Landscape and Urban Planning 148 (2016) 99–107
Contents lists available at ScienceDirect
Landscape and Urban Planning
journal homepage: www.elsevier.com/locate/landurbplan
Research paper
Does urban vegetation enhance carbon sequestration?
Erik Velasco a,∗ , Matthias Roth b , Leslie Norford a , Luisa T. Molina c
a
Singapore-MIT Alliance for Research and Technology (SMART), Center for Environmental Sensing and Modeling (CENSAM), 1 CREATE Way,
#09-03 CREATE Tower, Singapore 138602, Singapore
b
National University of Singapore (NUS), Department of Geography, Singapore, Singapore
c
Molina Center for Energy and the Environment (MCE2), La Jolla, CA, USA
h i g h l i g h t s
•
•
•
•
•
Limited evidence supports the carbon sequestration efficiency of urban vegetation.
Net CO2 flux measurements in urban areas suggest a limited greenery sink capacity.
Carbon sequestration depends on the characteristics of trees and pervious surfaces.
Soil respiration limits the potential of carbon sequestration by vegetation.
Greenery contributes −1.4% and 4.4% of the total CO2 flux at two suburban sites.
a r t i c l e
i n f o
Article history:
Received 20 March 2015
Received in revised form 16 October 2015
Accepted 5 December 2015
Keywords:
Carbon sequestration
Urban greenery
Urban forestry
Greenhouse gas emissions
Carbon dioxide
Eddy covariance
a b s t r a c t
Many cities are developing policies to promote greenery as a measure to reduce their net greenhouse gas
emissions. Studies suggest that urban forests may represent an important carbon reservoir. However, the
potential to directly remove carbon dioxide (CO2 ) from the atmosphere by urban vegetation is still poorly
supported by scientific evidence. Current assessments consider only the carbon accumulated by trees
and usually neglect the contribution from soil respiration and the emissions associated with greenery
management. Studies in mid-latitude cities suggest that the carbon uptake by urban vegetation is small
compared to the magnitude of the anthropogenic emissions. To investigate if the typically evergreen
vegetation in (sub)tropical cities has a larger potential for carbon sequestration, the CO2 flux data from two
residential neighborhoods of Singapore and Mexico City were analyzed. Results suggest that (sub)tropical
vegetation may act as either an emission source or sink depending on the species and characteristics of
the trees and the amount and conditions of pervious surfaces for soil respiration. The biogenic component
(vegetation and soil) was found to be a sink of 1 Mg km−2 day−1 of CO2 in Mexico City, but an emission
source of 0.8 Mg day−1 km−2 of CO2 in Singapore. The biogenic contribution to the total CO2 flux represents
−1.4% and 4.4% at both sites, respectively.
© 2015 Elsevier B.V. All rights reserved.
1. Introduction
Urban greenery has recently gained popularity as a climate
change adaptation/mitigation measure. Many city governments
have adopted policies promoting tree-planting, the preservation
of urban green spaces, and more recently green architecture (i.e.
green roofs and facades). The potential benefits and services provided by greenery to the urban ecosystem from a physical point of
view include reduction of greenhouse gas (GHG) emissions, thermal comfort, improved air quality, energy-use reduction, flood
∗ Corresponding author. Tel.: +65 6516 5236; fax: +65 66842118.
E-mail address: [email protected] (E. Velasco).
http://dx.doi.org/10.1016/j.landurbplan.2015.12.003
0169-2046/© 2015 Elsevier B.V. All rights reserved.
protection and improved runoff-water quality. From a social perspective, green spaces provide health and a range of recreational
and psychological benefits, create environmental awareness and
encourage positive actions toward climate change (US-EPA, 2008;
Pataki et al., 2011; Demuzere et al., 2014). The social benefits have
been relatively well documented. In contrast, some physical benefits are still poorly supported by scientific evidence. For example,
there is little data showing the effectiveness of urban vegetation to
reduce GHG emissions or concentrations of airborne pollutants.
The lack of data and models evaluated with observations, which
cover the large variability present among cities in terms of plant
species, urban morphology and climate setting, impede a proper
assessment of current greenery programs. There is some evidence
that vegetation through shading and transpiration can provide
100
E. Velasco et al. / Landscape and Urban Planning 148 (2016) 99–107
local cooling in hot regions, which translates into reduced energy
consumption for air conditioning and subsequently urban GHG
emissions (e.g., Akbari, 2002). Much less evidence is available to
demonstrate the direct removal of carbon dioxide (CO2 ) from the
atmosphere by urban vegetation. The few studies available from
mostly temperate cities in the US and Europe have estimated the
annual carbon sequestration at the scale of the entire city through
the application of allometric equations, relationships between
biomass (carbon stored) and physical dimensions (e.g., diameter
and height) of trees, and predictive growth models applied to tree
inventories (e.g., McPherson, Xiao, & Aguaron, 2013). The results
suggest that urban trees potentially have an important role in the
reduction of the carbon budget of cities. However, these studies
have not considered the contribution from soil respiration. Soil
CO2 efflux is mainly a result of autotrophic respiration by plant
roots and associated microorganisms, and heterotrophic respiration via microbial decomposition of soil organic matter (Hanson,
Edwards, Garten, & Andrews, 2000). These two biological sources
cannot be disassociated from the CO2 exchange of the aboveground
vegetation. A complete assessment of the contribution to carbon
sequestration by urban greenery needs to consider both the carbon
accumulated by trees and the soil respiration.
2015). Other studies have used measurements from a single tower
in combination with a detailed analysis of the land cover in the
sensor footprint as it changes with wind direction (e.g., residential
dwellings, parks, university campuses) (e.g., Kordowski & Kuttler,
2010; Järvi et al., 2012). Yet other studies have performed source
attribution of the measured fluxes using bottom-up approaches
(i.e. emission factors and activity data) to estimate the contribution from anthropogenic sources (e.g., vehicular traffic, domestic
heating, etc.) in combination with models of the biogenic CO2 flux
associated with photosynthesis and ecosystem respiration (e.g.,
Nemitz, Hargreaves, McDonald, Dorsey, & Fowler, 2002; Soegaard &
Møller-Jensen, 2003; Crawford & Christen, 2015). A common result
of all these studies carried out in mid-latitude cities is that the net
carbon uptake by urban greenery is small compared to the magnitude of the anthropogenic emissions. For example, Crawford and
Christen (2015), based on a comprehensive spatial analysis using
a high resolution land cover database and empirical models, found
that the biogenic component (vegetation and soil) offsets only 1.7%
of the total CO2 flux emitted to the atmosphere in a residential
neighborhood of Vancouver, Canada.
