ArticlePDF Available

FluxSAP 2010 experimental campaign over an heterogeneous urban zone, part 1: heat and vapour flux assessment

Authors:
Urban Projects
25
ISSUE NO. 40 JUNE 2011 INTERNATIONAL ASSOCIATION FOR URBAN CLIMATE
The experimental campaign FluxSAP 2010: Climatological measurements
over a heterogeneous urban area
In the framework of a research program on the role of
vegetation in sustainable urban development, the FluxSAP
campaign was conducted with the objective of quantifying
the importance of vegetation in the sensible heat and wa-
ter vapour fluxes from a mixed district. The 2010 campaign,
which took place around the permanent hydro/meteoro-
logical survey site of IRSTV, was principally devoted to the
feasibility of measurement over a very heterogeneous area,
by using in parallel five sets of sensors which all allow mea-
surement of fluxes. This creates the possibility, by evaluating
their origin, of separating the contributions of the various
land cover modes.
This report is a translation of an article published in La
Météorologie”, the bulletin of the Socié Météorologique
de France (Mestayer et al., 2011).
Context and Objectives
The study of the urban atmosphere is a relatively recent
topic. While wind measurements around isolated buildings
and wind tunnel simulations for comfort and material siz-
ing purposes were the object of numerous works during
the 20th century, it was only by the end of the 90’s that the
first models were launched allowing a predictive simulation
of the urban energy budget (Masson 2000, Grimmond and
Oke 2002) and the first experimental campaigns allowing
researchers to validate them. Logically, following the semi-
nal series of works performed by Tim Oke and his group in
Vancouver during the 80’s and 90’s (see, e.g., Grimmond and
Oke, 1999), the large cooperative experimental campaigns
were focused on large city centers where we could find both
the conditions most remote to natural soils – i.e., the dense
city with largely paved surfaces and elevated buildings – and
a relative horizontal homogeneity allowing the use of Mo-
nin–Obukhov (1954) similarity theory (MOST) to analyze the
lower atmosphere measurements.
In Europe, the campaigns BUBBLE
1
in Basel (Rotach et al.
2005), CLU-ESCOMPTE
2
in Marseille (Mestayer et al. 2005)
and CAPITOUL
3
in Toulouse (Masson et al. 2008) covered
large urban areas to explore the spatial structure of the ur-
ban boundary layer, but mainly the city center data have
been analyzed to assess the aerodynamic and thermo-ra-
diative inuence of buildings, to quantify the anthropic heat
uxes and to precisely describe the structure of the urban
heat island. Yet, the largest part of the cities is constituted
of mixed residential districts which superimpose impervious
surfaces and buildings on the one hand with gardens, parks
and non-impervious developments on the other hand, in a
patchwork whose spatial scales are usually much less than 1
km. On these heterogeneous urban zones the measurement
analysis encounters additional problems linked to this spa-
tial heterogeneity: sparse or discontinued vegetation pres-
ence, “footprint” evaluation, neighboring surface inuence,
ux divergence, reshaped grounds of ill-known nature, sur-
face water ows, etc.
The key factors of micrometeorology and quantitative hy-
drology are especially interdependent in these urban areas.
Evapotranspiration is a factor appearing in both the water
budget and the energy budget, while heat transfer by con-
duction in the soil strongly depends on the water content.
In urban zones the twinned processes are modied by the
presence of either semi-impervious surfaces (roads) or to-
tally impervious surfaces (roofs), which are totally or partially
connected to the hydrographic network (sewer pipes, gut-
ters, channeled brooks) but are also generators of horizontal
transfers by surface runo and by modern storm water man-
agement (inltration ditches or basins, urban streams).
This strong interdependence brought several groups of
IRSTV (www.irstv.fr) since 2004 to perform limited urban hy-
drometeorology measurement campaigns within the INSU
4
hydrology programs, and further set up long-term observato-
ry instrumentation over a district of Nantes (previously called
SAP, now ONEVU Nantes Observatory of Urban Environ-
ment). More recently IRSTV launched a cooperative program
with some 15 partners
5
aimed at understanding and quanti-
tatively assessing the vegetation impact in present and future
urban development projects (VegDUD, The role of vegeta-
tion in sustainable urban development; an approach linked
to climatology, hydrology, energy control, and environment),
funded by the French National Research Agency (ANR) over
4 years (2010-2013). Within this framework two campaigns
of at-ground and airborne measurements are organized, the
rst one in 2010 and the second one in 2012, in the vicinity of
the permanent observation sites (Figure 1).
1) Basel urban boundary layer experiment. 2) Couche limite urbaine - expérience
sur site pour contraindre les modèles de Pollution atmosphérique de transport
des émissions. 3) Canopy and aerosol particles interactions in Toulouse urban
layer. 4) Institut national des sciences de l’univers du CNRS. 5) LMF - École Cen-
trale de Nantes, GER - IFSTTAR, CAPE - CSTB, LPGN - Univ. Nantes, CESBIO - Tou-
louse, EPHYSE - INRA Bordeaux, LRC - IRSN, LSIIT - Strasbourg, LTHE - Grenoble,
OU - Norman, Oklahoma, USA, SAFIRE - CNRS-CNES-Météo France.
Figure 1. Location of the experimental domain (white disc)
within the Nantes urban area. The buildings are represented
in black, the water surfaces in blue, the main roads in red
and the administrative limits of Nantes Metropole in green.
Urban Projects
26
ISSUE NO. 40 JUNE 2011 INTERNATIONAL ASSOCIATION FOR URBAN CLIMATE
FluxSAP is altogether one of the experimental com-
ponents of the program VegDUD, in connection with the
modeling component, and a project of methodology and
measurement development in urban hydrometeorology
and remote sensing, both in the continuation of the previ-
ous hydrometeorology studies of the IRSTV groups and as a
follow-up of CLU-ESCOMPTE and CAPITOUL meteo-climatol-
ogy programs combining experimental campaigns and nu-
merical modeling. As such, FluxSAP is funded by the INSU.
The experimental set-up includes sensing systems for
temperature and water content in the soils, temperature and
humidity of air at 2-3 m above ground, turbulent uxes on
meteorological masts, scintillometers on elevated building
roofs, surface temperatures by airborne infrared remote sens-
ing, soil and surface identication by airborne hyperspectral
remote sensing, and passive tracer dispersion exercises.
Objectives
The program objectives are twofold, methodological and
quantitative. The methodological objective concerns the
feasibility of measuring the heat uxes in a heterogeneous
urban area: How can we measure them? How can we analyze
them? The quantitative objective is to obtain over one het-
erogeneous urban site some heat and water vapor transfer
reference data allowing us to assess models which take into
account the heterogeneity of urban grounds and the pres-
ence of networks; for this purpose it is necessary to perform
reliable and precise measurements but also to identify the
footprints of these measurements.
