value cannot be easily de-
termined from the grass samples since isotopic fractionation
occurs during assimilation of CO2into the plant. Note that
although 14C fractionation also occurs during assimilation,
this is corrected for mathematically in the 114C notation.
2.4 Atmospheric transport model
WindTrax (Thunder Beach Scientific, Nanaimo, Canada) is
a Lagrangian stochastic particle dispersion model, designed
for modeling short-range atmospheric dispersion (horizon-
tal distances of less than 1 km from source). The physics
is described by Flesch et al. (2004) and Wilson and Saw-
ford (1996). We run the model in forward mode, in which
CO2ff emissions are assumed known and gas concentrations
or mole fractions at any given location are unknowns to be
determined. We set the model to release 105particles from
the simulated stack at every time step, enough to reduce
model uncertainty to satisfactory levels (the more particles
released, the lower the uncertainty in model predictions).
Each particle is transported according to the model physics
and specified meteorological conditions. Concentration sen-
sors are placed at the observation locations, and the model
outputs a prediction of CO2ff mole fraction at each sensor
for each time step. The wind speed and direction (along with
other atmospheric conditions), combined with the prescribed
emission rate, determine the predicted CO2ff mole fraction
at that location and time. We then compare the model pre-
diction with the observed mole fractions. Alternatively, the
model can be run in backward mode to predict the emis-
sions from the observed CO2ff mole fractions, but this is
computationally more expensive, and forward mode allows
us to also investigate the broader patterns of the predicted
plume dispersion. The model is stochastic, not deterministic,
so the outcome of model runs will vary even with the same
initial conditions and parameters. This is a source of model
error that is quantified at each time step. The terrain elevation
varies by about 10m across our sampling area, and this is not
accounted for in our model simulations.
Daily emissions were provided by the Kapuni plant oper-
ator, Vector (P. Stephenson, personal communication, 2012).
We assume constant emissions for each 24 h period, although
Vector estimates that emission rates may vary by up to 3%
during that time. Emission rates were 3100–3750gC s1in
the two weeks of August 2012 preceding the day when the
first grass samples were collected, 2700–3500gC s1during
the two weeks prior to grass sample collection on 25 October
2012, and 3200gCs1on 26 October 2012 when the flask
samples were collected. Emissions are from two stacks, both
35m high, and 10m apart. In the model, we assume that
the stacks are close enough in space to be modeled as a sin-
gle point source. We release the emissions from the known
stack height of 35 m above ground level. We also tested an
alternative scenario where emissions were released from a
height of 45m above ground level, 10m higher than the ac-
tual stack height, to account for buoyant rise of the warm,
moving plume (Briggs, 1975). This value was determined us-
ing the known emission temperature (80–85 C) and stack di-
ameter (0.6m), and a velocity estimated from the CO2emis-
sion rate. Under the unstable atmospheric conditions during
Atmos. Chem. Phys., 14, 5001–5014, 2014
J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions 5007
our measurement campaigns, the difference in effective stack
height did not make a significant difference in our results.
For the flask samples, we use a model time step of ap-
proximately 10 s, commensurate with the wind data collected
from the Helikite at 10s resolution which was used as input
to the model. We note that previous studies have shown that
WindTrax performsbest over much longer averaging periods
(Flesch et el., 2004), on the order of 10–30 min, and using the
model with very short time intervals (i.e., less than 1min) is
problematic, as the relationships built into the model assume
atmospheric equilibrium, which might not be the case at such
short time scales. Our results should be interpreted with cau-
tion, as they might reflect the inability of the model to re-
solve atmospheric instability and rapid fluctuations at such
fine time scales.
In the case of the grass samples, 10-minute time steps were
used with wind data from the stationary meteorological sta-
tion installed close to the site (Fig. 1). Wind information was
not available from the local site for the August grass growth
period. Instead, we used the meteorological data obtained for
the week directly following the grass sampling. We justify
this by comparing data from the two weeks at a long-term
station 20 km away in Hawera, which provides hourly tem-
perature and pressure data. This long-term station data could
not be used as a proxy for wind speed and direction at our
Kapuni site because the particular locations and orientations
of the two sites relative to Mt. Taranaki result in large dif-
ferences in wind direction between the two sites. However,
we found that the weeks preceding and following our August
grass sampling had similar wind patterns at Hawera.
In the absence of detailed measurements of turbulence and
atmospheric stability, a general stability category was speci-
fied in WindTrax using the Pasquill-Gifford classes (Pasquill,
1961; Gifford, 1961). The Monin–Obukhov length L (a me-
teorological measurement of stability) was then calculated
by the model along with other related variables. As the mea-
surements necessary for a more exact quantification were
not made, we assumed “slightly unstable” conditions for the
flask samples, and “moderately unstable” conditions for the
grass sampling. The meteorological conditions during the
grass sampling periods may more correctly match neutral to
slightly unstable conditions, but we found that under these
conditions, the model underestimated the observed plume
dispersion. This is discussed further in Sect. 4.2.
The model output was sampled at the location and corre-
sponding time step(s) for each sample. As the Helikite moved
during the flask filling procedure, the modeled sensor was
moved both horizontally and vertically according to the lo-
cation obtained from the GPS on the kite. For the simulation
with the grass samples, the model was sampled at the GPS
location and 1.5m above ground level to avoid surface ef-
fects. We found that grouping several model sensors around
the GPS location and then averaging their output was more
accurate than using just one. In all results reported for the
grass samples, we used four sensors placed at the corners of
Figure 2. Measured CO2mole fraction across transect from south
to north. Large black circles indicate the flask sampling locations.
a square 30.5 ×30.5m, with the actual sample location in the
center. The model output was averaged over all daytime time
steps for the week prior to grass sampling to arrive at a final
predicted CO2ff mole fraction.
3 Observational Results
3.1 CO2measurements
During the Helikite sampling period on 26 October 2012, the
measured CO2mole fraction varied from 360ppm at ground
level to 592ppm in samples within the Kapuni emission
plume (Figs. 2 and 3). The emission plume moved during
the four-hour sampling period, so that our Helikite observed
the plume at different locations across the north–south tran-
sect at different times. The plume also moved in the vertical,
and was more dispersed at some times than at others.
In Fig. 3, strong photosynthetic drawdown can be seen in
the 7m above the surface, with CO2mole fractions as low
as 360 ppm, about 30ppm of drawdown. Above 7 m, the CO2
background can be estimated from the lowest CO2mole frac-
tions observed (Fig. 3). This background varied from 390–
395ppm with height, and also evolved during the 4-hour
measurement time. CO2mole fraction in the upwind surface
flask samples (collected three meters above ground) varied
from 386.3 to 387.4ppm over the 4-hour measurement pe-
riod. Atmos. Chem. Phys., 14, 5001–5014, 2014
5008 J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions
11: 00 12:00 13:00 14:00 15:00
Local time of day
Figure 3. Measured CO2mole fraction as a function of altitude.
Colors indicate the time of day.
We determine the CO2enhancement over background
(1CO2)in each CRDS or flask sample by subtracting the
estimated height-dependent CO2background value (Fig. 3)
from the observed CO2mole fraction. In this method, uncer-
tainty in the background mole fraction propagates directly
to uncertainty in the CO2enhancement. The CO2measure-
ment uncertainty is small relative to the background uncer-
tainty, so the total uncertainty in the enhancement for this
data set is determined from the range of background CO2val-
ues, and is ±15 ppm in the lowest 7 m, and ±2.5ppm above
7m. This level of uncertainty is quite significant for most of
the measurements, except those where the enhancements are
very large. The median CO2mole fraction for samples taken
above 7m was 397 ppm, so that the majority of measure-
ments are difficult to distinguish from the CO2background
of 390–395ppm.
3.2 CO2and CO2ff from 14C in flasks
CO2ff in the flask samples ranged from 0.6 to 52ppm, with
one-sigma uncertainties of 1.3 ppm (Fig. 4). Background
114C is not changed by CO2drawdown, and hence was less
variable than background CO2.1bg varied from 39.2±2.6
to 43.9±2.8‰, (using Student’s ttest, these values do not
differ significantly, p=0.39).