4. Evaluation of CO2 sequestration by urban vegetation in
(sub)tropical cities
2. Direct measurements of the CO2 exchange
Carbon dioxide flux measurements by eddy covariance (EC) (i.e.
flux towers) have been widely used to investigate the carbon cycle
in natural ecosystems, such as forests, crops, grasslands, etc. These
micrometeorological measurements include contributions from
aboveground vegetation and underground (soil) processes. Observations from over 400 sites indicate that the annual net carbon
exchange results from imbalances in the carbon uptake by photosynthesis and release by ecosystem respiration (from vegetation
and soil), which scale closely with one another (Baldocchi, 2008).
A key finding from this research that is potentially important for
urban greenery is that recently disturbed ecosystems tend to lose
carbon, unlike old-growth forests that usually act as carbon sinks
(e.g., Luyssaert et al., 2008; Stephenson et al., 2014). Since the late
nineties the EC method has also been used in urban environments
to measure CO2 exchange (Velasco & Roth, 2010). Until now, over 30
urban flux towers have been deployed, mostly in mid-latitude cities
located in the northern hemisphere. Instrumentation installed at
the top of these towers is able to directly measure the total CO2
exchange (including all major and minor anthropogenic and natural emission sources and sinks) within a sensor footprint that is
usually representative of the local scale (i.e. hundreds of meters
or similar to the size of a complete neighborhood). Results show
that urban areas are generally net sources of CO2 , in some cases
with reduced daytime magnitudes during the growing season when
anthropogenic emissions are partially offset by photosynthesis,
especially in suburban neighborhoods with abundant vegetation
(Velasco & Roth, 2010). Only one study conducted in a densely vegetated suburban area of Baltimore has reported an annual net carbon
uptake (Crawford, Grimmond, & Christen, 2011).
3. Methods to determine the CO2 flux associated with
urban greenery
Observations from a single EC flux tower are not sufficient to determine the influence of vegetation on the net
neighborhood-scale CO2 flux. A few studies have therefore
conducted simultaneous EC flux measurements over neighborhoods with different land cover characteristics within the same
metropolitan area (e.g., Coutts, Beringer, & Tapper, 2007; Bergeron
& Strachan, 2011; Ramamurthy & Pardyjak, 2011; Ward et al.,
The potential for carbon sequestration by urban vegetation has
yet to be investigated in (sub)tropical cities, which is similar to
other aspects of the urban ecology. Tropical and many subtropical
forests are usually evergreen and therefore have a larger potential
for CO2 assimilation than boreal and temperate forests (Baldocchi,
2008). To investigate whether (sub)tropical urban greenery indeed
sequesters more CO2 than its mid-latitude counterparts, respective
data from two EC flux towers installed in a residential neighbourhood of Singapore and a commercial/residential neighbourhood of
Mexico City (Velasco et al., 2013, 2014), have been analysed. The
former is located in a tropical rainforest climate whereas the latter
represents a subtropical highland climate (Table 1).
5. Methods
Two independent approaches were used to quantify the direct
removal of CO2 from the atmosphere by vegetation in the case of
Singapore and one in the case of Mexico City. Fig. 1 introduces both
approaches. For the residential neighbourhood of Singapore, the
first approach was based on the comparison of CO2 fluxes measured
directly by EC with emissions estimated by bottom-up approaches.
The latter includes contributions from vehicular traffic, household
combustion, soil respiration and human breathing. The soil CO2
efflux was estimated using the empirical Q10 model based on van’t
Hoff’s equation which assumes that soil respiration responds only
to temperature changes in the absence of soil moisture limitations
(Mahecha et al., 2010). The biogenic component of the CO2 flux from
aboveground vegetation was calculated as the difference between
estimated emissions and measured fluxes (see Velasco et al., 2013
for details). A tree survey was conducted for the second approach
which estimates the annual CO2 sequestration using allometric
equations and an alternative model of the metabolic theory of ecology for tropical forests (Muller-Landau et al., 2006). This model
predicts the growth rate of woody trees as a function of their size.
Palm trees, banana plants and turfgrass were also included in the
survey, but their annual CO2 uptake was obtained from published
average annual growth rates.
In the case of Mexico City the CO2 flux measured by EC was
compared with emissions taken from the gridded local emissions
inventory for the footprint covered by the flux tower. After adding
the contributions from human and soil respiration, it was found that
E. Velasco et al. / Landscape and Urban Planning 148 (2016) 99–107
101
Table 1
Summary of suburban building and vegetation characteristics in the studied neighborhoods of Escandón, Mexico City and Telok Kurau, Singapore.
Escandón, Mexico City
Telok Kurau, Singapore
Climate (Köppen classification)
Ambient temperature (◦ C)
Mean daily min.—Daily mean—Mean daily max.
Annual rainfall (mm)
Population density (inhabitants km−2 )
Local climate zonea
Subtropical highland (Cwb)
10–16–24
Tropical rainforest (Af)
25–27–32
820
8038
Compact mid-rise
2340
7491
Compact low-rise
Fraction of area covered by buildings, other impervious surfaces
(roads, sidewalks, etc.) and pervious (green) surfaces.
Mean building height and standard deviation (m)
Mean albedo and standard deviationb
Number of trees (trees km−2 )
Mean tree height and standard deviation (m)
Fraction of large trees (DBH ≥ 20 cm)c
Species
57%, 37%, 6%
39%, 46%, 15%
9.7 ± 4.6
0.112 ± 0.007
5276
10.4 ± 6.0
64%
97.2% woody trees
2.4% yuccas
0.4% palms
6330 (1725)
9.9 ± 4.0
0.158 ± 0.003
5856
7.3 ± 3.7
37%
60.6% woody trees
34.1% palms
5.3% banana plants
6337 (1727)
CO2 (carbon) storage (Mg km−2 )
a
b
c
Following classification proposed by Stewart and Oke (2012).
Based upon observations of incoming and outgoing shortwave radiation (K↑ and K↓) when solar elevation was >20◦ .
DBH—diameter at breast height.
the inventory-based approach over-predicted the observed flux as
a consequence of a probable overestimation of the traffic volume
for that part of the city. This over-prediction prevented the estimation of the aboveground vegetation flux as the difference between
estimated emissions and measured fluxes (see Velasco et al., 2014
for details). The CO2 sequestered by vegetation was obtained by
applying biomass allometric equations and a growth predictive
model to the trees inventoried within the study domain. The growth
prediction model of Peper, McPherson, and Mori (2001a, 2001b)
developed for trees in cities located in southern California was
selected because of the similarity of the tree species, warm weather
and almost year-round growing season in both places. The area
covered by turfgrass was negligible.