The heat and water budgets at the surface of an urban
fragment are usually described by the following twinned
equations:
R
n
+ F = H
S
+ LE + G
P = RET + R
S
+
I
where R
n
is net radiation (energy gain during daytime and
loss at night) and F anthropic heat contribution (produced
by human activity, heating, vehicles, industry), H
S
and LE the
aerodynamic uxes of sensible and latent heat, and G the
conduction heat ux into the ground and materials (build-
ings, vegetation). P is the water contribution of precipita-
tion (rain and condensation), RET the real evapotranspira-
tion, R
S
the surface runo ux and
I
the ground inltration
ux. The evapotranspiration RET is called real by reference to
the potential evapotranspiration PET furnished by the me-
teorological oce, Météo France, based on Penmann-Mon-
teith formulas (Climathèque, 2010). This term represents
the strongest coupling between the two budgets, since
LE=L
v
RET, where L
v
is the specic heat of vaporization. Very
low in the “mineral city” of the city centers and predominant
over the vegetated surfaces, its relative importance is a direct
function of the distribution of land uses and urban manage-
ment, and it is one of the keys of the urban climate since the
available energy which is restituted to the lower atmosphere
is the sum H
S
+ LE – the stronger the evapotranspiration, the
lower the air-warming sensible heat.
The measurement of surface uxes Several methods can
be used to measure surface uxes with presently available
instrumentation. They evaluate the uxes between the sur-
face and the atmosphere over very variable scales, from a
few cm
2
(gradients within the ground) to a few hectares (co-
variance of turbulent variables at the top of a meteorological
mast), or even several km
2
(satellite remote sensing or “bulk”
methods). Many works have demonstrated that they are
more or less equivalent to evaluate the heat ux of a homo-
geneous surface like large crops and ocean surfaces since
the ux is the same here and there, and constant over the
whole height of the atmospheric surface layer or constant
ux layer described by MOST. But can we use these methods
in an agreeable way in urban sites? What is their represen-
tativeness? What is the inuence of ground heterogeneity
upon elevated measurements? Our objective is to imple-
ment 5 ux measurement methods over one domain and to
compare their results.
The water table level, water content and temperature
measurements in the ground allow us to monitor water
and heat transfers through the soil layers. The temperature
proles allow us to evaluate the conduction heat ux with
the gradient harmonic method based on a Fourier analy-
sis of temperature time series at several depths. These are
point measurements and their locations have been chosen
to assess the behavior variability of the district open green
spaces.
The temperature and humidity measurements at the
surface (T
s
, q
s
) and at a height z of a few meters above the
surface (T
z
, q
z
) are analyzed by the method of the mean gra-
dients:
H
s
= ρC
p
C
H
U
z
(T
z
- T
s
); LE = L
v
C
v
U
z
(q
s
- q
z
)
where U
z
is the wind speed at height z, C
H
and C
v
transfer
coecients whose values are a function of surface char-
Figure 2. Measurement setup during the FluxSAP 2010 cam-
paign: the red frame indicates the limits of the geographi-
cal data base, the red dots the locations of T-RH sensors, the
blue symbols the turbulent ux sensors on masts or avail-
able supports, the orange lines scintillometer paths, and the
blue hexagon the Sodar. The sensors in the ground appear
in Figure 3.
Urban Projects
27
ISSUE NO. 40 JUNE 2011 INTERNATIONAL ASSOCIATION FOR URBAN CLIMATE
acteristics. These “bulk” formulas are commonly used for
homogeneous sea surfaces or over large crops but not for
heterogeneous areas with measurements at a low level (2-
3 m), therefore representative of small and near footprints.
The analysis of these measurements at the district scale re-
quires the computation of geomatic interpolations based on
ground cover modes and sensor environment classication.
Remote sensing measurements with airborne thermal in-
frared (TIR) cameras allow us to complete ground measure-
ments to determine the spatial distribution of the surface
temperatures.
Sensible heat and water vapor turbulent ux measure-
ments with fast sensors (at least 10 Hz) of turbulent uctua-
tions of wind speed, temperature, water vapor concentration
(and CO
2
concentration as well), at the top of meteorological
masts, are analyzed with the aerodynamic eddy correlation”
method, with footprints on the order of 10 ha depending on
wind direction and meteorological data.
• The measurements with elevated scintillometers, above
the urban canopy, allow us to evaluate the sensible heat
uxes, integrated over their path lengths of 1–2 km by using
several semi-empirical, supposedly universal surface layer
formulas based on the MOST assumptions and on Kolmogo-
rov-Obukhov (1946-1971) turbulence universality theory.
• Temperature, humidity and wind speed measurements
at several levels of the instrumented mast of the permanent
observation site allow us, in principle, to evaluate the uxes
with the mean vertical prole or gradient method.
Quantitative assessment & identication of footprints
The quantitative objective of the campaign is related to the
general objective of the VegDUD program, to assess the
vegetation contribution to the urban climatology. The rst
purpose of these experimental data are the validation of
urban meteorological and hydrological models, especially
ARPS-Canopy
6
for the atmospheric boundary layer, with
the drag-force model for the canopy “porous” layer (Maché
et al., 2009, 2010) and SM2U for the heat and water surface
transfers (Dupont and Mestayer, 2006; Dupont et al., 2006)
and the model URBS
7
for the water budgets and the urban
catchment hydrology (Berthier et al., 2006; Rodriguez et al.,
2007, 2008). For this purpose, the measurements are repeat-
ed at several points of a domain about 6 km
2
wide and it will
be necessary to establish if they show a sucient coherency
to ensure a high level of condence – because the meteorol-
ogy is the same over the whole domain – and sucient dif-
ferences to demonstrate the inuence of the dierent land
use distributions of their footprints since the dierences
are the signature vegetation contributions.
Used over an urban district, all these measurement meth-
ods require an analysis bearing on urban geographical data-
bases (digital elevation model, land uses and land covers) to
determine either measurement spatial representativeness,
transfer coecient values, surface TIR emissivities, ground
slopes, building heights and volumes, or roughness param-
eters used in footprint models. Due to the extreme hetero-
geneity of the materials covering the urban surfaces, the
commonly available geographical databases do not contain
Figure 3. Hydrologic instrumentation of the Pin Sec catch-
ment. The temperature proles in the ground were set at the
same places than the piezometers. G and D indicate the per-
manent meteorological masts at the Goss site (30 m) and on
the Dunant building roof. The black line indicates the Pin Sec
watershed. The pink/red dots give the measured soil drying
at 35 cm during the month of May 2010 (in moisture %).
enough information to document their radiative, thermody-
namic and hydrologic characteristics. Hyperspectral remote
sensing, at high spatial and spectral resolutions, allows us
to complement them, based on comparisons of their spec-
tral signatures to spectral reectivity banks generated from
measurements at ground and at the laboratory. The analy-
sis of high spatial resolution satellite data (SPOT, Quickbird)
with spatial segmentation methods also allow us to monitor
the land cover changes.