We compare the 14C-derived CO2ff with 1CO2using the
ratio 1CO2:CO2ff (RCO2:CO2ff) (Fig. 5). If the CO2emit-
ted from the Kapuni plant is entirely fossil-derived, then
1CO2should be equal to CO2ff, and RCO2:CO2ff equal to
one. We find that for the 15 Helikite samples, RCO2:CO2ff =
1.3±0.4ppm ppm1, suggesting that there may be a con-
Figure 4. CO2ff calculated from flask samples from the Helikite
and surface flasks on 26 October 2013.
Figure 5. RCO2:CO2ff in each flask sample. Blue points use the
assigned CO2background values, red points apply a 3ppm bias to
the CO2background values in calculating RCO2:CO2ff.
tribution of non-fossil CO2in the Kapuni emission plume.
However, when we increase the CO2background values by
3ppm (within the range of background variability), we find
RCO2:CO2ff =1.0±0.3ppm ppm1(blue points in Fig. 5),
indicating that the plume CO2is entirely fossil derived. The
variability in RCO2:CO2ff is therefore likely predominantly
due to uncertainties in determining the flask 1CO2and back-
ground CO2, and also to uncertainties in CO2ff which are
important in the lower mole fraction samples. The emitted
Atmos. Chem. Phys., 14, 5001–5014, 2014
J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions 5009
Figure 6. CO2ff in grass samples collected on 14 August 2012 (top)
and 24 October 2012 (bottom). The observed CO2ff derived from
114C is shown in yellow, and the modeled CO2ff prediction is
shown in white. Markers indicate the exact sampling location. A
wind rose showing the direction of the wind patterns over the previ-
ous one-week period is inset (bars indicate the direction the wind is
traveling to). In the lower panel the point indicated with the arrow
was measured 500m to the north-west, off the map.
plume appears to be entirely fossil-derived, within the uncer-
tainties of our measurements.
4 CO2ff from grass samples
The derived CO2ff in the grass samples for the two sampling
dates of 14 August and 24 October, 2012 are shown in Fig. 6.
Grass samples were measured to 114C precision of 1.1 to
1.5‰. This equates to 0.6 to 0.7ppm uncertainty in CO2ff.
CO2ff mole fraction derived from the grass samples varies
from 0.4 to 3.9ppm in the August samples, and 0.2 to
17.0ppm in the October samples. A negative CO2ff value is
non-physical, but 0.2ppm is within one sigma of zero. The
highest CO2ff values were observed in areas that were most
consistently downwind of the plant, and locations closer to
the plant typically had higher CO2ff values (Fig. 6). In Au-
gust, the wind direction was somewhat variable, dominantly
bringing the plume to the east or south of the plant, but sites
to the northwest were also occasionally downwind, resulting
in small CO2ff values at all these locations. In October, the
winds were consistently from the west, resulting in larger en-
hancements to the east of the Kapuni plant than in the August
samples, and no CO2ff detected in the sample to the north-
5 Comparison of observation and model CO2ff
5.1 Kite and surface flask samples
The model predicts CO2ff in the Helikite and surface flask
samples of 0 to 19.1 ppm. The model results for the kite sam-
ples are generally lower than the observed CO2ff (Fig. 7),
except for a few samples where observed CO2ff was quite
small. Other work shows that small errors in the simulated
wind direction can result in large errors in the modeled CO2ff
mole fraction for individual samples (Dresser and Huizer,
2011). Thus the model may frequently miss the location of
the plume over short time periods of a few minutes when the
wind direction is fluctuating rapidly. We deliberately sam-
pled in the center of the plume in the area of highest mole
fraction, and the likelihood of modeling this specific point
accurately is low given the error inherent in these types of
models. The assumption that the time-averaging interval rep-
resents an equilibrium state of the atmosphere is built into
the model. However, these conditions are most likely not
met in our simulations. As mentioned earlier, WindTrax is
known to perform poorly at time resolutions of less than
about 10 min (Flesch et al., 2004). Our results reflect the bias
in our sampling method as well as the model error associ-
ated with non-equilibrium conditions over very short time-
averaging periods. The agreement is much better for the sur-
face flasks, likely because of the longer averaging period of
15min (Fig. 8).
5.2 Grass samples
The modeled CO2ff values for the grass samples are shown in
Fig. 6 and range from 0.4 to 4.9ppm in the August samples
and 0.7 to 17.4ppm in the October samples. The modeled
predictions, like the observations, have the highest CO2ff val-
ues in the dominant wind direction, and samples taken closer
to the source have higher modeled CO2ff (Fig. 8). However,
it can be seen that the model significantly underestimates
CO2ff in a number of the October samples collected to the
southeast of the Kapuni plant. In August, the samples were
collected from sites surrounding the Kapuni plant in all di-
rections, and the wind direction was more variable, whereas
in the October case, the wind direction was consistently from
the west (Fig. 6). Discrepancies at the edge of the plume in Atmos. Chem. Phys., 14, 5001–5014, 2014
5010 J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions
Figure 7. Observed CO2ff versus modeled CO2ff for the kite (red)
and surface (grey) flask samples. Error bars are omitted for clarity.
the October samples suggest the model is not sufficiently dis-
persive in the horizontal.
The model simulations assumed moderately unstable at-
mospheric conditions. We tested the model with neutral and
slightly unstable conditions but found that under these con-
ditions, the model was even less dispersive in the horizontal
(results not shown). This was most apparent in the October
samples, where the slightly unstable model simulation pre-
dicted larger CO2ff in the samples taken directly west of the
plant, but very low CO2ff in the samples taken to the north
and south along the same transect shown in Figs. 1 and 6b.
We also tested our choice of effective stack height for the
emissions (45m), but found little change in the modeled re-
sults, with a significant change in the modeled result for only
one of the sampling locations. There was no change overall
in the coefficient of determination (r2)between model and
6 Uncertainty in emissions estimated from comparison
of observations and model
We further quantify the comparison between the model and
observations for the surface flasks and grass samples and
evaluate the model-observation mismatch by determining the
ratio CO2ffmodel :CO2ffobs (Rmodel:obs)for each individual
grass sample, and then calculate the mean Rmodel:obs. We also
report the one-sigma scatter of the individual Rmodel:obs re-
sults. Since the model does a very poor job of simulating the
Helikite observed CO2ff, we exclude these samples from this
Using all 21 measurements, including the four surface
flasks collected on 26 October 2012, eight grass samples
from August 2012, and nine grass samples from October
Figure 8. Comparison of observed and modeled CO2ff from the
August (red) and October (blue) grass samples, and 26 October sur-
face flasks (grey). The 1: 1 line is shown in black.
2012, we find good overall agreement between the model
and observations (r2=0.8, Fig. 8). The mean Rmodel:obs =
0.8±0.5. That is, on average, the modeled CO2ff is 20%
lower than the observed CO2ff, but with significant scatter
among the individual measurements. Using the August and
October grass samples alone, we find Rmodel:obs of 0.8±0.6.
Examining just the August data set (n=8), we find a less bi-
ased but more uncertain agreement between observation and
model, with mean Rmodel:obs of 1.0±0.7. The larger scatter
on this data set is mainly because the CO2ff values are quite
small for this data set. Looking at the October measurements
alone, we find a slightly larger underestimate in the model,
with Rmodel:obs =0.7±0.3. In all cases, the model remains
within one sigma of a 1: 1 match with observations.
To infer the uncertainty in emissions from our study, we
tested the model response across a wide range of emission
rates, and found that it is, as expected, linear. Thus, if model
transport were correct, we would infer from this that the re-
ported emission rate was too low by 30% in our worst case
for the October grass samples. For this experiment, we as-
sume that the reported emissions from the Kapuni plant are
correct, and therefore, any differences between the modeled
and observed CO2ff mole fractions must be due to uncertain-
ties in our methods and modeling. Thus we estimate, from
our worst case model-observation mismatch that the uncer-
tainty in emissions from our grass sample pilot study is 30 %
or better.
7 Discussion and recommendations
In this pilot study, we found that atmospheric 14C measure-
ments of 1 week integrated samples could be used to es-
timate CO2ff emissions from a point source to 30% or bet-
ter. This uncertainty estimate is derived from the comparison
Atmos. Chem. Phys., 14, 5001–5014, 2014
J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions 5011
of our top-down observations with forward modeling, so in-
corporates all sources of error, including model transport
errors, sampling biases and measurement uncertainties. We
now consider those various sources of error, the practicali-
ties of field measurements, and outline some improvements
that could substantially improve the method and reduce un-
certainties in the near future.
