The WUDAPT Platform as a Basis for Global Scale Urban Climate Modeling

Over the last few decades, urban climate models have evolved significantly from standalone systems to simulate the climate of specific locations, to modeling systems that simulate urban climate, air quality and human energy consumption across widely different climate types and socio-economic conditions globally.  Example modeling frameworks include regional and global scale community-based modeling systems such as the Community Earth System Models (CESM) Community Land Model (CLM) – Urban component (CLM-U), several Weather Research and Forecasting (WRF) urban components, the Town Energy balance Model (TEB), the Local Scale Meteorological Parameterization Scheme (LUMPS), and similarly configured community system models for air quality such as the Community Multi-scale Air Quality (CMAQ).  Ultimately each of these modeling frameworks aims to provide guidance towards meeting the challenges of climate change, air quality impacts, human comfort and urban sustainability.

While each of these models continue to refine model physics and dynamics, as well as improving temporal and spatial resolutions of their simulations, significant improvements in model performance will also require better model input parameters.  Included in this is information about city morphology, thermal properties of urban fabrics, human choices about energy consumption (e.g. use of air conditioning and heating), energy releases by transportation systems, and information about potential mitigation strategies.  Currently many input requirements are provided through top-down systematic data collection with satellite platforms.  However, many critical parameters are not as easy to collect as they depend on human decision-making and behaviors that cannot be observed remotely.  For example, building materials can have widely different thermal and radiative properties and energy consumption by buildings is in large part dependent on choices of heating and cooling systems and the relative amount of use of these systems as controlled by thermostats or local access to energy resources.  Since climate mitigation and adaptation schemes are heavily dependent of human energy consumption, these types of variables must be collected, not only to simulate urban climates in the present, but also to provide guidance for policy relevant assessments of human energy consumption and impacts on urban climates and air quality under different mitigation strategies. The World Urban Database and Access Portal Tools (WUDAPT) data collection effort is designed to provide the type regionally specific and relevant information that can be used by a variety of different models to better represent regional urban characteristics and human behavior within these regions. 

The World Urban Data and Access Portal Tools (WUDAPT) project aims to map cities into like neighborhood scale morphology classes and in addition to obtain information on human behaviour in cities that determine climate and energy outcomes. Using a combination of satellite data, the Local Climate Zone (LCZ) urban classification scheme (Stewart and Oke, 2012), and local expert knowledge cities are be classified into separate LCZ classes (Bechtel et al. 2015).  Once the areas are defined, typical LCZ urban morphology attributes can be attributed to these urban regions, or where applicable the local experts can narrow the information down to reflect the unique attributes to specific city locations.  The use of local experts also allows exploration of additional information about the LCZ regions.  For example, beyond urban morphology the local expert can provide information about the most common building materials used and typical uses of buildings within a specific LCZ.  This information can be translated into typical thermal and radiative properties representative of that particular setting.  Local experts may also be able to provide information on typical human use of the building, for example by specifying the types of heating and cooling systems typically used in a location and potentially the human actions that control these systems (e.g. typical thermostat settings).  

The focus of this paper is on the need to map urban parameters on the global scale.  Jackson et al. (2010), showed that improved representation of urban characteristic based on broad based generalizations can make significant differences in outcomes.  In this simulation with the CESM CLMU model, there is a systematic difference in the model output simulation of urban heat island calculations when a single city set of parameters (Vancouver based on values from Voogt and Grimmond, 2000) is compared to data derived by Jackson et al. (2010).  Results show that urban heat islands are significantly lower in the tropics when more representative data are used over 33 regions of the world.  The cause of the reduced heat island could be due to a number of factors, including the vegetative fraction in the urban fabric of cities, the morphology, and thermal and albedo values that characterize regional cities. 

While Jackson et al. (2010) was a first attempt to capture this information on a global scale, WUDAPT is an effort to greatly improve the quantity and quality of the data to characterize urban properties globally, with significant input from local experts.  Similarly, a study by Oleson et al. (2010) demonstrates the potential for models to assist with evaluating costs and benefits associated with policies to mitigate urban climate change.  Oleson et al. (2010) used a modified Jackson et al (2010) input data set to demonstrate that introducing high albedo roofs (white roofs) on a global scale would reduce urban heat islands in most regions compared to present conditions, while leading to an overall increase in energy consumption due to increased heating needs in cold regions.  Once again improved information about realistic scenarios from local experts can aid in improving experimental design.

We are now completing a set of experiments that further investigate these potential energy and heat island mitigation strategies on a global scale.  However, such studies can be greatly improved if there are better base line data; as WUDAPT is intended to do.  With more knowledge of present day city conditions and how urban climate policies are implemented in different regions; a process that can be made more realistic using WUDAPT data and information about local resources available for implementing urban climate mitigation and adaptation strategies.