Due to the strong heterogeneity of urban grounds the de-
termination of measurement footprints is crucial and com-
plex. If, for a radiation sensor at the top of a mast the foot-
print is easily determined by a geometry calculation, and
it does not change, this is not the case for a sensor system
measuring the turbulent ux of a scalar (heat, water vapor,
gas or aerosol concentration): its footprint depends not only
on the sensor height but also on the wind speed and direc-
tion, the ground roughness, and the thermal stratication;
in low winds it is located nearly at the mast foot but much
farther in strong wind conditions, and also farther when the
atmosphere is stratied. The experimental determination
bears on footprint models based on backward trajectory
or inverse plume computations (Leclerc and Thurtell, 1990;
Schmid, 1997, 2002) which in turn bear on the MOST clas-
sical assumptions, especially horizontal homogeneity. But
these assumptions are generally not veried in urban areas.
Passive tracer dispersion exercises allow us to test the valid-
ity of these models. These measurements will be analyzed
with a Lagrangian backward trajectory model implemented
within the ARPS atmospheric model including the drag-force
model for the porous canopy.
The feasibility questions Among the practical questions
raised by the experimental assessment of urban meteo/cli-
6) ARPS: Advanced regional prediction system, University of Oklahoma.
7) Urban runo branching structure.
Urban Projects
28
ISSUE NO. 40 JUNE 2011 INTERNATIONAL ASSOCIATION FOR URBAN CLIMATE
matology we expect to answer with the FluxSAP campaigns
the following queries:
Are the ux measurements operated on several masts
spread over a district suciently precise to altogether show
a sucient coherency to ensure a high level of condence
and include signicant dierences that can be attributed to
their footprint dierences?
• Due to the technical and administrative problems raised
by the setting of sensors within an urban fabric we may be
led to install provisional meteorological masts on platforms
which are not optimal as regards the theoretical recommen-
dations (Oke, 2004) but which are well secured as, e.g., build-
ing roofs: How can we determine the quality of the resulting
measurements? How can we take advantage of these “sub-
optimal” measurements?
• Is it possible to continuously monitor the uxes in an ur-
ban environment without meteorological masts? Is it possi-
ble to interpolate point measurements at ground to produce
maps of the uxes?
• What is the measurement reliability of scintillometers set
over building roofs? Can they be used to monitor a district?
Can they be integrated in a long-term observation system?
Do the measurements with slanted infrared cameras
allow us to separate the sensible heat ux contributions of
building facades and roofs (Hénon, 2008)?
The May 2010 set-up
The measurements have been performed around the
small urban watershed of Pin Sec, a district located in the
“second ring” between the XIXth century boulevards and the
rim speedway, with heterogeneous land uses including small
areas of collective and individual housing, athletic facilities,
schools, supermarkets and small industrial plants. The Pin
Sec catchment has been equipped with hydrology and me-
teorology sensors for several years in the framework of the
permanent observatory SAP/ONEVU of the IRSTV. Launch-
ing cooperative experimental campaigns on a permanent
observation site creates an especially interesting synergy,
allowing us on the one hand to extend to other seasons the
results obtained during the limited period of time of the
Figure 4. View of meteorological masts and turbulence sen-
sor supports.
campaign, and on the other hand to multiply the points of
measurement (and the points of view) which are limited by
necessity in the permanent setup.
The campaign spread over the 4 weeks of May 2010, with
a few technical operations by the end of April and beginning
of June.
8
In spite of a generally favorable reception by the in-
habitants and local authorities the selection of relevant and
secured measurement sites has been rather dicult, most
often due to the administrative circuits delivering the autho-
rizations. Six scintillometers have been set on 5 high build-
ing roofs, with accesses sometimes dicult but rather well
secured. But, in addition to the SAP permanent instrumenta-
tion we have been able to equip only 6 additional sites for
the turbulent ux measurements. We did not observe any
malevolence or deterioration; only one turbulent ux mea-
surement system did not work well due to decient cable
connectors and one ground temperature sensor was deteri-
orated during a rain storm. The campaign therefore attained
a success which passes beyond its primary objective of a
feasibility study. The meteorology was rather favorable since
over the measurement period we observed a range of situ-
ations from overcast with showers to very strong insolation.
Thanks to the good previsions of the met oce station of
Nantes airport the airborne measurement ights have been
successfully planned for the 3 most sunny days.
The setup is shown in Figure 2 with a zoom on the Pin
Sec central area in Figure 3, while Figures 4 and 5 illustrate
the sensor positioning. Eight masts or available supports
(one crane and one wind mill) have been equipped with
sonic anemo-thermometers at heights of 10 to 26 m above
ground level (agl), among which are two at 2 levels (G and
E) and six with H
2
O/CO
2
sensors (LiCor 7500 or 7000). Five
large aperture scintillometers have been set to form a tri-
angle and a cross, among which 2 were twinned in parallel
to test a method to obtain the friction velocity (Figure 5). A
small aperture scintillometer was set between two neighbor
buildings at the central site (D), with a 75 m long path. The
network of ground sensors included 10 piezometers com-
posed of a pressure sensor at the bottom of a hole, 8 water
Figure 5. Setup of scintillometers on high building roofs.
8) http://ftpa.ec-nantes.fr/FluxSAP/Rapport_de_campagne
Urban Projects
29
ISSUE NO. 40 JUNE 2011 INTERNATIONAL ASSOCIATION FOR URBAN CLIMATE
content sensors TDR CS625 connected to Campbell CR200
recorders to which 8 temperature prole systems have been
joined: 4 Taupe recorders with 3 sensors at the depths of 0,
-5 cm, -35 cm and 4 Taupe recorders with 4 sensors at the
depths of 0, -5 cm, -35 cm, -50 cm or -1m (Figure 3). The net-
work of air temperature and humidity (T-RH) sensors at 2-3
m agl included 10 systems composed of a sheltered sensor
and an autonomous Hydrolog-D recorder in addition to the
permanent SAP network which includes 4 similar sensors
over the campaign domain.
From May 21 to 23, 13 infrared remote sensing ights (150
legs) have been performed with two IRT cameras on board
the Piper Aztec 21 of the research and environment French
instrumented airplane service (SAFIRE) ying at 600 m agl;
one camera aimed at Nadir and the second one was slanted
50° backward, which allows us both to map the surface tem-
peratures and to evaluate the directional brightness tem-
perature anisotropy (Lagouarde et al., 2000). These measure-
ments were coordinated with measurements with a third IRT
camera from the top of Brittany Tower overlooking the city
center (70 images) and with 140 reference measurements at
the ground with 2 radio-thermometers (grounds, facades,
Erdre and Loire river surfaces). The ights were alternated to
document each hour of the day and the ight lines crossed
each other over Pin Sec on the one hand, and the city center
area observed from the Brittany Tower on the other hand.