The CO2 flux source attribution for the Mexican neighbourhood
was then performed assuming that emissions from anthropogenic
sources other than vehicular traffic (i.e. households, light industries,
stores and workshops) are properly predicted by the emissions
inventory. The traffic contribution was therefore calculated as the
difference between the measured EC flux and the predicted emissions from anthropogenic sources (excluding vehicles) plus the
contributions from human and soil respiration and CO2 sequestered
by vegetation. A previous study also based on EC flux measurements
Fig. 1. Methods used to quantify the CO2 uptake by urban vegetation. Method 1 is based on the comparison of CO2 fluxes measured by EC with anthropogenic emissions
estimated by bottom-up approaches and soil respiration efflux. This method was only applied in Singapore. Inconsistencies in the vehicular traffic emissions reported for
the Mexican neighborhood prevented its application. Method 2 is based on the application of biomass allometric equations and growth prediction models to the trees
inventoried within the study neighborhoods. Because of the lack of species-related biomass information for tropical urban trees, the carbon stored in boles and branches by
Singapore’s trees was estimated using equations for primary (large trees) and secondary (small trees) tropical forests. Leaf biomass was calculated following the relationship
of Chave et al. (2008). For the case of Mexico City, aboveground biomass allometric equations developed for trees in cities from southern California were selected because
of the similarity of species and warm weather. The root biomass in both locations was obtained from the model proposed by Cairns, Brown, Helmer, and Baumgardner
(1997). A factor of 1.2 was applied to the biomass production results according to Clark et al. (2001) to account for the sequestered carbon that is used for the maintenance,
reproduction and defense of the plant instead of biomass production, The CO2 uptake by palm trees, turfgrass and other minor species was obtained from published
average annual growth rates. For details see Velasco et al. (2013, 2014), Aguaron and McPherson (2012), Chave et al. (2005), van Breugel, Ransijn, Craven, Bongers, and Hall
(2011).
102
E. Velasco et al. / Landscape and Urban Planning 148 (2016) 99–107
Fig. 2. Average diurnal patterns of the energy budget in the residential neighborhoods of Telok Kurau, Singapore ((a)–(c)) and Escandón, Mexico City ((d)–(f)) for the
climatological seasons experienced in both cities. The sensible (QH ) and latent (QE ) heats were measured by EC. The four components of the net radiation (Q*) were directly
measured in Singapore. The incoming and outgoing shortwave radiation were only measured in Mexico City, the longwave components were estimated using the approach
described in Velasco et al. (2011) based on the model of Offerle, Grimmond, and Oke (2003) using cloudiness data from the city airport. For the case of Singapore, the
anthropogenic heat (QF ) was previously calculated by Quah and Roth (2012), and the storage heat (Qs ) was therefore calculated as the residual of the sum of Q* and QF and
the other two energy components. For the case of Mexico City, no estimation of QF was available and was not included in the computation of Qs . The data presented here
covers 21 and 15 months of measurements in Singapore (Oct. 2010–Jun. 2012) and Mexico City (Jun. 2011–Sep. 2012), respectively.
determined that the predicted emissions in the two-year preceding emissions inventory were essentially correct (Velasco et al.,
2009). With the exception of vehicular traffic, no major changes
were reported in the emissions inventory for that period.
In cities, the heterogeneous surface and multiple and varied
sources/sinks challenge the application of the EC method. However, by selecting a monitoring site with relatively homogenous
roughness elements (i.e. buildings and trees) and well-distributed
emission sources and sinks, the EC method is an effective alternative to investigate the urban carbon exchange at the local scale.
The instrument set-up and data post-processing are very similar to
those used in natural ecosystems (Velasco & Roth, 2010). Studies
which estimate net ecosystem exchange in forests need to consider
CO2 storage changes inside the canopy and apply gap filling procedures during periods when sensors fail or measuring conditions
are unfavorable (e.g., rain, strong atmospheric stability) to obtain
continuous time series. In densely built-up locations positive sensible heat flux off the warm surfaces and storage heat release at night
maintain fully turbulent conditions, reducing the magnitude of CO2
storage in the urban canopy layer to levels well below the intra-day
variability of hourly fluxes (Crawford & Christen, 2014). This intraday variability is primarily driven by human causes and is larger
than the generally small seasonal variability in (sub)tropical cities;
ensemble averages can therefore be computed without a need for
gap-filling. Further details about the EC flux measurements used
here are available in Velasco et al. (2013, 2014).
5.1. Complementary energy flux observations
The instrumentation for measuring CO2 fluxes measures simultaneously fluxes of energy. It is therefore common that studies
on the carbon cycle based on flux towers investigate also the
E. Velasco et al. / Landscape and Urban Planning 148 (2016) 99–107
103
6. Results and discussion
The net CO2 flux measured at both sites by EC is discussed at
the beginning of this section. The different diurnal patterns are
analyzed as a function of the neighborhoods’ characteristics and
human activities. Next, the CO2 flux partitioning according to emission sources and sinks is presented. Emphasis is given to the results
of CO2 uptake and soil respiration obtained at both locations to
determine the full role of greenery on the urban CO2 flux. The magnitude of the carbon uptake is then compared to that of forests
of similar species and climate, and the fate of the carbon stored
by urban vegetation is discussed. For a detailed explanation of the
anthropogenic emissions and the use of EC flux towers as a climate
change mitigation management tool to evaluate the accuracy of
bottom-up emission inventories, the reader is referred to Velasco
et al. (2013, 2014).
6.1. CO2 exchange
Fig. 3. Diurnal variability of CO2 fluxes measured by EC in the residential neighborhoods of Telok Kurau, Singapore (a) and Escandón, Mexico City (b) for different
seasons and days of the week. The gray shaded areas represent ± 1 standard deviation from the weekday flux average and gives an indication of the day-to-day
variability at every hour of the day. The measurements covered the periods indicated
in Fig. 2.
energy partitioning. Only small variations were observed in the
diurnal patterns of the energy fluxes (Fig. 2) during the climatological seasons experienced in both sites, especially compared to
those in temperate cities. The benign climatological conditions in
(sub)tropical locations explains the lack of clear seasonal patterns
in the energy balance, as well as in the CO2 flux. As shown in Fig. 3,
no major seasonal variability was found in the CO2 flux measured
in either neighborhood.