Large CO2ff mole fractions were observed in the “snap-
shot” Helikite samples collected over a few minutes, but the
WindTrax model was unable to accurately predict the high
observed CO2ff mole fractions. Conversely, the model was
quite skillful in predicting the observed CO2ff mole fractions
in the long-term averaged grass samples and in 15min av-
eraged surface flasks. The model performed best in captur-
ing the broad spatial pattern of emissions around the Kapuni
plant as shown in the August sampling pattern. The model
was less skillful at capturing the somewhat finer-scale pat-
tern of the October grass samples, which were predominantly
sampled in the same sector. This result is consistent with
findings from other studies (e.g. Dresser and Huizer, 2011)
that show that small errors in model transport of point source
emissions can result in the emission plume being incorrectly
located. Averaging the model results over time can reduce
the impact of these errors. This conclusion is likely broadly
applicable to local scale plume dispersion models such as
WindTrax, as well as to regional scale Lagrangian models.
The model skill will likely also be improved by more de-
tailed meteorological measurements, including estimates of
boundary layer turbulence and atmospheric stability.
We found the long-term averaged grass sampling better
suited to the model skill than the snapshot Helikite flask sam-
ples. Grass sampling is also far cheaper and easier than flask
sampling, particularly from an elevated platform such as the
Helikite. Flask sampling requires flasks and the equipment
to fill them, which may run to many thousands of dollars per
sampling system. Elevated platforms such as the Helikite, as
well as aircraft and unmanned aerial vehicles can also be ex-
pensive, can only operate under specific meteorological con-
ditions and may be subject to air traffic regulations. In con-
trast, plant sampling can be done simply and quickly using
only a few plastic bags and a handheld GPS to record loca-
tions. Whereas flask samples may be limited in size by the
practicalities of flask volumes and pumping rates, large plant
material samples can be collected to facilitate replicate mea-
surements if required. Plant sampling is also less intrusive
and less visible to the power plant operators and the public,
which may be advantageous in some situations. The labora-
tory 14C preparation, while slightly different for each sample
type, is of similar complexity and cost. However, plant sam-
pling may suffer biases that cannot be easily quantified. The
particular environment at our Kapuni site is conducive to the
grass sampling method, with rapid grass growth and regular
grazing to consistently remove old growth before flowering.
Biases in the 114C of grass or other plant material due to
the details of plant CO2assimilation through time may make
this method challenging in other locations. In future work,
we will examine alternative integrated sampling techniques
such as NaOH absorption that provide a similar integration
period, but allow more control over when CO2is collected.
This type of sampler would require more complex field sam-
pling than plant material, but could still be relatively easily
deployed in the field, allows collection of large amounts of
CO2, and laboratory methods are well-established.
Uncertainties in observed CO2ff for the grass samples
come from the 14C measurement uncertainty, uncertainties
in the CO2assimilation period represented by the grass sam-
ple, and from the choice of background. Large gains in 14C
measurement uncertainty are unlikely in the near future, but
the impact of measurement uncertainties could be reduced by
measuring multiple aliquots of each sample, or preferably by
collecting and measuring more samples at higher spatial and
temporal resolutions to obtain a greater data resolution for
the model – observation comparison. As already discussed,
uncertainties in the CO2assimilation period could be reduced
by using an alternative integrated sampling technique such as
NaOH absorption. We prepared and measured an individual
grass leaf from each sample, but homogenizing and measur-
ing a mixture of several leaves, possibly from several grass
plants, might give a more representative sample. There is also
potential for bias in observed CO2ff from the choice of back-
ground. We used a single background measurement for each
data set, and any bias in that background will result in a con-
sistent bias in Rmodel:obs calculated for each sample. Making
several background measurements would reduce background
uncertainty and bias.
The Kapuni plant emissions are quite small (two orders of
magnitude) relative to many fossil fuel power plants around
the world. Larger point sources will produce larger observed
CO2ff mole fractions, and hence relatively smaller measure-
ment uncertainties. However, many large power plants will
have much higher stack emission heights of 100 to 800m,
and plume buoyancy due to hot emissions might raise the
effective emission height even further. Therefore surface or
near-surface measurements might need to be made further
downwind to observe the plume. At these larger distances
from the source, the plume will be more dispersed and hence
observed CO2ff mole fractions will be reduced, likely to sim-
ilar magnitude to those we observed at Kapuni. A further
modeling consideration is that in this study we used only
daytime measurements. It is well-known that atmospheric
transport models perform best in the mid-afternoon when
the boundary layer is well-mixed, but have difficulty in accu-
rately representing the nocturnal boundary layer. Thus most
researchers utilize only daytime measurements. This repre-
sents an unresolved difficulty in assessing overall emissions
for power plants, which may have significant diurnal vari-
ability in their emission rates.
We deliberately selected a location with reasonably flat
terrain where atmospheric transport is straightforward. More
complex terrain will make the transport modeling more Atmos. Chem. Phys., 14, 5001–5014, 2014
5012 J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions
difficult and will require additional care in selection of op-
timal sampling locations and times. Extending this work to
sites with multiple emission sources will also complicate in-
From this pilot experiment, we believe that it is realistic to
substantially reduce the uncertainty in atmospheric determi-
nation of point source CO2ff emissions in the near future. A
goal of 10–20% overall uncertainty appears realistic.
Acknowledgements. This work was funded by GNS Science
Strategic Development Fund and public research funding from
the Government of New Zealand. Jenny Dahl, Kelly Lyons and
Johannes Kaiser assisted with 14C measurements. Ross Martin
assisted with analyzing the vertical profile data from the Helikite
radiosonde. We wish to thank Peter Stephenson and the staff at
the Vector Kapuni processing plant providing necessary details
on the plant’s CO2emissions and their interest in the research.
Darryl and Alison Smith, Roger Luscombe, Brent and Kevin
Parrett all generously allowed us access to their land for sampling
and provided helpful information on local conditions. Thanks to
the two reviewers, Zoe Loh and Felix Vogel, for their thoughtful
comments and suggestions.
Edited by: M. Heimann
Ackerman, K. V. and Sundquist, E. T.: Comparison of two US
power-plant carbon dioxide emissions data sets, Environ. Sci.
Technol., 42, 5688–5693, 2008.
Andres, R. J., Boden, T. A., Bréon, F.-M., Ciais, P., Davis, S., Er-
ickson, D., Gregg, J. S., Jacobson, A., Marland, G., Miller, J.,
Oda, T., Olivier, J. G. J., Raupach, M. R., Rayner, P., and Trean-
ton, K.: A synthesis of carbon dioxide emissions from fossil-fuel
combustion, Biogeosciences, 9, 1845–1871, doi:10.5194/bg-9-
1845-2012, 2012.
Australian Government: Starting emissions trading on 1 July 2014,
Canberra, Australia, 2013.
Baisden, W. T. and Keller, E. D.: Synthetic constraint of soil C dy-
namics using 50 years of radiocarbon and Net Primary Produc-
tion (NPP) in a New Zealand grassland site, Radiocarbon, 55,
1071–1076, 2013.
Baisden, W. T., Prior, C. A., Chambers, D., Canessa, S., Phillips, A.,
Bertrand, C., Zondervan, A., and Turnbull, J. C.: Radiocarbon
sample preparation and data flow at Rafter: accommodating en-
hanced throughput and precision, Nucl. Instrum. Methods, B294,
194–198, 2013.
Bozhinova, D., Combe, M., Palstra, S. W. L., Meijer, H. A. J., Krol,
M. C., and Peters, W.: The importance of crop growth modeling
to interpret the 14CO2signature of annual plants, Global Bio-
geochem. Cy., 27, 792–803, doi:10.1002/gbc.20065, 2013.
Brenninkmeijer, C. A. M. and Röckmann, T.: Russian doll type
cryogenic traps: improved design and isotope separation effects,
Anal. Chem., 68, 3050–3053, doi:10.1021/ac960208w, 1996.
Briggs, G.: Plume rise predictions, in: Lectures on Air Pollution
and Environmental Impact Analysis, American Meteorological
Society, Boston, Massachusetts, 59–111, 1975.
Conway, T. J., Lang, P. M., and Masarie, K. A.: Atmospheric carbon
dioxide dry air mole fractions from the NOAA/ESRL Carbon
Cycle Global Cooperative Network, 1968–2010, version 2011-
06-21, available at:
(last access: 9 September 2013), 2011.