The hyperspectral ight (20 parallel ight lines at a height of
1600 m agl) took place on May 21 between 10 and 12 UTM
with Hyspecs cameras VNIR (400–1000 nm, 160 bands of 4
nm, 17° fov, 0.6 m spatial resolution at ground) and SWIR
(1000–2500 nm, 256 bands of 6 nm, 12° fov, 1.2 m resolu-
tion). Simultaneously 95 reference measurements were per-
formed at the ground with a portable spectrometer.
The local meteorology was documented by the perma-
nent observation site data (Figure 6) and by those from Mé-
téo France station at the Nantes Atlantique airport. A Sodar
was run on the CSTB site on the other side of the Erdre river
(Figure 2) but it was operated during only 2 days due to the
noise nuisance. The wind rose is rather representative of the
dominant wind regimes over the year in Nantes. Note that
the dispersion exercises (weeks 20 and 21) were performed
with dierent stable wind regimes, respectively from N, N NE
and SW. For the remote sensing ights during the Whitsun
week-end the wind remained stable from the NE sector with
a strong insolation. May 10 and 29 stand out with a strong
cloud coverage. The temperatures varied largely, sometimes
rapidly (by 2 to 30°C). The rain showers were numerous and
regularly spread, but the precipitation quantity was about
the half of those of the same period in the preceding years
(on average 41 mm over Pin Sec).
Preliminary Results
The air temperature time series of 3 m agl T-RH network
sensors during the campaign appear in Figure 7. We have
selected a two week period when the temperature rose pro-
gressively. We also indicate the measurements obtained at
Figure 6. Meteorology during the campaign: temperatures, solar and infrared incoming radiations, wind direction, water
vapor density and rain.
Urban Projects
30
ISSUE NO. 40 JUNE 2011 INTERNATIONAL ASSOCIATION FOR URBAN CLIMATE
the permanent station on the roof of
building D (15 m agl) and at the foot
of the permanent mast (G). The en-
semble of sensors correctly represent
the general evolution of temperature,
but one can note dierences in the
diurnal cycles which, although not
systematic, reect some tendencies.
During the nights the dierences are
smaller, within the measurement un-
certainty range, except for one site
with a very vegetated nearby envi-
ronment (T-RH 11) which causes a
2°C cooler temperature at sunset.
During daytime the dispersion is
larger, with extreme dierences between sites of up to 5°C
during the most sunny days. A ner documentation of the
near environment of each sites is being performed to better
characterize them and to relate the dierences to the foot-
print characteristics.
As for the sensible heat and vapor uxes, the partial re-
sults which are presented here have been obtained from
15-minute samples obtained at 3 sites among the 8: G at 26
and 21 m, D at 3 m above roof level (18 m agl), and M at 10
m. Figure 8 shows:
- a good coherency between the uxes at 26 m and 21 m
on mast G ;
- a good coherency also between the heat uxes at masts
G and M ;
- sensible heat and latent heat uxes of the same order of
magnitude during this month of May for this mixed zone;
- neatly lower ux measurements at the building D roof,
especially for the water vapor ux. Is this a systematic bias
due to the low position above a big building ? The analyses
will need to answer this question which is important for the
urban site instrumentation.
The ground temperatures have been measured at 3 or 4
depths. To illustrate the spatial variability of temperatures
and storage uxes in the upper layers, Figure 9 displays the
proles obtained during two typical days (see Figure 6): May
9 after a relatively cool and humid period and May 23 at the
end of the hot period. The site 7 (close to sensors WAF) is
most of the time in shadow (in May it “seesthe sun from
9h30 to 11h only), while the site 4 (WPS) is open and the
temperature variation amplitude is larger. The phase shift
between the surface and the depth 5 cm is about 1 hour for
the two sites. At the surface the temperature is directly de-
pendent on the local insolation condition; thus the storage
ux, associated to the surface gradient, is negative during
the night and may stay so during a large part of the day at
a masked site (site 7) while it is largely positive all day long
at an open, sunlit site (site 4 on May 23). Besides, Figure 9
shows that the temperature variations at 1 m in the ground
are small at the day and week scales, which validates the
choice of this maximal depth for future studies.
The ground hydrological sensor network worked well from
May 7. The precipitations have been relatively low and the
observation period belongs to the decline phase of ground
saturation levels, after a high water-table period which end-
ed by the end of March in this area. The measurements in the
ground indicate that the saturation level decreased by 22 cm
on average during the period and the ground water content
decreased by 2.4 % on average. A noticeable spatial variabil-
ity has been observed between the various measurement
points, located on public and semi-public green spaces as
shown by Figure 3, which indicates that the ground drying
during the month of May is not homogeneous on these sites
and varies from 0 to 5% of the volumetric soil moisture. Over
this period two signicant rain events have been observed:
8.5 mm in 4 hours on May 10, and 26.5 mm in 4 hours on
June 6. The smaller rain event did not induce noticeable
variations of the soil moisture at this depth: on May 10 only
one sensor records a weak variation (Figure 10). On the other
hand, the event of June 6 generated a ground moistening
visible on half of the sensors, with very dierent amplitudes:
some sensors indicate moisture jumps of 10 to 20 % over a
few hours while some others hardly vary. This observation
conrms the high variability of the sites with respect to wa-
ter inltration. The sub-surface ows are indeed inuenced
by the presence of buried networks and that of roots, and
the morphology of the instrumented green spaces is vari-
able, at the surface as well as in the subsoil. Last, the most
intense drying period has been observed during the sunny
period from May 20 to 27; it is also over this period that soil
moisture daily oscillations are observed; they are related to
the time variation of ground drying by evapotranspiration
during the diurnal cycle, and therefore to the daily prole of
latent heat ux. The data from these TDR sensors, which are
sensitive to temperature, have been corrected thanks to the
temperature measurements at the same depth.
First lessons and outlook for 2012
Several lessons can already been drawn about the feasi-
bility:
- It is rather dicult to nd suciently open spaces to set
a meteorological mast satisfying the criteria proposed by
Oke (2004), and available securely and without nuisance for
Figure 7. Air temperatures from the T-RH network sensors.