The differences between the energy components from both sites
are explained by their locations and urban morphologies. In addition to the latitude and altitude, the net radiation (Q*) depends also
on the cloudiness and atmospheric pollution. The lower latent heat
(QE ) in Mexico City is explained by its drier climate; the annual
average rainfall is about a third of that in Singapore. The sensible
heat (QH ) is the energy component with the smallest difference
between both sites, in contrast to the heat stored in the urban
surface (Qs ). The storage heat is slightly lower than QH during
daytime in the low-rise neighborhood of Singapore studied here,
but clearly higher in the neighborhood of Mexico City. This difference responds to the fractions of area covered by buildings and
roads, and the characteristics and size of those buildings. The builtup and impervious surfaces represent 94% and 85% of the plan area
of those neighborhoods in Mexico City and Singapore. Although the
release of heat in the Mexican neighborhood is twice as large as in
the Singapore neighborhood, in both cases it is enough to maintain
turbulent conditions throughout the night (Velasco et al., 2013,.,
2014), as mentioned above.
The Mexico City and Singapore neighborhoods are net CO2
sources of 24,500 and 6500 Mg km−2 year−1 , respectively. The
abundant commercial activities and heavy traffic present in the
Mexico City neighborhood are mainly responsible for the difference
compared to the Singapore neighborhood. Clear diurnal patterns
of CO2 flux with morning and evening peaks in phase with the
rush-hour traffic were observed during weekdays (Singapore and
Mexico City) and weekends (Singapore) (Fig. 3) and reveals the
importance of vehicular traffic as the main emission source. The
diurnal pattern observed in Mexico City is similar to those reported
for densely built-up locations and city centers elsewhere (e.g.,
Matese, Gioli, Vaccari, Zaldei, & Miglietta, 2009; Pawlak, Fortuniak,
& Siedlecki, 2011; Liu, Feng, Järvi, & Vesala, 2012) where the CO2
flux is always positive. The pattern on weekdays follows the shape
of a large mountain with two small peaks at the beginning and end
of the working day. Unlike in Singapore, the weekend CO2 fluxes in
Mexico City were clearly lower due to less traffic and anthropogenic
activities. The diurnal pattern observed in Singapore resembles the
shape of a valley surrounded by two mountains. Similar to suburban neighborhoods with abundant vegetation in mid-latitude
cities during the growing season (e.g., Bergeron & Strachan, 2011;
Buckley, Mitchell, McHale, & Millard, 2014; Ward et al., 2015), the
fluxes become negative around midday, more frequently so during
the NE monsoon season. The seasonal variability is due to shifts
in the sensor footprint following the prevailing monsoon winds
resulting in slightly different CO2 source and sink distributions during the NE and SW monsoon seasons, respectively (Velasco et al.,
2013).
6.1.1. Anthropogenic CO2 flux contribution
As expected vehicular traffic, including passenger cars, light and
heavy good vehicles, buses, motorcycles and taxis, is the main
contributor of CO2 to the atmosphere in both neighborhoods as
shown in Table 2. The sum of emissions from households, commercial establishments and light industries is the second major
contribution category in the neighborhood of Mexico City. The
residential nature and reduced number of commercial establishments in the Singapore neighborhood makes human metabolic
respiration the second largest source of CO2 while it is the third
largest source in the Mexican neighborhood. Although the per
capita emission from human breathing represents a tiny fraction of
the total flux, the high population density in both neighborhoods
(>7000 inhabitants km−2 , see Table 1) makes human respiration an
important component to consider. The inflow of people, especially
during daytime when the population density decreases significantly in the Singapore neighborhood, explains the difference in
the estimated contribution by human breathing between both sites.
104
E. Velasco et al. / Landscape and Urban Planning 148 (2016) 99–107
Table 2
Partitioning of CO2 fluxes during weekdays in the residential neighborhoods of Escandón, Mexico City and Telok Kurau, Singapore. Percentages are contributions of each
source/sink to the total flux. Magnitude of biogenic (i.e. vegetation and soil) fluxes is the same on weekdays and weekends.
Flux contributor
Escandón, Mexico City
Telok Kurau, Singapore
Vehicular traffic CO2 (carbon) emissions (Mg km−2 day−1 )
53.2 (14.5)
71.8%
17.8 (4.9)
24.0%
4.15 (1.1)
5.6%
−1.5 (−0.4)
−2.0%
0.4 (0.1)
0.6%
−1.0 (−0.3)
−1.4%
74.1 (20.2)
100%
12.8 (3.5)
71.6%
1.3 (0.4)
7.2%
3.0 (0.8)
16.8%
−1.4 (−0.4)
−7.8%
2.2 (0.6)
12.2%
0.8 (0.2)
4.4%
17.9 (4.9)
100%
Point sources and housing CO2 (carbon) emissions (Mg km−2 day−1 )a
Human respiration CO2 (carbon) (Mg km−2 day−1 )
Aboveground vegetation CO2 (carbon) uptake (Mg km−2 day−1 )b
Soil CO2 (carbon) efflux (Mg km−2 day−1 )
Total biogenic CO2 (carbon) flux (Mg km−2 day−1 )c
Total CO2 (carbon) flux (Mg km−2 day−1 )
a
Point sources include workshops, restaurants, dry cleaners, bakeries, and any other commercial establishment. Housing emissions are restricted to natural gas and
liquefied petroleum gas burning for cooking and water heating in Singapore and Mexico City, respectively.
b
A negative sign indicates sequestration.
c
Biogenic flux including contributions from aboveground vegetation and soil respiration. A negative (positive) sign indicates sequestration (emission).
Similarly, the lower population density during daytime explains the
smaller contribution from households in Singapore. The respective
emissions are due to the consumption of fossil fuels (natural gas in
Singapore, liquefied petroleum gas in Mexico City) for cooking and
water heating.
