Urban Surface Energy Balance Models: Model Characteristics and Methodology for a Comparison Study - Meteorological and Air Quality Models for Urban Areas - page 101

Table 11.11 Methods to calculate turbulent sensible heat flux (Q H )

Model Turbulent sensible heat flux methods

BEP02, BEP05,
SUNBEEM

For walls based on Clarke (1985)

CAT Resistance between canyon surfaces and air based on Hagishima

and Tanimoto (2003); at canyon top depends on stability, using
empirical parameterization

CLMU Resistance network accounting for differences between surfaces

ENVImet From turbulence model (wall function) and surface energy balance

GCTTC Calculated for each surface based on the attenuated radiation by the

CTTC factor

LUMPS deBruin and Holtslag (1982) modified Penman Monteith modified

for urban areas (Grimmond and Oke, 2002)

HIRLAM-U, MM5u Parametric formulation based on the specific heat capacity for

moist air, the density of the atmosphere, the surface friction
velocity and the surface temperature scale

MOSES2T, MOSES1T Standard resistance, based upon MO similarity theory

MOUSES Resistance network based on Harman et al. (2004)

MUCM MO or Jurges

MUKLIMO From surface energy balances at the soil, walls and roofs using MO

laws

NSLUCM MO based on Louis (1979) and Jurge's formulation, and calculated

from each surface

SM2U MO Resistance (Guilloteau, 1998; Zilitinkevich, 1995)

SUEB MO similarity Louis (1979) modified by Mascart et al. (1995)

SUMM Resistance (top-down method, Kanda et al., 2005)

TEB, TEB07 Resistance

TUF2D, TUF3D Resistances based on flat-plate heat transfer coefficients (vertical

patches) and based on MO similarity (horizontal patches)

UCLM Exchange based on canyon air and surface temperature difference,

wind speed and prescribed heat transfer coefficient.

VUCM Parametric formulation

The storage heat flux methods involve, amongst others, empirically-based
approaches such as objective hysteresis model (OHM)(e.g. MM5u, LUMPS,
HIRLAM-U, CAT) and thermal diffusion approaches (Table 11.12). Models use
varying numbers of layers to represent substrate materials, and as noted in the model
inputs, materials are described in a wide variety of ways with implications for how
the heat storage term can be calculated.
The methods used to calculate drag include roughness length approaches and
distributed drag within the atmosphere (Table 11.13). Those that distribute the drag
within the canopy might be expected to require more computational time and have
greater data needs to describe the urban morphology.
A wide range of approaches are used to calculate the latent heat fluxes reflecting
a range of possible representations of vegetated and/or wet surfaces. Some models
assume that the urban area is dry and therefore ignore the latent heat flux completely;
others have wet built surfaces but no vegetation; and some include vegetation as
either a separate tile or as integrated (Table 11.3, 11.7, and 11.14). As with the