Urban Projects
31
ISSUE NO. 40 JUNE 2011 INTERNATIONAL ASSOCIATION FOR URBAN CLIMATE
the users;
- For installing “heavy” set-
ups (masts, scintillometers…)
the authorizations are easier to
obtain from private landlords
or lessors than from public ser-
vices, except those with whom
a cooperation has been previ-
ously established;
- Inversely, the provisional
installation of autonomous,
“light” sensors (T-RH, ground)
are easily obtained;
- We observed no deteriora-
tion, although some sensors
were hardly protected and/or
at publicly open places;
- The building roofs provide
well secured platforms but they
are rarely of easy access;
- Ground measurements at
apparently similar green spaces
may be inuenced by the lo-
cal conguration (presence of
trees, drainage by buried net-
works…) which may generate a
variability of hydrologic and thermal functioning.
The rst results are encouraging and especially show a
rather strong coherency between the various measurement
sites. For the 2012 campaign, we plan to invest eorts on
the following points:
- to better document the experimental domain ;
- to better quantify the role and the functioning of veg-
etation, in the traditional arrangements as well as in the
ecologically innovative management zones.
With parallel developments of models, we think that this
requires us to rene the understanding of the evapotranspi-
ration ux, notably:
- to better document the connections of impervious sur-
faces to the rainwater network to know better the contribu-
tion of the runo from these surfaces to the water ow at
the network outlet;
- to evaluate the water storage in the ground by the mea-
surement of at least one vertical prole of soil moisture and
suction;
- to dierentiate more clearly the neighboring sites, with
strongly mineral footprints on the one hand, strongly veg-
etal ones on the other hand;
- to follow the uxes in the ground, not only of non-cov-
ered spaces but also under the pavement of roads and park-
ing lots;
- to use the scintillometers on two long paths to estimate
the heat ux of the bulk of the district, as well as on shorter,
eventually very short, paths to evaluate the uxes of homo-
geneous and well-dened sub-districts ;
- to measure the turbulent uxes from footprints corre-
sponding to these sub-districts, which implies adapting the
measurement point height to that of vegetation and build-
ings and to have at our disposal more secured supports and
more water vapor sensors.
It also seems interesting:
- to extend the instrumented zone to the neighboring
district of Bottière-Chenaie, recently developed as an eco-
district by the city of Nantes;
- to instrument a neighborhood building with tempera-
ture sensors to monitor its energy budget;
- to better document the wind, temperature and humidity
proles in the lowest atmosphere, e.g. with a 0-200 m pro-
ler based on a small tethered balloon;
- to document the dierences between concentration
and ux footprints (Schmid, 1997) by developing a passive
tracer ux measurement system.
Last, we hope to implement one or two water vapor pro-
totype scintillometers.
Acknowledgements
We wish to thank all our correspondents in the services
and institutions who helped us to set up the FluxSAP 2010
campaign: Direction of sewerage and Aubinière pole of
Nantes Metropole, Service of green spaces of the city of
Nantes (SEVE), Nantes Habitat, La Nantaise d’Habitation, the
Cabinet Coudray Lorraine, the companies Goss, Defontaine
and Societe Generale Securities Services.
Figure 8. Sensible and latent heat uxes measured by the eddy covariance method at
three sites.
Urban Projects
32
ISSUE NO. 40 JUNE 2011 INTERNATIONAL ASSOCIATION FOR URBAN CLIMATE
References
Berthier, E., S. Dupont, P. G.
Mestayer et H. Andrieu, 2006:
Comparison of two evapo-
transpiration schemes on a
sub-urban site. J. Hydrology,
328, 635–646.
Climathèque, 2010: Fiche
méthode PET quotidienne
en points de grille, version
1.1, Météo France, Direction
de la climatologie (http://cli-
matheque.meteo.fr/)
Dupont, S. et P. G. Mestayer,
2006: Parameterisation of the
Urban Energy Budget with the
Submesoscale Soil Model. J.
Applied Meteor. and Clim., 45,
1744–1765.
Dupont, S., P. G. Mestayer,
E. Guilloteau, E. Berthier et H.
Andrieu, 2006: Parameterisa-
tion of the Urban Water Bud-
get with the Submesoscale
Soil model. J. Applied Meteor.
and Clim., 45, 624–648.
Grimmond, S., et T. Oke, 2002: Turbulent heat uxes in
urban areas: observations and a Local-scale Urban Meteo-
rological Parameterization Scheme (LUMPS). J. Applied Me-
teor., 41, pp. 792–810.
Grimmond, S., et T. Oke, 1999: Heat storage in urban ar-
eas: Observations and evaluation of a simple model. J. Appl.
Meteor., 38, 922–940.
Hénon A., 2008: Températures mesurées, modélisées, et
observées par télédétection infrarouge dans la canopée ur-
baine: modélisation aéro-thermo-radiative des ux de chal-
eur urbains, Thèse de Doctorat, École Centrale de Nantes,
251 pp.
Lagouarde, J.-P., A. Hénon, B. Kurz, P. Moreau, M. Irvine, J.
A. B. Voogt et P. G. Mestayer, 2010: Modelling daytime ther-
mal infrared directional anisotropy over Toulouse city cen-
tre. Remote Sens. Environ. 114, 87-105
Leclerc, M. Y., et G. W. Thurtell, 1990: Footprint predictions
of scalar uxes using a Markovian analysis. Bound.-Layer Me-
teor., 52, 247–258.
Maché, M., I., Calmet et J.-F. Sini, 2009: A high-resolution
large-eddy simulation of turbulent ow within and above
urban-like canopies. 9th annual meeting of European Me-
teorological Society (9th EMS/9th ECAM), 28 Sept. – 2 Oct.
2009, Toulouse, France, Book of abstracts EMS2009-196.
Maché M., I. Calmet et J.-F. Sini, 2010: Analysis of the dy-
namical interactions between atmosphere and urban cano-
pies of dierent densities using a drag force approach. 13th
International Conference on Harmonisation within Atmospher-
ic Dispersion Modelling for Regulatory Purposes (HARMO13),
June 1-4 2010, Paris, France. Proc. HARMO13, 668–672.
Masson V., L. Gomes, G. Pigeon, C. Liousse, V. Pont, J.-P. La-
gouarde, J. Voogt, J. Salmond, T. R. Oke, J. Hidalgo, D. Legain,
O. Garrouste, C. Lac, O. Connan, X. Briottet, S. Lachérade et P.
Tulet, 2008: The Canopy and Aerosol Particles Interactions in
TOulouse Urban Layer (CAPITOUL) experiment. Meteor. and
Atmos. Physics 102, 135–157.
Masson, V., 2000: A Physically-Based Scheme for the Ur-
ban Energy Budget in Atmospheric Models. Bound.-Layer
Meteorol., 94, 357–397.
Mestayer, P. G., P. Durand, P. Augustin, S. Bastin, J.-M. Bon-
nefond, B. Bénech, B. Campistron, A. Coppalle, H. Delbarre, B.