6.1.2. Biogenic CO2 flux contribution
Because of heavy traffic and other intense anthropogenic activities the available vegetation in the neighborhood of Mexico City
was unable to significantly offset daytime emissions. In contrast,
with essentially the same amount of biomass or carbon storage
(Table 1), the much less intense anthropogenic emissions in the
suburban neighborhood studied in Singapore were almost counterbalanced by sequestration around noon resulting in a very small
net CO2 flux at that time (Fig. 3). The two approaches used to
quantify the carbon sequestration in Singapore agree within 2% of
each other (510 and 500 Mg km−2 year−1 , respectively) and suggest that aboveground vegetation sequesters 7.8% of the total
emitted CO2 (Table 2). The approach based on EC flux measurements and emission estimates using emission factors and activity
data yields a net CO2 sequestration rate of 1.4 Mg km−2 day−1
calculated as the difference between the daily uptake by photosynthesis (3.95 Mg km−2 ) and release by plant respiration at night
(2.55 Mg km−2 ). However, when soil respiration is included, the
biogenic component becomes an emission source accounting for
4.4% of the total CO2 flux to the atmosphere (Table 2). For the
neighborhood in Mexico City a CO2 uptake of 1.5 Mg km−2 day−1
(541 Mg km−2 year−1 ) was estimated by allometry, which represents 2.0% of the observed flux by EC. Due to the large extension
of impervious surfaces, soil respiration contributes only 0.6%, and
the biogenic component takes up only 1.4% of the total CO2 flux
(Table 2).
Excluding the CO2 flux from the soil, the annual carbon uptake
by aboveground vegetation of these two (sub)tropical neighborhoods ranks at the top compared to sequestration rates reported
in the literature for urban areas in temperate cities. For example,
McPherson et al. (2013) estimated an annual CO2 sequestration
of 312 and 95 Mg year−1 for medium to high density residential
areas of Sacramento and Los Angeles, California, respectively. Both
annual rates are about 60% and 20% of those reported here for
Singapore and Mexico City.
Although there are more trees per unit area at the Singapore
site (by 11%), the Mexico City aboveground vegetation assimilates
slightly more CO2 (6%). This is primarily due to the predominance
of large woody trees at the latter site, which are capable of
sequestering more CO2 . The carbon uptake of palms, which are
more common in Singapore, is limited by their small wood specific
density, which on average is less than half of that of woody trees.
Defining large trees as those with diameter at breast height (DBH)
≥ 20 cm, 18 palms sequester the same amount of CO2 as one
large woody tree in the Singapore context (Velasco et al., 2013).
Similarly, 23 small trees are needed to replace one large tree
in terms of CO2 uptake (Velasco et al., 2013). This is consistent
with the finding that ecosystems sequestering the most carbon
are usually evergreen mature forests (Baldocchi, 2008). However,
when considering the total biogenic CO2 flux, the Singapore site
is a net emitter, which is primarily due to its larger (by 2.5 times)
proportion of pervious surfaces available for soil respiration. It
is important to recall that soil respiration is the main emission
source in natural ecosystems (Mahecha et al., 2010).
These findings are not surprising given that natural ecosystems
can act as emission sources or sinks of carbon depending on their
characteristics, age, management regime and climate. Forests that
have experienced recent disturbance via logging, fire, drainage or
wind-throw are prone to lose carbon by ecosystem respiration
(Luyssaert et al., 2007; Baldocchi, 2008).
The above results point to a limited or null impact on carbon
mitigation by urban greenery when both, carbon sequestration by
photosynthesis and respiration by trees and soil are considered. In
summary:
• In the case of Mexico City, the total biogenic component of the
CO2 flux in the studied neighborhood is a sink due to the presence
of large woody trees and the absence of pervious surface cover.
However, its magnitude is insignificant and does not offset the
anthropogenic emissions.
• Abundant trees in the neighborhood of Singapore remove a
significant fraction of the anthropogenic CO2 emissions by photosynthesis. However, soil efflux from the perennial warm and
humid soil in the extensive green areas (e.g. lawns and parks)
cancels such carbon uptake, making the biogenic component a
net emission source.
6.2. Magnitude of the urban carbon uptake compared to that of
forests
Synthesis studies from global networks of CO2 flux measurement systems have revealed that a large amount of the terrestrial
carbon sequestration is achieved by forests, with the duration
of the growing season being one of the main parameters driving
E. Velasco et al. / Landscape and Urban Planning 148 (2016) 99–107
the carbon uptake (Luyssaert et al., 2007; Baldocchi, 2008). The
tropical and subtropical trees in Singapore and Mexico City have
the benefit of being evergreen in addition to often being well
irrigated and fertilized. However, their total carbon uptake is
small compared to the magnitude of the anthropogenic emissions.
Given the lack of data for forests near Mexico City, the mean
CO2 uptake of 1393 ± 268 Mg km−2 year−1 reported by Luyssaert
et al. (2007) from eight Mediterranean warm evergreen forests
is used as a reference because of their similar characteristics, to
calculate a forest area which would need to be 16–23 times larger
to offset the anthropogenic emissions of the Mexican neighborhood. Similarly, based on the mean CO2 sequestration rate of
1478 ± 374 Mg km−2 year−1 from seven tropical humid evergreen
forests reported by the same authors, a forest area 4–6 times larger
is needed to assimilate the anthropogenic carbon emitted in the
Singapore neighborhood.
Considering the entire city and based on the same sequestration rates reported by Luyssaert et al. (2007), a forest between
15 and 18 times the size of Mexico City (1486 km2 , considering only the Federal District) is needed to offset the emission of
30,731 Gg CO2eq (CO2 accounted for 81%) reported for 2012 in the
official emissions inventory (SMA-GDF, 2013). For Singapore, a forest area that is 30–50 times larger than the city-state (710 km2 )
is needed to offset the 38,790 Gg CO2eq (CO2 accounted for 97%)
reported for 2000 as part of its Second National Communication
to the United Nations Framework Convention on Climate Change
(National Environmental Agency (NEA), 2010).
6.3. Fate of the carbon stored by urban greenery
Over long timescales the fate of the carbon stored in the urban
ecosystem depends on the degree of urban expansion, greenery
management and carbon allocation to biomass and soil organic
matter. The bulk of carbon is allocated to the production of biomass
in foliage, wood and roots. Carbon incorporated into wood has a residence time of years to centuries within living trees, whereas the
carbon deposited in the foliage and fine roots has much shorter residence times in the order of months to years (Luyssaert et al., 2007).
Each year part of the standing biomass is transferred to the soil
as litter and/or removed from the urban ecosystem through pruning and debris collection by the municipal greenery and cleaning
service. Intense pruning programs, as well as urban planning decisions where large and mature trees are removed or replaced by
young trees, accelerate and enhance the return to the atmosphere
of all or part of the carbon that has been accumulated over the years.
In some cases, this return occurs as non-respiratory CO2 fluxes (e.g.,
debris incineration).