Currie, K. I., Brailsford, G., Nichol, S., Gomez, A., Sparks, R.,
Lassey, K. R., and Riedel, K.: Tropospheric 14CO2at Welling-
ton, New Zealand: the world’s longest record, Biogeochemistry,
104, 5–22, doi:10.1007/s10533-009-9352-6, 2009.
Dresser, A. L. and Huizer, R. D.: CALPUFF and AERMOD model
validation study in the near field: Martins Creek revisited, J. Air
Waste Manage., 61, 647–659, doi:10.3155/1047-3289.61.6.647,
Flesch, T. K., Wilson, J. D., Harper, L., Crenna, B., and Sharpe, R.:
Deducing ground-to-air emissions from observed trace gas con-
centrations: a field trial, J. Appl. Meteorol., 43, 487–502, 2004.
Ghannam, K. and El-Fadel, M.: Emissions characterization and
regulatory compliance at an industrial complex: an integrated
MM5/CALPUFF approach, Atmos. Environ., 69, 156–169,
doi:10.1016/j.atmosenv.2012.12.022, 2013.
Gifford, F. A.: Use of routine meteorological observations for esti-
mating atmospheric dispersion, Nucl. Safe., 2, 47–57, 1961.
Government of India: India: Taking on Climate Change Post-
Copenhagen domestic actions, Ministry of Environment and
Forests, New Delhi, 2010.
Graven, H. D. and Gruber, N.: Continental-scale enrichment of at-
mospheric 14CO2from the nuclear power industry: potential im-
pact on the estimation of fossil fuel-derived CO2, Atmos. Chem.
Phys., 11, 12339–12349, doi:10.5194/acp-11-12339-2011, 2011.
Graven, H. D., Guilderson, T. P., and Keeling, R. F.: Methods for
high-precision 14C AMS measurements of atmospheric CO2at
LLNL, Radiocarbon, 49, 349–356, 2007.
Gurney, K. R., Mendoza, D. L., Zhou, Y., Fischer, M. L.,
Miller, C. C., Geethakumar, S., and de la Rue du Can, S.: High
resolution fossil fuel combustion CO2emission fluxes for the
United States, Environ. Sci. Technol., 43, 5535–5541, 2009.
Hsueh, D. Y., Krakauer, N. Y., Randerson, J. T., Xu, X., Trum-
bore, S. E., and Southon, J. R.: Regional patterns of radiocarbon
and fossil fuel-derived CO2in surface air across North America,
Geophys. Res. Lett., 34, L02816, doi:10.1029/2006gl027032,
Karlen, I., Olsson, I. U., Kllburg, P., and Kilici, S.: Absolute de-
termination of the activity of two 14C dating standards, Arkiv
Geofysik, 4, 465–471, 1968.
Keller, E. D., Baisden, W. T., Timar, L., Mullan, B., and Clark,
A.: Grassland production under global change scenarios for New
Zealand pastoral agriculture, Geosci. Model Dev. Discuss., 7, 1–
59, doi:10.5194/gmdd-7-1-2014, 2014.
LaFranchi, B. W., Pétron, G., Miller, J. B., Lehman, S. J., An-
drews, A. E., Dlugokencky, E. J., Miller, B. R., Montzka, S. A.,
Hall, B., Neff, W., Sweeney, C., Turnbull, J. C., Wolfe, D. E.,
Tans, P. P., Gurney, K. R., and Guilderson, T. P.: Constraints
on emissions of carbon monoxide, methane, and a suite of
hydrocarbons in the Colorado Front Range using observa-
tions of 14CO2, Atmos. Chem. Phys. Discuss., 13, 1609–1672,
doi:10.5194/acpd-13-1609-2013, 2013.
Atmos. Chem. Phys., 14, 5001–5014, 2014
J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions 5013
Levin, I.: A novel approach for independent budgeting of fossil fuel
CO2over Europe by 14CO2observations, Geophys. Res. Lett.,
30, 2194, doi:10.1029/2003gl018477, 2003.
Levin, I. and Rödenbeck, C.: Can the envisaged reductions of fos-
sil fuel CO2emissions be detected by atmospheric observa-
tions?, Naturwissenschaften, 95, 203–208, doi:10.1007/s00114-
007-0313-4, 2007.
Levin, I., Naegler, T., Kromer, B., Diehl, M., Francey, R. J., Gomez-
Pelaez, A. J., Steele, L. P., Wagenbach, D., Weller, R., and Wor-
thy, D. E.: Observations and modelling of the global distribution
and long-term trend of atmospheric 14CO2, Tellus B, 62, 26–46,
doi:10.1111/j.1600-0889.2009.00446.x, 2010.
Loh, Z., Leuning, R., Zegelin, S., Etheridge, D., Bai, M.,
Naylor, T., and Griffith, D.: Testing Lagrangian atmo-
spheric dispersion modelling to monitor CO2and CH4leak-
age from geosequestration, Atmos. Environ., 43, 2602–2611,
doi:10.1016/j.atmosenv.2009.01.053, 2009.
Meijer, H. A. J., Smid, H. M., Perez, E., and Keizer, M. G.: Isotopic
characterization of anthropogenic CO2emissions using isotopic
and radiocarbon analysis, Phys. Chem. Earth, 21, 483–487, 1996.
Miller, J. B., Lehman, S. J., Montzka, S. A., Sweeney, C.,
Miller, B. R., Wolak, C., Dlugokencky, E. J., Southon, J. R.,
Turnbull, J. C., and Tans, P. P.: Linking emissions of
fossil fuel CO2and other anthropogenic trace gases us-
ing atmospheric 14CO2, J. Geophys. Res., 117, D08302,
doi:10.1029/2011JD017048, 2012.
Nisbet, E. and Weiss, R.: Top-Down Versus Bottom-Up, Science,
328, 1241–1243, doi:10.1126/science.1189936, 2010.
NZMED: New Zealand’s Energy Outlook 2010, New Zealand
Ministry of Economic Development, Wellington, New Zealand,
Palstra, S. W. L., Karstens, U., Streurman, H.-J., and Mei-
jer, H. A. J.: Wine ethanol 14C as a tracer for fossil fuel CO2
emissions in Europe: measurements and model comparison, J.
Geophys. Res., 113, D21305, doi:10.1029/2008JD010282, 2008.
Pasquill, F.: The estimation of the dispersion of windborne material,
Meteorol. Mag., 90, 33–49, 1961.
Randerson, J. T.: Seasonal and latitudinal variability of troposphere
114CO2: post bomb contributions from fossil fuels, oceans, the
stratosphere, and the terrestrial biosphere, Global Biogeochem.
Cy., 16, 1112, doi:10.1029/2002GB001876, 2002.
Riley, W. J., Hsueh, D. Y., Randerson, J. T., Fischer, M. L.,
Hatch, J. G., Pataki, D. E., Wang, W., and Goulden, M. L.:
Where do fossil fuel carbon dioxide emissions from Califor-
nia go? An analysis based on radiocarbon observations and an
atmospheric transport model, J. Geophys. Res., 113, G04002,
doi:10.1029/2007jg000625, 2008.
Roberts, A. and Thomson, N.: Seasonal distribution of pasture pro-
duction in New Zealand XVII. South Taranaki, New Zealand J.
Experiment. Agr., 12, 83–92, 1984.
Ryerson, T. B., Trainer, M., Holloway, J., Parrish, D. D., Huey, L.,
Sueper, D., Frost, G. J., Donnelly, S., Schauffler, S., Atlas, E.,
Kuster, W., Golden, P., Hubler, G., Meagher, J., and Fehsen-
feld, F.: Observations of ozone formation in power plant plumes
and implications for ozone control strategies, Science, 292, 719–
723, 2001.
Shibata, S., Kawano, E., and Nakabayashi, T.: Atmospheric
[14C]CO2variations in Japan during 1982–1999 based on 14C
measurements of rice grains, Appl. Radiat. Isotop., 63, 285–290,
doi:10.1016/j.apradiso.2005.03.011, 2005.
Stuiver, M. and Polach, H. A.: Discussion: reporting of 14C data,
Radiocarbon, 19, 355–363, 1977.
Suess, H. E.: Radiocarbon concentration in modern wood, Science,
122, 414–417, 1955.