Dousset, P. Drobinski, A. Druilhet, E. Fréjafon, S. Grimmond,
D. Groleau, M. Irvine, C. Kergomard, S. Kermadi, J.-P. Lag-
ouarde, A. Lemonsu, F. Lohou, N. Long, V. Masson, C. Mop-
pert, J. Noilhan, B. Oerle, T. Oke, G. Pigeon, V. Puygrenier, S.
Roberts, J.-M. Rosant, F. Saïd, J. Salmond, M. Talbaut, J. Voogt,
2005: The Urban Boundary Layer Field Campaign in Marseille
(UBL/CLU-ESCOMPTE): Set-up and First Results. Bound.-Layer
Meteorol., 114, 315–365.
Mestayer, P.G., J.-M. Rosant, F. Rodriguez, J.-M. Rouault,
2011: La campagne expérimentale FluxSAP 2010; mesures
de climatologie en zone urbaine hétérogène, La Météorolo-
gie, 73, 33-43
Monin, A. S. and A. M. Obukhov, 1954: Osnovnye zako-
nomernosti turbulentnogo peremeshivanija v prizemnom
sloe atmosfery (Basic Laws of Turbulent Mixing in the At-
mosphere Near the Ground). Trudy geoz. inst. AN SSSR, 24
(151), 163–187.
Obukhov, A. M., 1946: Turbulentnost’ v temperaturnoj
neodnorodnoj atmosfere. Trudy Inst. Theor. Geoz. AN SSSR,
Figure 9. Temperature proles in the ground at two sites.
Urban Projects
33
ISSUE NO. 40 JUNE 2011 INTERNATIONAL ASSOCIATION FOR URBAN CLIMATE
P. G. Mestayer
1,2
, J. -M. Rosant
1,2
, F. Rodriguez
1,3
and J. -M. Rouaud
1,3
1) Institut de recherche en sciences et techniques de la ville, FR CNRS 2488, Nantes, France
2) Laboratoire de Mécanique des Fluides, UMR CNRS 6598, École centrale de Nantes, Nantes, France
3) Département géotechnique, eau et risques, Institut français des sciences et technologies des transports, de
l’aménagement et des réseaux, Bouguenais, Loire-Atlantique, France
patrice.mestayer@ec-nantes.frjean-michel.rosant@ec-nantes.fr,
fabrice.rodriguez@ifsttar.frjean-marc.rouaud@ifsttar.fr
1, 95–115.
Obukhov, A. M., 1971: Turbulence in an Atmosphere with
a Non-uniform Temperature, Bound.-Layer Meteorol., 2, 7–29.
Oke, T., 2004: Initial guidance to obtain representative me-
teorological observations at urban sites. IOM Rep. 81 WMO/
TD 1250, World Meteorol. Organ., Geneva.
Rodriguez, F., F. Morena, H. Andrieu et G. Raimbault, 2007:
Introduction of innovative stormwater techniques within
a distributed hydrological model and the inuence on the
urban catchment behaviour. Water Practice and Tech. 2,
doi10.2166/wpt.2007.0048.
Rodriguez, F., H. Andrieu et F. Morena, 2008: A distributed
hydrological model for urbanized areas - Model development
and applications to case studies. J. Hydrology, 351, 268–287.
Rotach, M. W., R. Vogt, C. Bernhofer, E. Batchvarova, A.
Christen, A. Clappier, B. Feddersen, S.-E. Gryning, G. Martucci,
H. Mayer, V. Mitev, T. R. Oke, E. Parlow, H. Richner, M. Roth, Y.
A. Roulet, D. Rueux, J. Salmond, M. Schatzmann, J. Voogt,
2005: BUBBLE - an Urban Boundary Layer Meteorology Proj-
ect. Theor. Appl. Climatol. 81, 231–261.
Ruban, V., F. Rodriguez, K. Lamprea-Maldonado, I. Bagga,
J.-M. Rosant, T. Piquet, M.-L. Mosini, L. Lebouc, P. Pichon, L.
Letellier, J.-M. Rouaud, L. Martinet, G. Dormal, V. Panhalleux,
D. Demare, P. Mestayer et H. Andrieu, 2010: Le secteur atelier
pluridisciplinaire (SAP), un observatoire de l’environnement
urbain - Présentation des premiers résultats du suivi hy-
drologique et microclimatique du bassin du Pin Sec (Nantes).
Bull. Lab. Ponts et Chaussées, 277, 5-18, 2010.
Schmid, H.P., 1997: Experimental design for ux measure-
ments: matching scales of observations and uxes. Agric.
and Forest Meteor., 87, 179–200.
Schmid, H.P., 2002: Footprint modeling for vegetation at-
mosphere exchange studies: a review and perspective. Agric.
and Forest Meteor., 113, 159–184.
Figure 10. Variations of ground water content at 35 cm during the month of May 2010.
... The catchment has been part of the French observatory ONEVU (Observatoire Nantais de l'EnVironnement Urbain) since 2006, which aims to monitor water and pollutant fluxes and soil-atmosphere exchanges in Nantes urban environment. The site has been instrumented continuously for a decade and has therefore a complete database spanning over a long time period (Mestayer et al., 2011). The following measurements are available (Fig. 2): rainfall, discharges in the storm water and wastewater sewer networks, soil water content, ground water level, turbulent fluxes (latent and sensible heat fluxes), micro-meteorological data (wind speed and direction, air temperature and humidity, atmospheric pressure, incoming solar radiation). ...
Article
Full-text available
Nature-based solutions (NBS) are increasingly promoted to mitigate urbanization effects, such as the urban heat island. The release of latent heat requires the availability of water in the urban soils. Models able to represent both detailed water and energy budgets are needed for a reliable evaluation of NBSs performances. The TEB-Hydro model is a recent hydro-microclimate model that extends the physics of the urban microclimate model TEB-Veg to water processes in urban subsoil in order to represent more realistically coupled water and energy budgets. Hence, the aim of this paper is to evaluate the TEB-Hydro model regarding how the water processes affect the energy balance. The model is applied to an urban French catchment for which both hydrological and microclimate data are continuously collected. The model shows general good performances in both simulating latent and sensible heat fluxes. Nevertheless, soil water contents are slightly underestimated during wet periods and overestimated during dry periods. Compared to the previous version of the model (TEB-Veg) with a simplified water budget, TEB-Hydro tends to more overestimate latent heat fluxes than TEB-Veg during dry periods. During wet periods, however, TEB-Hydro simulates better sensible heat fluxes and latent heat fluxes.
