URBAN HEAT ISLANDS ■ 179TABLE 7.1
Regional Climate Model Experiments
Run name PeriodAnthropogenic heat
fl ux (W m−2) Notes
a 1971–90 0 Urban fractions set to zero; urban
temperatures calculated at every
location within the model domain, but
only nonurban tile climates feed back
onto the modeled atmosphere
b 1971–90 0 Fully coupled urban areas
c 1971–90 15/25 As run (b) plus anthropogenic heat fl ux
d 1971–90 45/75 As run (b) plus tripled anthropogenic
heat fl ux
e 2041–60 0 Fully coupled urban areas
f 2041–60 15/25 As run (e) plus anthropogenic heat fl ux
g 2041–60 45/75 As run (e) plus tripled anthropogenic
heat fl uxNote: RCM control run with rural surfaces only feeding back to modeled climate. Urban climate calculated
but not used in the simulation. The urban fractions have all been set to zero, so surface fractions of rural
tiles have been increased where necessary so they sum to 1. (b) Same as run (a) but urban areas included
fully in simulation. A comparison of runs (a) and (b) allows any feedbacks between the urban climate and
the larger modeled climate to be quantifi ed. Run (c) as (b) but with an anthriopogenic heat source to the
urban tile included. Run (d) as (c) but the anthropogenic heat source is tripled. Runs (e), (f), and (g) are
repeats of runs (b), (c), and (d), respectively, for the future period 2041–2060.
tures to climate change and, in the case of urban areas, to diff erent assumptions
regarding anthropogenic heat release. For all the regional climate simulations,
suitable boundary conditions were supplied from a climate projection for the
period 1950–2099 created with the global model HadCM3 (Collins and others
2006). Th is global model simulation used greenhouse gas emissions from a
medium-high emissions scenario (A1B; Nakićenović and Swart 2000). Th is
scenario assumes rapid introduction of new and effi cient technologies, with
a balance between fossil fuel use and alternative energy sources (IPCC 2007).
Results
Th e results obtained from the model experiments are discussed next.
Validation of Modeled Temperatures
Simulations of surface air temperatures from run (a) (see table 7.1) are compared
with observations in fi gure 7.1 to validate the model. Th ese observations have