Tans, P. P., De Jong, A. F., and Mook, W. G.: Natural atmospheric
14C variation and the Suess effect, Nature, 280, 826–828, 1979.
Tans, P. P., Fung, I. Y., and Takahashi, T.: Observational constraints
on the global atmospheric CO2budget, Science, 247, 1431–
1438, 1990.
Thornton, P., Law, B. E., Gholz, H. L., Clark, K. L., Falge, E.,
Ellsworth, D., Goldstein, A. H., Monson, R. K., Hollinger, D.,
Falk, M., Chen, J. M., and Sparks, J.: Modeling and measuring
the effects of disturbance history and climate on carbon and water
budgets in evergreen needle leaf forests, Agr. Forest Meteorol.,
113, 185–222, 2002.
Thornton, P., Running, S., and Hunt, E.: Biome-BGC: Terrestrial
Ecosystem Process Model, Version 4.2. Data model. , Numerical
Terradynamic Simulation Group, School of Forestry, University
of Montana, Missoula, Montana, USA, 2005.
Thornton, P., Running, S., and Hunt, E.: Biome-BGC: Terrestrial
Ecosystem Process Model, Version 4.2. Data model, Numerical
Terradynamic Simulation Group, School of Forestry, University
of Montana, Missoula, Montana, USA, 2005.
Turnbull, J. C., Tans, P. P., Lehman, S. J., Baker, D., Chung, Y.,
Gregg, J. S., Miller, J. B., Southon, J. R., and Zhao, L.: At-
mospheric observations of carbon monoxide and fossil fuel
CO2emissions from East Asia, J. Geophys. Res., 116, D24306,
doi:10.1029/2011JD016691, 2011a.
Turnbull, J. C., Karion, A., Fischer, M. L., Faloona, I., Guilder-
son, T., Lehman, S. J., Miller, B. R., Miller, J. B., Montzka, S.,
Sherwood, T., Saripalli, S., Sweeney, C., and Tans, P. P.: Assess-
ment of fossil fuel carbon dioxide and other anthropogenic trace
gas emissions from airborne measurements over Sacramento,
California in spring 2009, Atmos. Chem. Phys., 11, 705–721,
doi:10.5194/acp-11-705-2011, 2011b.
Turnbull, J. C., Rayner, P. J., Miller, J. B., Naegler, T.,
Ciais, P., and Cozic, A.: On the use of 14CO2as a tracer
for fossil fuel CO2: quantifying uncertainties using an at-
mospheric transport model, J. Geophys. Res., 114, D22302,
doi:10.1029/2009JD012308, 2009.
Turnbull, J. C., Lehman, S. J., Miller, J. B., Sparks, R. J.,
Southon, J. R., and Tans, P. P.: A new high precision 14CO2time
series for North American continental air, J. Geophys. Res., 112,
D11310, doi:10.1029/2006JD008184, 2007.
Turnbull, J. C., Miller, J. B., Lehman, S. J., Tans, P. P., Sparks, R. J.,
and Southon, J. R.: Comparison of 14CO2, CO and SF6as tracers
for determination of recently added fossil fuel CO2in the atmo-
sphere and implications for biological CO2exchange, Geophys.
Res. Lett., 33, L01817, 10.1029/2005GL024213, 2006.
Ummel, K.: CARMA revisited: an updated database of carbon diox-
ide emissions from power plants worldwide, Center for Global
Development, Washington, D.C., 2012.
USEPA: 2008 National Emissions Inventory, version 2 Technical
Support Document, US Environmental Protection Agency, Of-
fice of Air Quality Planning and Standards, Air Quality Assess-
ment Division, Research Triangle Park, North Carolina, 2012. Atmos. Chem. Phys., 14, 5001–5014, 2014
5014 J. C. Turnbull et al.: Atmospheric measurement of point source fossil fuel CO2emissions
Van Der Laan, S., Karstens, U., Neubert, R. E. M., Van Der
Laan-Luijkx, I. T., and Meijer, H. A. J.: Observation-based es-
timates of fossil fuel-derived CO2emissions in the Nether-
lands using 114C, CO and 222Radon, Tellus B, 62, 389–402,
doi:10.1111/j.1600-0889.2010.00493.x, 2010.
Vogel, F. R., Hammer, S., Steinhof, A., Kromer, B., and Levin, I.:
Implication of weekly and diurnal 14C calibration on hourly
estimates of CO-based fossil fuel CO2at a moderately pol-
luted site in southwestern Germany, Tellus B, 62, 512–520,
doi:10.1111/j.1600-0889.2010.00477.x, 2010.
Wheeler, D. and Ummel, K.: Calculating CARMA: global estima-
tion of CO2emissions from the power sector, Center for Global
Development, Washington, D.C., 2008.
Wilson, J. D. and Sawford, B. L.: Review of Lagrangian stochastic
models for trajectories in the turbulent atmosphere, Bound.-Lay.
Meteorol., 78, 191–210, 1996.
Atmos. Chem. Phys., 14, 5001–5014, 2014
... The 14 C content of plant organic matter is primarily controlled by the incorporation of CO 2 by photosynthesis (Suess, 1955;Turnbull et al., 2014), and D 14 C values calculated from measurements of either bulk or specific fractions of plant organic matter are routinely used to represent mean daytime D 14 C values in the atmosphere during their growth period (Lichtfouse et al., 2005;Rakowski et al., 2008;Djuricin et al., 2012;Beramendi-Orosco et al., 2013;Bozhinova et al., 2013;Turnbull et al., 2014;Xu et al., 2015). In this manner, D 14 C results from foliage of different plant types (e.g. ...
... The 14 C content of plant organic matter is primarily controlled by the incorporation of CO 2 by photosynthesis (Suess, 1955;Turnbull et al., 2014), and D 14 C values calculated from measurements of either bulk or specific fractions of plant organic matter are routinely used to represent mean daytime D 14 C values in the atmosphere during their growth period (Lichtfouse et al., 2005;Rakowski et al., 2008;Djuricin et al., 2012;Beramendi-Orosco et al., 2013;Bozhinova et al., 2013;Turnbull et al., 2014;Xu et al., 2015). In this manner, D 14 C results from foliage of different plant types (e.g. ...
... In this manner, D 14 C results from foliage of different plant types (e.g. C 3 and C 4 herbaceous plants, deciduous and evergreen trees) have been used in studies that trace atmospheric CO 2ff (Alessio et al., 2002;Lichtfouse et al., 2005;Hsueh et al., 2007;Riley et al., 2008;Zhou et al., 2014;Turnbull et al., 2014;Niu et al., 2015;Beramendi-Orosco et al., 2015). Actually, the carbon isotope composition of plant organic matter is affected not only by carbon assimilation during photosynthesis (Hsueh et al., 2007;Bozhinova et al., 2013), but also by carbon consumption during respiration (O'Leary, 1981;Gillon and Griffiths, 1997). ...
Radiocarbon (¹⁴C) has been widely used for quantification of fossil fuel CO2 (CO2ff) in the atmosphere and for ecosystem source partitioning studies. The strength of the technique lies in the intrinsic differences between the ¹⁴C signature of fossil fuels and other sources. In past studies, the ¹⁴C content of CO2 derived from plants has been equated with the ¹⁴C content of the atmosphere. Carbon isotopic fractionation mechanisms vary among plants however, and experimental study on fractionation associated with dark respiration is lacking. Here we present accelerator mass spectrometry (AMS) radiocarbon results of CO2 respired from 21 plants using a lab-incubation method and associated bulk organic matter. From the respired CO2 we determine Δ¹⁴Cres values, and from the bulk organic matter we determine Δ¹⁴Cbom values. A significant difference between Δ¹⁴Cres and Δ¹⁴Cbom (P < 0.01) was observed for all investigated plants, ranging from −42.3‰ to 10.1‰. The results show that Δ¹⁴Cres values are in agreement with mean atmospheric Δ¹⁴CO2 for several days leading up to the sampling date, but are significantly different from corresponding bulk organic Δ¹⁴C values. We find that although dark respiration is unlikely to significantly influence the estimation of CO2ff, an additional bias associated with the respiration rate during a plant's growth period should be considered when using Δ¹⁴C in plants to quantify atmospheric CO2ff.