Article
Full-text available
The paper describes the observational and modeling efforts performed under the Bolzano Tracer Experiment (BTEX). BTEX focused on the basin surrounding the city of Bolzano, at the junction of three tributary valleys on the southern side of the Alps, to characterize the ground-level impact of pollutants emitted by a waste incinerator close to the city, and atmospheric factors controlling dispersion processes in the whole basin, under different winter weather situations. As part of the experiment, two controlled releases of a passive gas tracer (sulfur hexafluoride, SF 6 ) were performed through the stack of the incinerator on 14 February 2017 at two different times, starting respectively at 0700 and 1245 LST, representative of distinct phases of the daily cycle. Samples of ambient air were collected at target sites, and later analyzed using a mass spectrometer, allowing a detectability limit down to 30 ppt. Meanwhile, meteorological conditions were continuously monitored by means of a high-resolution, nonconventional network of ground-based instruments, including 15 weather stations, one temperature profiler, one sodar, and one Doppler wind lidar. Data from the above measurements represent one of the rare examples of integrated datasets available to the community for the characterization of dispersion processes in a typical mountainous environment. In particular, they offered a reference benchmark for testing and calibrating a series of combined numerical modeling suites for weather prediction and pollutant dispersion simulation in such a complex terrain, as shown in the paper.
Conference Paper
Full-text available
FluSAP 2010, a part of a large federative research program Vegdud (2010–2013) funded by the French National Research Agency, is an experimental campaign aiming to improve our knowledge on the dispersion of a plume in heterogeneous urban zones. For this, the French Institute for Radiological Protection and Nuclear Safety (IRSN) conducted an experimental campaign in the city of Nantes (France) to study atmospheric dispersion. A gas tracer SF6 was released in the atmosphere and measured at different altitudes (from 1 to 100 m) by using a mast and a small tethered balloon. This allows determining the plume vertical dispersion in an urban area as a function of atmospheric turbulence. A sampling system was also used at ground (1 m height) in order to evaluate the plume horizontal dispersion. All these systems were placed at a distance from the emission point between 20 and 1150 m: 30 SF6 emissions were performed between May 18 and May 27, 2010. High compatibility was found between experimental horizontal and vertical widths of the plume and Briggs urban model for the stability class B, C and D according to Pasquill classification. However, this compatibility was only confirmed for small distances from the source (till 370 m). For higher distances from the source, it is hard to draw significant conclusions. Keywords: atmospheric dispersion, experimental campaign, urban zone, Briggs urban model.
Article
An atmospheric tracer dispersion study was conducted in a heterogeneous district of the city of Nantes, France, in May 2010 and June 2012 during the measurement campaigns FluxSAP. Vertical (0-100 m) and horizontal (≈ 300 m) profiles were measured at distances of 20 to 1200 m from the tracer gas release location. The vertical profiles show vertical mixing occurring progressively with distance: the 0-100 m air column is homogenous beyond 600 m, whereas at shorter distances a maximum of concentration is observed, either at ground level for distances less than 200 m, or at altitude for intermediate distances. Our measurements are compared with the Briggs-urban model which may be used in emergency situations: the model gives coherent results with our measurements in terms of vertical and horizontal dispersion parameters but we observe concentration differences up to an order of magnitude between model and measurements. Based on our results we propose new relationships for the horizontal and vertical dispersion parameters for distances from 0 to 1 km. For emission-measurement distances above 1 km, differences appear between the trends deduced from our measurements and those of the Briggs-urban model: it will be necessary to acquire data at greater distances, 2000 m and more, to verify and validate these trends.
Article
Full-text available
Un suivi à long terme d'un petit bassin versant séparatif urbain a démarré en janvier 2006 à Nantes (France) dans le cadre d'un réseau fédératif de recherche regroupant plusieurs laboratoires (IRSTV). L'objectif de cette étude est d'établir un bilan des flux d'eau, de polluants et d'énergie de ce bassin versant sur plusieurs années pour contribuer à l'évaluation de modèles hydrométéorologiques intégrés en milieu urbain, intégrant les compartiments sol et atmosphère. Les différents dispositifs expérimentaux mis en place sont présentés, ils incluent un suivi en continu, du débit et de la qualité à l'exutoire des deux réseaux d'assainissement, un suivi détaillé de paramètres météorologiques, ainsi qu'un suivi de la nappe. Les premiers résultats montrent que l'établissement du bilan hydrologique du bassin versant du Pin Sec nécessite de faire un certain nombre d'hypothèses qui génèrent des incertitudes non négligeables. Des expérimentations futures permettront d'affiner ce bilan. Concernant le suivi de la qualité, il apparait que les principaux polluants véhiculés par les eaux de ruissellement sont les matières en suspension, le zinc, le cuivre, le plomb et le glyphosate ; un traitement des eaux avant rejet au milieu naturel pourrait s'avérer nécessaire. L'analyse des flux turbulents intervenant dans les bilans énergétique et hydrique reste délicate à cause de l'hétérogénéité du site ; ces flux résultent de surfaces sources, variables en localisation et en taille selon la hauteur du capteur et les conditions météorologiques. Une documentation détaillée de l'occupation des sols est nécessaire pour bien évaluer la contribution de ces sources locales.
Thesis
Full-text available
In the case of an urban fragment the presence of horizontal, vertical, oblique, and very heterogeneous surface elements yields complex definitions of the surface and its associated temperatures. The objective is to establish parametric relationships between the sensible heat and infrared fluxes, which are both related to the surface temperature. Aero-thermo-radiative simulations of the urban canopy were carried out using the SOLENE model, by applying it to configurations of urban fragments of downtown dense areas documented during two series of measurements in two French cities: Marseilles and Toulouse. The detailed analysis of the model variables then makes it possible to understand the thermo-radiative interaction processes of the studied fragment. Moreover, the ability to change the parameter values around those of the real scene allows sensitivity studies, bringing information on both the model and the processes. In the case of Marseilles, the general thermo-radiative behavior of the various surface types during a summer diurnal cycle is studied first, in terms of temperatures and local and integrated fluxes. In the case of Toulouse, the results of the model are validated in a more precise way, the parameterization is improved by developing and optimizing a model of wind inside the canopy and a model of convection coefficient, and then the new tools are qualified. Last, the improvements to the evaluation of the sensible heat and upward infrared fluxes made it possible to highlight their correlation and to propose a method allowing to calculate the sensible heat flux from the measured upward infrared flux.