6.3.1. Carbon stored by vegetation: Allocation and spatial
distribution
The vegetation in the Singapore neighborhood stores
1727 Mg C km−2 in woody trees (95.2%), palm trees (2.7%),
turfgrass (2.1%) and banana plants (0.004%). Boles and branches
account for 83.4% of the carbon stored in woody trees, roots
and leaves 15.4% and 1.2%, respectively (Velasco et al., 2013).
This carbon allocation in trees is similar to that reported in the
literature for tropical forests (e.g., Poorter et al., 2011). Despite the
differences in land cover and tree species, the Mexico City neighborhood stores essentially the same amount of carbon per unit of
area (1725 Mg C km−2 ). Aboveground vegetation accounts for 77%
of the carbon and roots 23%. The contribution from plants other
than woody tress is negligible. For instance, yucca plants (Yucca
orientalis), the most abundant non-woody species, account for 2.5%
of the trees, but store only 0.4% of the total carbon in vegetation
(Velasco et al., 2014). The carbon stored by vegetation in the two
densely populated (sub)tropical neighborhoods of Singapore and
105
Mexico City is similar to that reported for low density residential
areas of US cities, but clearly higher than measured in medium to
high density districts (McPherson et al., 2013).
As mentioned above, the predominance of large woody trees
(64%) in the neighborhood of Mexico City enhances the annual
carbon sequestration and the carbon storage in the long term. In
the case of Singapore, although large trees represent only 36.8%
of all trees, they contain 95.3% of the tree biomass and thus carbon. Fourteen Ahuhuete trees (Taxodium mucronatum) located in a
small park within the flux tower footprint of the Mexican neighborhood demonstrate the importance of large/old trees in the urban
carbon budget. According to our estimations based on allometric equations and confirmed by local dwellers and archives of the
city, the youngest of those trees is no less than 50 years old. Six
of them are apparently older than 100 years. True to their high
age, these trees contain the highest individual amount of CO2
(7.5 ± 6.4 Mg tree−1 ) and have the largest CO2 sequestration rate
(428 ± 183 kg tree−1 year−1 ). Despite representing only 1.4% of all
trees, they store and sequester 8.4% and 5.4%, respectively of the
total carbon in the neighborhood. On average, their individual mass
was equivalent to the mass of 69 small trees (i.e., DBH < 20 cm), and
their annual carbon gain was equivalent to that of nearly nine small
trees (Velasco et al., 2014).
Changes in urbanization can be tracked through the age, size,
physical condition, density and spatial distribution of trees. In case
of the Mexico City neighbourhood 4% of the actual trees date back
from pre-urbanization times (i.e. > 60 years) and 6% were planted
during the main urbanization period, 40–60 years ago. The majority
of trees, 60%, were planted after the neighbourhood was established
between 10 and 40 years ago. The remaining 30% of the trees are
younger than 10 years old.
In both neighborhoods the trees with the highest carbon stocks
and sequestration rates are generally located along the main roads
and in public parks. This pattern is more evident in Singapore,
where in recent years large trees have been replaced by young trees
and palms to make space for new constructions and carparks. In
general, the spatial distribution of species depends on who performs the greenery management. Trees in public spaces and along
roads are maintained by municipal workers, local dwellers often
take care of trees along secondary roads and always of those inside
their premises. The variety of species is therefore usually larger
along secondary roads and in private lawns. Three of the five most
common species in the Mexico City neighborhood correspond to
species used for reforestation programs. Evergreen ashes (Fraxius uhdei), box elders (Acer negundo) and quaking aspens (Populus
tremuloides) are preferred because of their physical characteristics and resilience and adaptability to the local weather and soil
properties.
6.3.2. Carbon in soil
The carbon in soil is subject to decomposition by heterotrophic
respiration. This process includes decomposition of fresh litter, but
also of organic matter accumulated over the years (Luyssaert et al.,
2007). In the case of urban ecosystems the role the ubiquitous
impervious surfaces play in the soil efflux is not yet clear. Some
authors suggest that soils beneath impervious surfaces and next to
roadways and buildings are prone to lose their capacity to retain
and store carbon as a consequence of direct release to the atmosphere, aqueous losses and severe degradation during construction
(e.g., Raciti, Hutyra, & Finzi, 2012). Urban soils are often highly compacted and exposed to abrupt physical and chemical barriers to
rooting depth and contamination by construction debris (De Kimpe
& Morel, 2000). In contrast, other authors suggest that organic carbon underneath impervious surfaces does not decompose because
of the lack of oxygen, and thus old towns may contain important
pools of carbon in deep layers (e.g., Churkina, 2012). In urban forests
106
E. Velasco et al. / Landscape and Urban Planning 148 (2016) 99–107
the gain or loss of carbon from soil is associated with the climate
and greenery management (i.e. fertilization, irrigation, debris collection). Some studies have found that soils in public parks and
managed lawns of temperate cities have the capacity to accumulate an important amount of carbon compared to native soils (e.g.,
Pouyat, Yesilonis, & Nowak, 2006; Pouyat, Yesilonis, & Golubiewski,
2009). Although the carbon content in urban soils can be quite variable, urban forests may represent an important carbon reservoir. In
US cities, the amount of belowground carbon is estimated to be 2–4
times larger than the carbon aboveground (Pouyat et al., 2006).
So far the carbon storage in soils of (sub)tropical cities has
not been assessed. In the case of the Singapore neighborhood
the intense soil respiration enhanced by the prevailing warm and
humid conditions may inhibit the accumulation of carbon. In the
case of the Mexican neighborhood the small extent of pervious surfaces (6%) may not be sufficient to take up a significant amount of
organic matter. Clearly, measurements of the organic carbon content are needed to evaluate the carbon storage capacity of these
and other cities.
7. Conclusions
No method currently exists that can directly evaluate the CO2
uptake by urban greenery. All approaches described herein rely
on indirect estimates and are prone to uncertainties (e.g., McHale,
Burke, Lefsky, Peper, & McPherson, 2009). However, our studies in
two (sub)tropical cities together with those from temperate cities,
suggest that the impact of urban vegetation to reduce GHG emissions directly through carbon sequestration is very limited or null.