... We aim to demonstrate the ability to detect CCS leaks with quantification of the magnitude of the leak being of only secondary concern. Our main observational dataset is from the Kapuni natural gas processing plant in rural Taranaki, New Zealand (Turnbull et al., , 2014 and we also use results from other similar studies (Donders et al., 2013;Cook et al., 2001). Although these datasets are previously published elsewhere, we include explanations of the sampling methods and calculations to illustrate the 14 C method. ...
... For continental-scale studies, 14 C from nuclear power plants can bias calculated CO 2 ff low by several ppm (Graven and Gruber, 2011), and heterotrophic respiration/biomass burning can bias CO 2 ff low by up to about 1 ppm (Turnbull et al., 2009). However, for the small spatial scales applicable to CCS monitoring, ␤ can be set to zero as long as the choice of bg incorporates all these additional sources (Turnbull et al., 2014). ...
... To demonstrate the capability of the radiocarbon method for detection of CCS leaks, we use an analog site with existing 14 C/CO 2 ff observations that have been reported elsewhere (Turnbull et al., , 2014Keller et al., 2016;Norris, 2015). In this section, we summarize the site and sampling methods to provide context for our interpretation of the results as they apply to CCS leak detection. ...
We outline the methodology for detection of carbon dioxide (CO2) leaks to the atmosphere from carbon capture and storage (CCS) using measurements of radiocarbon in CO2. The radiocarbon method can unambiguously identify recently added fossil-derived CO2 such as CCS leaks due to the very large isotopic difference between radiocarbon-free fossil derived CO2 and natural CO2 sources with ambient radiocarbon levels. The detection threshold of 1 ppm of fossil-derived CO2 is comparable to other proposed atmospheric detection methods for CCS leakage. We demonstrate that this method will allow detection of a 1000 ton C yr⁻¹ leak 200–300 m from the source during the day and more than 600 m away at night. Using time-integrated sampling techniques, long time periods can be covered with few measurements, making the method feasible with existing laboratory-based radiocarbon measurement methods We examine the method using previously published observations and new model simulations for a case study in Taranaki, New Zealand. Plant material faithfully records the radiocarbon content of assimilated CO2 and we show that short-lived grass leaves and cellulose from tree rings provide effective time-integrated collection methods, allowing dense spatial sampling at low cost. A CO2 absorption sampler allows collection at controlled times, including nighttime, and gives similar results.
... A major source of uncertainty in the top-down method is the atmospheric transport models that translate the observed atmospheric mole fractions to the emission flux rate by describing the movement of air (14). Previous work has shown that these models cannot adequately simulate the short-term turbulent atmospheric variability needed to interpret individual "grab samples," flasks of air filled over a period of a few minutes (23). ...
... Four NaOH samplers experienced leakage problems, and data were discarded. CO 2 ff was determined from the Δ 14 C values (23) such that ...
... Δ obs is our observed Δ 14 C value for each time-integrated sample, and Δ bg is the background Δ 14 C value. We assume that at this site, Δ bg incorporates all other CO 2 sources, so no other corrections need be applied (23). Δ bg was taken from the monthly mean at Baring Head, Wellington, except for the December 2013 samples, for which the Baring Head Δ 14 C was 5 ‰ lower than several of the measured background samples. ...
Significance The 1,000 largest power plants comprise 22% of total global fossil fuel CO 2 emissions, making them an obvious target for regulating and reducing emissions. The success of existing and upcoming regulations and emission trading schemes requires reliable monitoring and verification of emissions, preferably using independent, objective evaluation to establish trust and transparency. However, such methodology has thus far been elusive, and emissions reporting currently relies solely on self-reported “bottom-up” inventory data. We demonstrate a method using time-integrated atmospheric observations and modeling to reliably quantify fossil fuel CO 2 emissions from point sources to within 10%. This level of uncertainty is a marked improvement over current ∼20% uncertainties for individual power plants and allows independent evaluation of reported emissions.
... CO 2 ff was determined following Turnbull et al. (2014) from the isotopic difference between the measured tree ring and clean air background CO 2 measured at Baring Head, Wellington (41.4167 • S, 174.8667 • E; Currie et al., 2011;extended with unpublished data). Baring Head, located at the southern end of New Zealand's North Island and approximately 300 km south of Kapuni, was chosen as the background for this study over more local sites because it provides a long-term record of background CO 2 and 14 C, dating back to the early 1970s. ...
... Comparison of this record with tree rings collected 3 km upwind of our source showed no difference from the Wellington record. A small correction, β, accounts for the fact that the 14 C of CO 2 from other sources may be slightly different from that of the atmosphere; in our case we set β to zero since the proximity to the coast and consistent winds suggest that other CO 2 is negligible in this location (Turnbull et al., 2014). Uncertainty in CO 2 ff is dominated by 14 C measurement uncertainty in both background and the observed sample and is typically ∼ 1 ppm for this data set. ...
... WindTrax was chosen for this study because it is easy to use and the distance scale is appropriate for our site. We previously used WindTrax to estimate CO 2 ff in grass samples at the Kapuni site (Turnbull et al., 2014), demonstrating that the model is capable of providing reasonable estimates of observed CO 2 ff. Here, we take the same approach to model CO 2 ff measured in tree rings. ...
Full-text available
We examine the utility of tree ring 14C archives for detecting long-term changes in fossil CO2 emissions from a point source. Trees assimilate carbon from the atmosphere during photosynthesis, in the process faithfully recording the average atmospheric 14C content in each new annual tree ring. Using 14C as a proxy for fossil CO2, we examine interannual variability over six years of fossil CO2 observations between 2004–2005 and 2011–2012 from two trees growing near the Kapuni Gas Treatment Plant in rural Taranaki, New Zealand. We quantify the amount of variability that can be attributed to transport and meteorology by simulating constant point-source fossil CO2 emissions over the observation period with the atmospheric transport model WindTrax. We compare model simulation results to observations and calculate the amount of change in emissions that we can detect with new observations over annual or multi-year time periods, given both the measurement uncertainty of 1ppm and the modelled variation in transport. In particular, we ask, what is the minimum amount of change in emissions that we can detect using this method, given a reference period of six years? We find that changes of 42 % or more could be detected in a new sample from one year at the same observation location or 22 % in the case of four years of new samples. This threshold is reduced and the method becomes more practical the more the size of the signal increases. For point sources 10 times larger than the Kapuni plant (a more typical size for power plants worldwide), it would be possible to detect sustained emissions changes on the order of 10 %, given suitable meteorology and observations.
... Sophisticated studies (e.g. [3][4][5][6][7][8][9]) typically rely on higher-frequency or continuous measurements of trace gas concentrations (including CO 2 and CO), higher frequencies of measurements of 14 C in ambient air, and ancillary measurements of atmospheric dynamics (i.e., boundary layer height), or satellite observations of atmospheric column CO 2 . In these studies, isotopic analysis is preferably carried out on air-CO 2 samples captured in pre-evacuated flasks [10,11], or zeolite molecular sieve traps [12]. ...
Full-text available
Fossil fuel-derived CO2 (Cff) emission patterns and their point sources across the Rio de Janeiro megacity and state were estimated from a single regional-scale Δ14C distribution map based on isotopic measurements of ipê leaves (Tabebuia, a popular flowering deciduous perennial tree). Data from multi-year sampling (i.e., 2014–2016) was renormalized to reflect 14C signatures of the 2015 calendar year. Spatial variability in Δ14C ranges from a maximum of 27.1 ± 0.4‰ (city of Petrópolis, a higher-elevation municipality) to a minimum of −43.6 ± 1.4‰ (i.e., approximately 27.6 ± 1 ppm of Cff — Santo Cristo, a district within the Rio de Janeiro city). Overall, higher Δ14C values correlate well with green habitats and high elevation areas, while lower values are associated with Cff emissions in densely populated areas with higher industrial and traffic footprints. Cff emissions are higher where local air circulation is poor, such as the area surrounding Guanabara Bay. Other areas with significantly higher Cff emissions were the Paraíba Valley and Mountain regions. These results may be explained by atmospheric transport of CO2 from neighboring states, such as São Paulo and Minas Gerais, and by the predominant west winds and the limited regional air flow created by large topographic features. Lower Cff emissions were observed in the Northwest and Lakes regions, which are dominated by agriculture and tourism activities. Our results highlight the potential of directly estimating Cff for studying urban landscapes in the southern region of Brazil through 14C time-integrated distribution mapping of ipê leaves. The method could also be used to augment greenhouse gas (GHG) emissions inventory studies trends in partitioning Cff from CO2 of bio-template sustainable sources.