Article
Full-text available
This paper presents the hydrological component of the Submesoscale Soil Model, urbanized version (SM2-U). This model is an extension of the rural Interactions between Soil, Biosphere, and Atmosphere (ISBA) soil model to urban surfaces. It considers in detail both rural and urban surfaces. Its purpose is to compute the sensible heat and humidity fluxes at the canopy atmosphere interface for the computational domain lower boundary condition of atmospheric mesoscale models in order to simulate the urban boundary layer in any weather conditions. Because it computes separately the surface temperature of each land use cover mode while the original model computes a unique temperature for the soil and vegetation system, the new version is first validated for rural grounds by comparison with experimental data from the Hydrological Atmospheric Pilot Experiment-Modélisation du Bilan Hydrique (HAPEX-MOBILHY) and the European Field Experiment in a Desertification Threatened Area (EFEDA). The SM2-U water budget is then evaluated on the experimental data obtained at a suburban site in the Nantes urban area (Rezé, France), both on an annual scale and for two stormy events. SM2-U evaluates correctly the water flow measured in the drainage network (DN) at the annual scale and for the summer storm. As for the winter storm, when the soil is saturated, the simulation shows that water infiltration from the soil to the DN must be taken care of to evaluate correctly the DN flow. Yet, the addition of this soil water infiltration to the DN does not make any difference in the simulated surface fluxes that are the model outputs for simulating the urban boundary layer. Urban hydrological parameters are shown to largely influence the available water on artificial surfaces for evaporation and to influence less the evapotranspiration from natural surfaces. The influence of the water budget and surface structure on the suburban site local climatology is demonstrated.
Article
Full-text available
Two different urban evapotranspiration (ET) schemes are compared with the dataset of an instrumented sub-urban site, during 5 years at the annual, monthly and hourly time scales. The first scheme comes from atmospheric sciences and details the exchanges with the atmosphere; it is used in the soil model SM2-U. The second one comes from hydrological studies and details the surface and soil water content; it is used in the hydrological model UHE. For both models, the annual ET from the site comes mainly from natural surfaces, which cover 55% of the surface area. At the hourly scale, the ET predominance from natural surfaces is not always verified, especially just after rainfall events. At the annual scale, both models indicate also that the soil water status is not a limiting factor for the ET above this site. However, a large amplitude difference appears between ET simulated by the two models: simulated with UHE the annual ET is 37% higher than with SM2-U. This is explained by uncertainties in some parameter values and by differences in some parameterizations of common variables. These results, illustrate the strong remaining uncertainties in ET estimation above urban areas and the need of an enhanced research effort to measure and model the urban ET factors.
Article
This study presents the implementation of innovative stormwater techniques into a distributed hydrological model. This model, is able to represent hydrological processes usually neglected in urban modelling, such as evapotranspiration, infiltration in roads, or direct infiltration of soil water in sewers. The introduction of innovative stormwater techniques such as flat roofs, permeable reservoir pavements or tree plantation can be easily performed in this very modular model. A small urban catchment in Nantes (France) is taken as case study to analyse the influence of these innovative techniques on the hydrological catchment behaviour. The influence of innovative stormwater techniques on the distribution of runoff sources in the catchment, and on total runoff, varies between scenarios. The biggest runoff reduction is delivered by widespread application of flat roofs.
Article
The footprint of a turbulent flux measurement defines its spatial context. With the onset of long-term flux measurement sites over forests and other inherently inhomogeneous areas, and the development of the FLUXNET program, the need for flux footprint estimations has grown dramatically. This paper provides an overview of existing footprint modeling approaches in the critical light of hindsight and discusses their respective strengths and weaknesses. The second main objective of this paper is to establish a formal connection between micrometeorological measurements of scalar fluxes and their mass conservation equation, in a surface-vegetation-atmosphere volume. An important focus is to identify the limitations of the footprint concept and to point out situations where the application of footprint models may lead to erroneous conclusions, as much as to demonstrate its utility and power where warranted. Finally, a perspective on the current state-of-the-art of footprint modeling is offered, with a list of challenges and suggestions for future directions.
Article
The urban microclimate is a result of the interactions between regional climatology, morphology of urban fabric and thermo-radiative and hydric properties of the materials constituting the urban surfaces. Modern urban planning tries to compact cities for avoiding spreading of the town area in order to reduce the need of transports. The climatic effects of this politic are not well known yet. This work is the first step of a study whose main purpose is to propose an urban planning politic which considers the urban heat island within the context of the climate change. Therefore it aims to understand the effects of building configuration on the momentum, heat and humidity turbulent transfers. Urban structures are quite heterogeneous and complex. To analyse the interactions between urban canopies and the atmosphere, the resolution has to be high enough that mean flow characteristics between the buildings can be observed as well as turbulent structures that develop above roof level. For that purpose, the Large-Eddy Simulation atmospheric model ARPS (Advanced Regional Prediction System) is employed. The subgrid scales are modelled by means of the resolution of an equation for the subgrid-scale turbulent kinetic energy. The version used is this study is based on the developments of Dupont & Brunet (2008) who introduced a drag force approach to simulate the influence of the vegetal covers on the low atmosphere. The application of this model on urban canopies requires adjusting the drag coefficient (Cd) in function of the frontal density (Af) and of the building arrangement. Its dependency on configuration (staggered or aligned arrays) and on frontal density is investigated. In order to put in light the main characteristics of the flow field and to validate the model in urban configurations, the turbulent flow above arrays of obstacles is investigated here for different arrangements and packing densities. The grid spacing is set to a few tens of meters in both horizontal directions and one meter in vertical direction, within the canopy. Above the canopy, grids are stretched vertically to reach four meters in average. The surface density Af is calculated in function of the geometry of the obstacles present in a grid cell. The effect of Cd values on the turbulent statistics is compared with results from wind tunnel experiments or building resolving simulations. The ability of the model to reproduce the major turbulent organized structures above the canopy and the influence of the building configuration is also studied, in terms of intensity, size and orientation of the structures. Further research will be done to investigate the effect of configuration of buildings with more complex geometries, like variable building height, on turbulent flow characteristics. To understand the climatologic features of a real city this approach will be completed by the introduction of a parametrization for heat and humidity exchanges inside the canopy.
Article
The circulation of rainwater within urban areas has not yet been described in a detailed manner, as studies on this topic often remain limited to the runoff on impervious surfaces. The need for innovative and sustainable methods of water management has incited increased research efforts on the hydrological processes at work in urban areas. A distributed hydrological model based on information supplied by current urban databanks has been developed in this aim. The components of rainwater flux (i.e. surface runoff, soil runoff, drainage flow via the sewer, and evapotranspiration) as well as information on the hydric state of the urban soil (saturation level, storage capacity) are modeled at the parcel scale, and then coupled with a detailed description of the hydrographic network. This model runs continuously and has been intended to reproduce hydrological variables over very long time series. In order to evaluate this model, it has been applied at two different scales, on two urban catchments of various land use, where hydrological data were available. This evaluation is based on the comparison of observed and simulated flowrates and saturation levels, and details the various compartments (soil, impervious or natural areas) to the outflow. This study shows the importance of water fluxes often neglected in urban hydrology, such as the evapotranspiration or the soil infiltration into sewers. This first evaluation has highlighted the capability of mapping most of the hydrological fluxes on urban catchments, such as the capacity of soil to store water.