When considering vegetation and soil together, the biogenic component was found to reduce the total CO2 flux in a neighborhood
of Mexico City by 1.4%, but add 4.4% extra CO2 in the case of a
neighborhood in Singapore. A more complete assessment should
include emissions associated with greenery management (i.e. pruning, mowing, watering, fertilizing, debris removing, etc.) which
could further offset any carbon reduction (e.g., Townsend-Small
& Czimczik, 2010). Climate change mitigation policies based on
promoting tree-planting, preservation of green spaces, and green
architecture may overestimate their GHG reduction goals if the
complete biogenic component (vegetation and soil) and its associated maintenance activities are not properly considered. Although
there are many environmental and social benefits to urban greenery, current research points to a limited role as an effective measure
to enhance carbon sequestration.
Acknowledgements
The research in Singapore was supported by the National
Research Foundation Singapore through the Singapore MIT Alliance
for Research and Technology’s CENSAM research program and a
National University of Singapore grant to MR (research grant R109-000-091-112). The research in Mexico City was supported by
the National Institute of Ecology and Climate Change and the Air
Quality Monitoring Network of Mexico City of the Ministry of Environment of the Federal District Government.
References
Aguaron, E., & McPherson, G. (2012). Comparison of methods for estimating carbon
dioxide storage by Sacramento’s urban forest. In R. Lal, & B. Augustin (Eds.),
Carbon sequestration in urban ecosystems. Netherlands: Springer.
Akbari, H. (2002). Shade trees reduce building energy use and CO2 emissions from
power plants. Environmental Pollution, 116, S119–S126.
Baldocchi, D. (2008). Breathing of the terrestrial biosphere: Lessons learned from a
global network of carbon dioxide flux measurement systems. Australian Journal
of Botany, 56, 1–26.
Bergeron, O., & Strachan, I. B. (2011). CO2 sources and sinks in urban and suburban
areas of a northern mid-latitude city. Atmospheric Environment, 45, 1564–1573.
Buckley, S. M., Mitchell, M. J., McHale, P. J., & Millard, G. D. (2014). Variations in
carbon dioxide fluxes within a city landscape: Identifying a vehicular influence.
Urban Ecosystems, 17, 1–20.
Cairns, M. A., Brown, S., Helmer, E., & Baumgardner, G. A. (1997). Root biomass
allocation in the world’s upland forests. Oecologia, 111, 1–11.
Chave, J., Andalo, C., Brown, S., Cairns, M. A., Chambers, J. Q., Eamus, D., et al. (2005).
Tree allometry and improved estimation of carbon stocks and balance in tropical
forests. Oecologia, 145, 87–99.
Chave, J., Olivier, J., Bongers, F., Châtelet, P., Forget, P. M., van der Meer, P., et al.
(2008). Above-ground biomass and productivity in a rain forest of eastern South
America. Journal of Tropical Ecology, 24, 355–366.
Churkina, G. (2012). Carbon cycle of urban ecosystems. In R. Lal, & B. Augustin (Eds.),
Carbon sequestration in urban ecosystems. Netherlands: Springer.
Clark, D. A., Brown, S., Kicklighter, D. W., Chmabers, J. Q., Thomlinson, J. R., Ni, J., et al.
(2001). Net primary production in tropical forests: An evaluation and synthesis
of existing field data. Ecological Applications, 11, 1154–1171.
Coutts, A. M., Beringer, J., & Tapper, N. J. (2007). Characteristics influencing the variability of urban CO2 fluxes in Melbourne, Australia. Atmospheric Environment,
41, 51–62.
Crawford, B., Grimmond, C. S. B., & Christen, A. (2011). Five years of carbon dioxide
fluxes measurements in a highly vegetated suburban area. Atmospheric Environment, 45, 896–905.
Crawford, B., & Christen, A. (2014). Spatial variability of carbon dioxide in the urban
canopy layer and implications for flux measurements. Atmospheric Environment,
98, 308–322.
Crawford, B., & Christen, A. (2015). Spatial source attribution of measured
urban eddy covariance CO2 fluxes. Theoretical and Applied Climatology, 119,
733–755.
De Kimpe, C. R., & Morel, J. L. (2000). Urban soil management: A growing concern.
Soil Science, 165, 31–40.
Demuzere, M., Orru, K., Heidrich, O., Olazabal, E., Geneletti, D., Orru, H., et al. (2014).
Mitigating and adapting to climate change: Multi-functional and multi-scale
assessment of green urban infrastructure. Journal of Environmental Management,
146, 107–115.
Hanson, P. J., Edwards, N. T., Garten, C. T., & Andrews, J. A. (2000). Separating root
and soil microbial contributions to soil respiration: A review of methods and
observations. Biogeochemistry, 48, 115–146.
Järvi, L., Nordbo, A., Junninen, H., Riikonen, A., Moilanen, J., Nikinmaa, E., et al. (2012).
Seasonal and annual variation of carbon dioxide surface fluxes in Helsinki,
Finland, in 2006–2010. Atmospheric Chemistry and Physics, 12, 8475–8489.
Kordowski, K., & Kuttler, W. (2010). Carbon dioxide fluxes over an urban park area.
Atmospheric Environment, 44, 2722–2730.
Luyssaert, S., Inglima, I., Jungs, A. D., Richardson, A. D., Reichsteins, M., Papale, D.,
et al. (2007). CO2 balance of boreal, temperate, and tropical forests derived from
a global database. Global Change Biology, 13, 2509–2537.
Luyssaert, S., Schulze, E. D., Börner, A., Knohl, A., Hessenmöller, D., Law, B. E., et al.
(2008). Old-growth forests as global carbon sinks. Nature, 455, 213–215.
Liu, H. Z., Feng, J. W., Järvi, L., & Vesala, T. (2012). Four-year (2006–2009) eddy
covariance measurements of CO2 flux over an urban area in Beijing. Atmospheric
Chemistry and Physics, 12(17), 7881–7892.
Mahecha, M. D., Reichstein, M., Carvalhais, N., Lasslop, G., Lange, H., Seneviratne, S.
I., et al. (2010). Global convergence in the temperature sensitivity of respiration
at ecosystem level. Science, 329(5993), 838–840.
Matese, A., Gioli, B., Vaccari, F. P., Zaldei, A., & Miglietta, F. (2009). CO2 emissions of the
city center of Firenze, Italy: Measurement, evaluation and source partitioning.
Journal of Applied Meteorology and Climatology, 48(9), 1940–1947.
McHale, M. R., Burke, I. C., Lefsky, M. A., Peper, P. J., & McPherson, E. G. (2009).
Urban forest biomass estimates: Is it important to use allometric relationships
developed specifically for urban trees? Urban Ecosystems, 12, 95–113.
McPherson, E. G., Xiao, Q., & Aguaron, E. (2013). A new approach to quantify and map
carbon stored, sequestered and emissions avoided by urban forests. Landscape
and Urban Planning, 120, 70–84.
