Premium
A building energy demand and urban land surface model
Author(s) -
Lipson Mathew J.,
Thatcher Marcus,
Hart Melissa A.,
Pitman Andrew
Publication year - 2018
Publication title -
quarterly journal of the royal meteorological society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.744
H-Index - 143
eISSN - 1477-870X
pISSN - 0035-9009
DOI - 10.1002/qj.3317
Subject(s) - environmental science , energy consumption , meteorology , neighbourhood (mathematics) , climate model , urban climate , atmospheric sciences , latent heat , energy flux , energy budget , urban heat island , climate change , urban planning , geography , civil engineering , mathematics , engineering , ecology , geology , mathematical analysis , physics , astronomy , electrical engineering , biology
Cities are unique environments where anthropogenic waste heat from energy consumption changes the dynamics of the boundary layer, affecting temperatures, pollution dispersion and buoyancy‐driven flows. Although urban environments are important for societal wellbeing, there are relatively few models that link predictions of waste heat variability with urban climate interactions. This study presents the Urban Climate and Energy Model (UCLEM): a new physically based model representing important heat transfer processes between the atmosphere, external and internal urban environments, combined with a statistical model of human behaviours relating to energy use. The aim of UCLEM is to efficiently predict the climatology of different urban areas, as well as the energy used to maintain comfortable temperatures within buildings. The model is designed to be easily adaptable to a wide range of urban settings with adjustable parameters including building height, density, material thermal and radiative characteristics and vegetation fractions. We assess UCLEM's ability to predict energy consumption for a neighbourhood of Melbourne, Australia, forced by local flux tower observations and evaluated at half‐hourly intervals over 12 months. Results are presented in four development stages to assess various levels of physical and behavioural model complexity. We show that more complete physical representations can improve average daily energy consumption predictions; however, sub‐daily patterns of energy use are improved only by combining the physics‐based model with a statistical model of human behaviour. At the final stage, as well as predicting surface–atmosphere radiant and turbulent fluxes, UCLEM estimates neighbourhood energy demand with a normalised mean error of 11.5% and a computation time on a single processor of about 1 s per simulation year.