Urban Surface Energy Balance Models: Model Characteristics and Methodology for a Comparison Study - Meteorological and Air Quality Models for Urban Areas - page 93

11.2.3 Model Inputs
Inputs consist of three general types: (1) the parameters to describe the site; (2) the
variables required to drive the model; and (3) the initial conditions required to ini-
tiate the model runs. Obviously, the first set of information is needed in any type
of model run, so there is no difference between a typical run and an offline model
evaluation run. For the second set of data, for a normal online run, would be the
meso- (or larger-) scale model with data 'passed down' to the surface scheme to
force it. These data are therefore updated at each time step. The third set of input,
initial conditions, are explicitly related to the surface scheme rather than to wider
models of which these schemes may be a part. For some models and some variables
these may be the same as (2) but for others there are additional requirements.
The complexity of urban areas demand a large number of model parameters;
here they are sub-divided into two groups: the built environment (Table 11.6) and
urban vegetation (Table 11.7). In terms of the morphometric characteristics of the
built form, inputs vary greatly depending on whether basic information is used (e.g.
height and width) from which the required parameters are calculated (e.g. canyon
aspect ratio, sky view factor), or if 'higher' level parameters are inputs. This can
give the impression that there are larger differences in the model inputs than there
really are. Other parameters of the built environment relate to the nature of the
materials used and include parameters that are concerned with radiative transfer
(e.g. albedo, emissivity) and conductive characteristics. These characteristics may
be specified in different ways relative to urban form; for example, relative to mass
(specific heat capacity) or volume (volumetric heat capacity). Alternatively, mate-
rials may be specified and model 'look-up' tables used to assign the appropriate
parameters. The details of the tiled-models that draw on vegetation schemes are not
summarised in Table 11.7 because they are extensively evaluated (see Sect. 11.2.2).
Typically, the vegetation characteristics are assigned by using a default number of
classes (e.g. 11 in SM2U, 5 in MOSES).
Anthropogenic heat flux is dealt with in a wide variety of ways. For example,
some schemes capture this flux by specifying fixed internal temperatures and traffic
counts (Table 11.6), which provides temporal dynamics to the flux estimates. Other
models require the flux to be provided as a direct input (e.g. GCTTC, CAT).
The variables that are used to drive the models are listed in Table 11.8 . These
relate to wind, temperature, humidity, radiation, and soil characteristics and as noted
above, some models also require the anthropogenic heat flux to be directly sup-
plied as an input. From the nature of the inputs, it is evident that a wide variety of
approaches is used, for example, to determine the radiative forcing. Some models
calculate radiation, others take the short and long wave radiative fluxes as a direct
input, while others add further detail by differentiating between direct and diffuse
components (e.g. MUCM, CLMU, VUCM, CAT, SUNBEEM).
The final set of inputs relates to the initial conditions. The information required
about temperature profiles within a building or the soil may be the most significant
for some of the models. Such data are typically difficult to obtain. One consequence