... These research endeavors combine multiple observing and modeling techniques. Among these are ground-level atmospheric concentration measurement, satellite measurement of columnar concentration, inverse modeling, combustion flux monitoring, and urban socioeconomic modeling [11][12][13][14][15]. ...
Full-text available
The 'Hestia Project' uses a bottom-up approach to quantify fossil fuel CO2 (FFCO2) emissions spatially at the building/street level and temporally at the hourly level. Hestia FFCO2 emissions are provided in the form of a group of sector-specific vector layers with point, line, and polygon sources to support carbon cycle science and climate policy. Application to carbon cycle science, in particular, requires regular gridded data in order to link surface carbon fluxes to atmospheric transport models. However, the heterogeneity and complexity of FFCO2 sources within regular grids is sensitive to spatial resolution. From the perspective of a data provider, we need to find a balance between resolution and data volume so that the gridded data product retains the maximum amount of information content while maintaining an efficient data volume. The Shannon entropy determines the minimum bits that are needed to encode an information source and can serve as a metric for the effective information content. In this paper, we present an analysis of the Shannon entropy of gridded FFCO2 emissions with varying resolutions in four Hestia study areas, and find: (1) the Shannon entropy increases with smaller grid resolution until it reaches a maximum value (the max-entropy resolution); (2) total emissions (the sum of several sector-specific emission fields) show a finer max-entropy resolution than each of the sector-specific fields; (3) the residential emissions show a finer max-entropy resolution than the commercial emissions; (4) the max-entropy resolution of the onroad emissions grid is closely correlated to the density of the road network. These findings suggest that the Shannon entropy can detect the information effectiveness of the spatial resolution of gridded FFCO2 emissions. Hence, the resolution-entropy relationship can be used to assist in determining an appropriate spatial resolution for urban CO2 flux studies. We conclude that the optimal spatial resolution for providing Hestia total FFCO2 emissions products is centered around 100 m, at which the FFCO2 emissions data can not only fully meet the requirement of urban flux integration, but also be effectively used in understanding the relationships between FFCO2 emissions and various social-economic variables at the U.S. census block group level.
... Finally, the errors in the measurements in our study should be fully independent of the inverse modelling framework. The 1 ppm measurement error for FFCO 2 gradients between sites corresponds to typical values based on the analysis of air samples by accelerator mass spectrometry (AMS) for 14 CO 2 (2-3‰, (Vogel et al., 2010;Turnbull et al., 2014)) and by typical analyzers for continuous CO 2 samples (Chen et al., 2010;Turnbull et al., 2011). Apart from these errors, various fluxes that influence the atmospheric 14 CO 2 , such as those from cosmogenic production, ocean, biosphere and nuclear facilities, make the direct conversion into FFCO 2 gradients bear complex uncertainties whose typical values may exceed 1 ppm for some locations and periods of times (Hsueh et al., 2007;Bozhinova et al., 2013;Vogel et al., 2013). ...
Full-text available
National annual inventories of CO2 emitted during fossil fuel consumption (FFCO2) bear 5–10% uncertainties for developed countries, and are likely higher at intra annual scales or for developing countries. Given the current international efforts of mitigating actions, there is a need for independent verifications of these inventories. Atmospheric inversion assimilating atmospheric gradients of CO2 and radiocarbon measurements could provide an independent way of monitoring FFCO2 emissions. A strategy would be to deploy such measurements over continental scale networks and to conduct continental to global scale atmospheric inversions targeting the national and one-month scale budgets of the emissions. Uncertainties in the high-resolution distribution of the emissions could limit the skill for such a large-scale inversion framework. This study assesses the impact of such uncertainties on the potential for monitoring the emissions at large scale. In practice, it is more specifically dedicated to the derivation, typical quantification and analysis of critical sources of errors that affect the inversion of FFCO2 emissions when solving for them at a relatively coarse resolution with a coarse grid transport model. These errors include those due to the mismatch between the resolution of the transport model and the spatial variability of the actual fluxes and concentrations (i.e. the representation errors) and those due to the uncertainties in the spatial and temporal distribution of emissions at the transport model resolution when solving for the emissions at large scale (i.e. the aggregation errors). We show that the aggregation errors characterize the impact of the corresponding uncertainties on the potential for monitoring the emissions at large scale, even if solving for them at the transport model resolution. We propose a practical method to quantify these sources of errors, and compare them with the precision of FFCO2 measurements (i.e. the measurement errors) and the errors in the modelling of atmospheric transport (i.e. the transport errors). The results show that both the representation and measurement errors can be much larger than the aggregation errors. The magnitude of representation and aggregation errors is sensitive to sampling heights and temporal sampling integration time. The combination of these errors can reach up to about 50% of the typical signals, i.e. the atmospheric large-scale mean afternoon FFCO2 gradients between sites being assimilated by the inversion system. These errors have large temporal auto-correlation scales, but short spatial correlation scales. This indicates the need for accounting for these temporal auto-correlations in the atmospheric inversions and the need for dense networks to limit the impact of these errors on the inversion of FFCO2 emissions at large scale. More generally, comparisons of the representation and aggregation errors to the errors in simulated FFCO2 gradients due to uncertainties in current inventories suggest that the potential of inversions using global coarse-resolution models (with typical horizontal resolution of a couple of degrees) to retrieve FFCO2 emissions at sub-continental scale could be limited, and that meso-scale models with smaller representation errors would effectively increase the potential of inversions to constrain FFCO2 emission estimates.
... Carbon dioxide (CO 2 ) gas was produced from the pretreated residue of samples and purified for graphitization. Samples were combusted at 900°C for 4 h in evacuated, sealed quartz tubes with cupric oxide and silver wire (Turnbull et al 2014). The cupric oxide provides oxygen for the combustion and the silver isolates sulfur and halogens in a solid form. ...
Full-text available
Carbon isotopic evidence revealed Deepwater Horizon (DWH) oil entering coastal planktonic and lower terrestrial food webs. The integration of spilled oil into higher terrestrial trophic levels, however, remains uncertain. We measured radiocarbon (14 C) and stable carbon (13 C) in seaside sparrow (Ammodramus maritimus) feathers and crop contents. Lower 14 C and 13 C values in feathers and crop contents of birds from contaminated areas indicated incorporation of carbon from oil. Our results, although based on a small sample of birds, thus reveal a food-web link between oil exposure and a terrestrial ecosystem. They also suggest that the reduction in reproductive success previously documented in the same population might be due to the (direct) toxic effect of oil exposure, rather than to (indirect) ecological effects. We recommend future studies test our results by using larger samples of birds from a wider area in order to assess the extent and implications of DWH oil incorporation into the terrestrial food web.
... This new modeling framework, however, is still unable to reproduce the variability in observed Δ 14 CO 2 in the most polluted areas, which was also the case in the study by Riley et al. (2008). Plant samples can be useful for the investigation of point sources (Turnbull et al. 2014a), but not all studies yet try to quantify the effect of the variable plant growth and its effect on the Δ 14 CO 2 signature of the assimilated CO 2 . An additional complication when dealing with perennial plants (Park et al. 2013;Sakurai et al. 2013;Baydoun et al. 2015) could be the re-allocation of carbon assimilated from previous seasons for the initialization and maintenance of the current season growth. ...
Atmospheric Δ14CO2 measurements are useful to investigate the regional signals of anthropogenic CO2 emissions, despite the currently scarce observational network for Δ14CO2. Plant samples are an easily attainable alternative, which have been shown to work well as a qualitative measure of the atmospheric Δ14CO2 signals integrated over the time a plant has grown. Here, we present the 14C analysis results for 89 individual maize (Zea mays) plant samples from 51 different locations that were gathered in the Netherlands in the years 2010 to 2012, and from western Germany and France in 2012. We describe our sampling strategy and results, and include a comparison to a model simulation of the Δ14CO2 that would be accumulated in each plant over a growing season. Our model simulates the Δ14CO2 signatures in good agreement with observed plant samples, resulting in a root-mean-square deviation (RMSD) of 3.30‰. This value is comparable to the measurement uncertainty, but still relatively large (20–50%) compared to the total signal. It is also comparable to the spread in Δ14CO2 values found across multiple plants from a single site, and to the spread found when averaging across larger regions. We nevertheless find that both measurements and model capture the large-scale (>100 km) regional Δ14CO2 gradients, with significant observation-model correlations in all three countries in which we collected samples. The modeled plant results suggest that the largest gradients found in the Netherlands and Germany are associated with emissions from energy production and road traffic, while in France, the 14CO2 enrichment from nuclear sources dominates in many samples. Overall, the required model-based interpretation of plant samples adds additional uncertainty to the already relatively large measurement uncertainty in Δ14CO2, and we suggest that future fossil fuel monitoring efforts should prioritize other strategies such as direct atmospheric sampling of CO2 and Δ14CO2.
