Urban Soil-Canopy-Atmosphere Exchanges at Submesoscales: Learning from Model Development, Evaluation, and Coupling with LES - Meteorological and Air Quality Models for Urban Areas - page 41

- extending the tiling approach by adding in each cell the separate computation of
surface budgets for each typical urban surface cover mode, pavement with or with-
out trees, grounds with dispersed vegetation, and building roofs, considering the
(semi-)impervious surfaces as water reservoirs with run-off to the neighbouring
tiles and/or to the drainage network (Guilloteau, 1999);
- introducing Guilloteau's (1998) non-iterative algorithm to compute the fluxes with
non-equal momentum and heat roughness lengths;
- Bottema's models to compute roughness length from the building morphological
parameters (Bottema, 1996, 1997; Bottema and Mestayer, 1998; Mestayer and
Bottema, 2002).
In this initial “flat canopy” approach, only the (quasi-)horizontal surfaces were
modelled. Guilloteau and Dupont (2000) demonstrated that this was not sufficient to
simulate correctly the diurnal cycle of the urban surface energy components, which
requires the inclusion of the thermo-radiative influence of building walls. The sec-
ond stage consisted in adding model equations for :
- energy transfer processes for walls and roads including storage,
- shadowing and radiation trapping between neighbouring buildings, with
geometry-dependent “effective” albedo and emissivity of streets (Dupont, 2001),
and
- influence of canopy shape in the heat transfer through the canopy layer (Piringer
et al., 2002) by means of Zilitinkevich's (1995) heat roughness length formula.
The most recent development consisted in adding a scheme for computing the
energy budget of the (coastal) sea surface, based on the LKB model (Liu et al.,
1979) and Smith's (1988) formula for the momentum roughness length (Leroyer,
2006).
6.3 Model Validation and Implementation
The UBL-Escompte campaign over Marseille urban area (Mestayer et al., 2005)
provided with observational data sets for evaluating urban surface energy budget
(SEB) schemes. SM2U was tested and validated against the data obtained at the city
centre (Grimmond et al., 2004) (Figs. 6.2 and 6.3).
The model allows evaluation of the role of the relative influences of the cover
types. Figure 6.4 shows the phase shift between components, with the slow rise of
the turbulent sensible heat flux in the morning due to the heat storage in the mate-
rials. After 3 pm and during the night the heat stored is released. At night this is
nearly equivalent to the radiation loss. This behaviour is not apparent in the vegeta-
tion fraction fluxes (“natural” surfaces), where the available energy is about equally-
partitioned between the sensible and latent heat fluxes with little heat storage and a
small nocturnal negative radiation flux compensated by the sensible heat flux. For