Muller-Landau, H. C., Condit, R. S., Chave, J., Thomas, S. C., Bohlman, S. A., Bunyavejchewin, S., et al. (2006). Testing metabolic ecology theory for allometric scaling
of tree size, growth and mortality in tropical forests. Ecology Letters, 9, 575–588.
National Environmental Agency (NEA). (2010). Singapore’s second national communication under the United Nations Framework Convention on Climate Change.
Singapore Government.
Nemitz, E., Hargreaves, K. J., McDonald, A. G., Dorsey, J. R., & Fowler, D. (2002).
Micrometeorological measurements of the urban heat budget and CO2 emissions on a city scale. Environmental Science & Technology, 36, 3139–3146.
Offerle, B., Grimmond, C. S. B., & Oke, T. R. (2003). Parameterization of net all-wave
radiation for urban areas. Journal of Applied Meteorology, 42, 1157–1173.
Pataki, D. E., Carreiro, M. M., Cherrier, J., Grulke, N. E., Jennings, V., Pincetl, S.,
et al. (2011). Coupling biogeochemical cycles in urban environments: Ecosystem services, green solutions, and misconceptions. Frontiers in Ecology and the
Environment, 9(1), 27–36.
Pawlak, W., Fortuniak, K., & Siedlecki, M. (2011). Carbon dioxide flux in the centre
of Łódź, Poland—Analysis of a 2-year eddy covariance measurement data set.
International Journal of Climatology, 31(2), 232–243.
Peper, P. J., McPherson, E. G., & Mori, S. M. (2001a). Predictive equations for
dimensions and leaf area of coastal southern California street trees. Journal of
Arboriculture, 27(4), 169–180.
Peper, P. J., McPherson, E. G., & Mori, S. M. (2001b). Equations for predicting diameter,
height, crown width, and leaf area of San Joaquin Valley street trees. Journal of
Arboriculture, 27(6), 306–317.
E. Velasco et al. / Landscape and Urban Planning 148 (2016) 99–107
Poorter, H., Niklas, K. J., Reich, P. B., Oleksyn, J., Poot, P., & Mommer, L. (2011). Biomass
allocation to leaves, stems and roots: Meta-analyses of interspecific variation
and environmental control. New Phytologist, 193, 30–50.
Pouyat, R. V., Yesilonis, I. D., & Golubiewski, N. (2009). A comparison of soil organic
carbon stocks between residential turfgrass and native soil. Urban Ecosystem, 12,
45–62.
Pouyat, R. V., Yesilonis, I. D., & Nowak, D. J. (2006). Carbon storage by urban soils in
the United States. Journal of Environmental Quality, 35, 1566–1575.
Quah, A. K., & Roth, M. (2012). Diurnal and weekly variation of anthropogenic
heat emissions in a tropical city, Singapore. Atmospheric Environment, 46,
92–103.
Raciti, S. M., Hutyra, L. R., & Finzi, A. C. (2012). Depleted soil carbon and
nitrogen pools beneath impervious surfaces. Environmental Pollution, 164,
248–251.
Ramamurthy, P., & Pardyjak, E. R. (2011). Toward understanding the behavior of
carbon dioxide and surface energy fluxes in the urbanized semi-arid Salt Lake
Valley, Utah, USA. Atmospheric Environment, 45, 73–84.
Secretaria del Medio Ambiente – Gobierno del Distrito Federal (SMA-GDF). (2013).
Registro de emisiones de gases de efecto invernadero del Distrito Federal: 2012.
Gobierno del Distrito Federal.
Soegaard, H., & Møller-Jensen, L. (2003). Towards a spatial CO2 budget of a metropolitan region based on textural image classification and flux measurements. Remote
Sensing of Environment, 87, 283–294.
Stewart, I. D., & Oke, T. R. (2012). Local climate zones for urban temperature studies.
Bulletin of the American Meteorological Society, 93, 1879–1900.
Stephenson, N. L., Das, A. J., Condit, R., Russo, S. E., Baker, P. J., Beckman, N. G., et al., &
Coomes. (2014). Rate of tree carbon accumulation increases continuously with
tree size. Nature, 507, 90–93.
107
Townsend-Small, A., & Czimczik, C. I. (2010). Carbon sequestration and greenhouse
gas emissions in urban turf. Geophysical Research Letters, 37, L02707.
United States – Environmental Protection Agency (US-EPA). (2008). Trees and vegetation. In Reducing urban heat islands: Compendium of strategies. USA: Climate
Protection Partnership Division—EPA’s Office of Atmospheric Programs.
van Breugel, M., Ransijn, J., Craven, D., Bongers, F., & Hall, J. S. (2011). Estimating
carbon stock in secondary forests: Decisions and uncertainties associated with
allometric biomass models. Forest Ecology and Management, 262, 1648–1657.
Velasco, E., Pressley, S., Grivicke, R., Allwine, E., Coons, T., Foster, W., et al. (2009).
Eddy covariance flux measurements of pollutant gases in urban Mexico City.
Atmospheric Chemistry and Physics, 9(19), 7325–7342.
Velasco, E., & Roth, M. (2010). Cities as net sources of CO2 : Review of atmospheric
CO2 exchange in urban environments measured by eddy covariance technique.
Geography Compass, 4(9), 1238–1259.
Velasco, E., Pressley, S., Allwine, E., Grivicke, R., Molina, L. T., & Lamb, B. (2011).
Energy balance in urban Mexico City: Observation and parameterization during
the MILAGRO/MCMA-2006 field campaign. Theoretical and Applied Climatology,
103, 501–517.
Velasco, E., Roth, M., Tan, S. H., Quak, M., Nabarro, S. D. A., & Norford, L. (2013).
The role of vegetation in the CO2 flux from a tropical urban neighborhood.
Atmospheric Chemistry and Physics, 13, 10185–10202.
Velasco, E., Perrusquia, R., Jiménez, E., Hernández, F., Camacho, P., Rodríguez, S., et al.
(2014). Sources and sinks of carbon dioxide in a neighborhood of Mexico City.
Atmospheric Environment, 97, 226–238.
Ward, H. C., Kotthaus, S., Grimmond, C. S. B., Bjorkegren, A., Wilkinson, M., Morrison,
W. T. J., et al. (2015). Effects of urban density on carbon dioxide exchanges:
Observations of dense urban, suburban and woodland areas of southern England.
Environmental Pollution, 198, 186–200.