... the unsaturated zone, shallow groundwater). Monitoring above the surface in the low atmosphere could also provide spatial information of a leakage, as shown for example by Turnbull et al. (2014) , who examined the point source CO 2 emission of a gas treatment plant up to an altitude of about 100 m above ground. However, several challenges exist for this approach. ...
Full-text available
This synthesis discusses the emissions of carbon dioxide from fossil-fuel combustion and cement production. While much is known about these emissions, there is still much that is unknown about the details surrounding these emissions. This synthesis explores our knowledge of these emissions in terms of why there is concern about them; how they are calculated; the major global efforts on inventorying them; their global, regional, and national totals at different spatial and temporal scales; how they are distributed on global grids (i.e., maps); how they are transported in models; and the uncertainties associated with these different aspects of the emissions. The magnitude of emissions from the combustion of fossil fuels has been almost continuously increasing with time since fossil fuels were first used by humans. Despite events in some nations specifically designed to reduce emissions, or which have had emissions reduction as a byproduct of other events, global total emissions continue their general increase with time. Global total fossil-fuel carbon dioxide emissions are known to within 10 % uncertainty (95 % confidence interval). Uncertainty on individual national total fossil-fuel carbon dioxide emissions range from a few percent to more than 50 %. This manuscript concludes that carbon dioxide emissions from fossil-fuel combustion continue to increase with time and that while much is known about the overall characteristics of these emissions, much is still to be learned about the detailed characteristics of these emissions.
Full-text available
We adapt and integrate the Biome-BGC and Land Use in Rural New Zealand models to simulate pastoral agriculture and to make land-use change, intensification of agricultural activity and climate change scenario projections of New Zealand's pasture production at time slices centred on 2020, 2050 and 2100, with comparison to a present-day baseline. Biome-BGC model parameters are optimised for pasture production in both dairy and sheep/beef farm systems, representing a new application of the Biome-BGC model. Results show up to a 10% increase in New Zealand's national pasture production in 2020 under intensification and a 1-2% increase by 2050 from economic factors driving land-use change. Climate change scenarios using statistically downscaled global climate models (GCMs) from the IPCC Fourth Assessment Report also show national increases of 1-2% in 2050, with significant regional variations. Projected out to 2100, however, these scenarios are more sensitive to the type of pasture system and the severity of warming: dairy systems show an increase in production of 4% under mild change but a decline of 1% under a more extreme case, whereas sheep/beef production declines in both cases by 3 and 13%, respectively. Our results suggest that high-fertility systems such as dairying could be more resilient under future change, with dairy production increasing or only slightly declining in all of our scenarios. These are the first national-scale estimates using a model to evaluate the joint effects of climate change, CO2 fertilisation and N-cycle feedbacks on New Zealand's unique pastoral production systems that dominate the nation's agriculture and economy. Model results emphasise that CO2 fertilisation and N-cycle feedback effects are responsible for meaningful differences in agricultural systems. More broadly, we demonstrate that our model output enables analysis of decoupled land-use change scenarios: the Biome-BGC data products at a national or regional level can be re-sampled quickly and cost-effectively for specific land-use change scenarios and future projections.
Time-series radiocarbon measurements have substantial ability to constrain the size and residence time of the soil C pools commonly represented in ecosystem models. ¹⁴ C remains unique in its ability to constrain the size and turnover rate of the large stabilized soil C pool with roughly decadal residence times. The Judgeford soil, near Wellington, New Zealand, provides a detailed 11-point ¹⁴ C time series enabling observation of the incorporation and loss of bomb ¹⁴ C in surface soil from 1959–2002. Calculations of the flow of C through the plant-soil system can be improved further by combining the known constraints of net primary productivity (NPP) and ¹⁴ C-derived C turnover. We show the Biome-BGC model provides good estimates of NPP for the Judgeford site and estimates NPP from 1956–2010. Synthesis of NPP and ¹⁴ C data allows parameters associated with the rapid turnover “active” soil C pool to be estimated. This step is important because it demonstrates that NPP and ¹⁴ C can provide full data-based constraint of pool sizes and turnover rates for the 3 pools of soil C used in nearly all ecosystem and global C-cycle models.
Anyone involved with diffusion calculations becomes well aware of the strong dependence of maximum ground concentrations on the “effective stack height,” he. For most conditions χmax is approximately proportional to he−2, as has been recognized at least since 1936 (Bosanquet and Pearson). Making allowance for the gradual decrease in the ratio of vertical to lateral diffusion at increasing heights, the exponent is slightly larger, say χmax~ he−2.3. In inversion breakup fumigation, the exponent issomewhat smaller; very crudely, χmax~ he−1.5 In any case, for an elevated emission the dependence of χmax on he is substantial.
Time-series radiocarbon measurements have substantial ability to constrain the size and residence time of the soil C pools commonly represented in ecosystem models. 14C remains unique in its ability to constrain the size and turnover rate of the large stabilized soil C pool with roughly decadal residence times. The Judgeford soil, near Wellington, New Zealand, provides a detailed 11-point 14C time series enabling observation of the incorporation and loss of bomb 14C in surface soil from 1959-2002. Calculations of the flow of C through the plant-soil system can be improved further by combining the known constraints of net primary productivity (NPP) and 14C-derived C turnover. We show the Biome-BGC model provides good estimates of NPP for the Judgeford site and estimates NPP from 1956-2010. Synthesis of NPP and 14C data allows parameters associated with the rapid turnover "active" soil C pool to be estimated. This step is important because it demonstrates that NPP and 14C can provide full data-based constraint of pool sizes and turnover rates for the 3 pools of soil C used in nearly all ecosystem and global C-cycle models. © 2013 by the Arizona Board of Regents on behalf of the University of Arizona.
Greenhouse gas emissions are currently quantified from statistical data without testing the results against the actual increases of these gases in the atmosphere. This is like dieting without weighing oneself. Data are produced by greenhouse gas emitters of all sizes, from factory or farm to nation, and are quoted to high precision—yet misreporting occurs, whether by simple error, ignorance, or intention. But now scientists on both sides of the Atlantic are arguing that regulation of greenhouse gas emissions can have integrity only if verified by direct atmospheric measurements (1, 2).
[1] The 14C/C abundance in CO2(Δ14CO2) promises to provide useful constraints on regional fossil fuel emissions and atmospheric transport through the large gradients introduced by anthropogenic activity. The currently sparse atmospheric Δ14CO2 monitoring network can potentially be augmented by using plant biomass as an integrated sample of the atmospheric Δ14CO2. But the interpretation of such an integrated sample requires knowledge about the day‒to‒day CO2 uptake of the sampled plants. We investigate here the required detail in daily plant growth variations needed to accurately interpret regional fossil fuel emissions from annual plant samples. We use a crop growth model driven by daily meteorology to reproduce daily fixation of Δ14CO2 in maize and wheat plants in the Netherlands in 2008. When comparing the integrated Δ14CO2 simulated with this detailed model to the values obtained when using simpler proxies for daily plant growth (such as radiation and temperature), we find differences that can exceed the reported measurement precision of Δ14CO2(∼2‰). Furthermore, we show that even in the absence of any spatial differences in fossil fuel emissions, differences in regional weather can induce plant growth variations that result in spatial gradients of up to 3.5‰ in plant samples. These gradients are even larger when interpreting separate plant organs (leaves, stems, roots, or fruits), as they each develop during different time periods. Not accounting for these growth‒induced differences in Δ14CO2 in plant samples would introduce a substantial bias (1.5–2 ppm) when estimating the fraction of atmospheric CO2 variations resulting from nearby fossil fuel emissions.