Environmental Engineering Reference
In-Depth Information
of an hour-by-hour evaluation along with comparisons over longer (daily, monthly,
seasonal, annual) time periods. The statistics used will include a range of metrics
(mean, standard deviation, probability distribution function, linear regression, root
mean square error (systematic, unsystematic), index of agreement, mean absolute
error, mean bias error, correlation coefficient, coefficient of determination, etc.).
Each provides insight into different aspects of model performance.
The analysis will also assess the ability of each model to simulate known urban
climatological phenomena. It is now well documented that the energy exchange
processes in cities (relative to a rural area) are modified by the presence of buildings
and other anthropogenic structures in the following key ways:
•
urban areas reflect less shortwave radiation, due to trapping and multiple reflec-
tions between buildings (Arnfield, 1982)
•
materials used in urban areas have high thermal heat capacity, i.e. there is 'storage
heat flux' into the buildings by day (Grimmond and Oke, 1999a; Offerle et al.,
2005) and significant release at night
•
the surface area is increased, which combined with materials of a high heat capac-
ity, means that more heat is absorbed and emitted (Harman et al., 2004b)
•
the morphology of buildings affects flow, and thus determines the rate of
exchange of heat with the air above (Barlow and Belcher, 2002; Barlow et al.,
2004)
•
buildings and traffic generate additional anthropogenic sources of heat
(Grimmond, 1992; Sailor and Lu, 2004; Offerle et al., 2005)
•
evaporative cooling is decreased due to reduction in vegetation cover; thus the
latent heat flux may be relatively small (Grimmond and Oke, 1991; Grimmond
and Oke, 1999b)
•
positive sensible heat fluxes are more probable at night in highly built up areas
(Grimmond and Oke, 2002)
Many of the models under consideration also predict variables beyond the surface
energy balance terms. It is often these variables that are of specific interest in many
applications. For example, air temperature and humidity are reported as part of a
weather forecast but are also of interest for health and air quality applications. Sim-
ilarly, atmospheric stability and wind speed are of interest in pollution dispersion
applications.
By staging the comparison and considering model performance individually and
across groups of models defined in terms of key attributes, we will aim to address
the following four key questions:
1)
What are the main physical processes that need to be resolved to simulate
realistically, urban energy balance exchanges?
This will be addressed by
grouping models in terms of the processes that they represent, to determine
whether models which represent certain processes produce significantly better
results. By staging the comparisons so that the experiment finishes with the
optimisation of the parameters for each model, it will be possible to determine
Search WWH ::
Custom Search