During the past few decades, the urban development’s impact on urban climatic condition has been reported throughout the world. It is well recognized that site morphology and geometry can modify the local climatic condition and form its unique climatic characteristics. Thus, describing physical site properties in a scientific and precise way is important and necessary for urban climatic analysis and application into urban planning. Tim Oke initiated a scheme of urban climate zone (UCZ) in 2004 (Oke, 2004 & 2006) and later in 2009 Iain Stewart and him further developed a refined comprehensive approach of “local climate zone (LCZ)” classification by using a standardized and quantitative way to present physical surface properties of sites and their local climate features(Stewart & Oke,2009 & 2012).

LCZ classification has 17 standard types including two subsets: 10 built types and 7 land cover types. Each LCZ can be defined quantitatively by using a standard set of parameters. Some case studies have been conducted in cities of Sweden, Japan and Canada, which are located low density city. Unfortunately, there has been no study testing LCZ classification in high-dense cities. The study wishes to fill this knowledge gap by focusing on high density built-up types and land cover types of Hong Kong. Using field measurement, computational numerical simulations, information regarding the metadata and meteorological data could be collated.

First, this study will collect site metadata through the field visit study and up to 20 sites would be surveyed. The survey sample sites will take into account all parameters of LCZ including sky view factor, canyon aspect ratio, building surface fraction, impervious surface fraction, pervious surface fraction, surface admittance, surface albedo and anthropogenic heat flux. Secondly, an experimental study of defining the thermal source area by using thermal sensor and infrared cameras will be also conducted. Thirdly, the study will use both local and micro scale models to do the numerical simulations to capture the thermodynamic features and surface energetics of each survey site. Then, the collected data and simulation result will be analyzed and classified into ‘LCZ’ of Hong Kong. It would allow not only climatologists, but also planners and governors to gain insight into the “how much, where and what” questions of local climatic condition. The information would allow the formulation of urban planning guidelines and climatic spatial design strategies based on a set of threshold value of urban morphology and its climatic impact.

Data collected by Earth Observation (EO) satellites provide a valuable source of information for understanding, monitoring, modelling and thus protecting the environment. The increasing availability of EO systems and the advances in remote sensing techniques increase the opportunities for monitoring the urban environment and its thermal behaviour. Several parameters related to the urban climate can be derived from EO data, providing valuable support for advanced urban studies and climate modelling. Recently, attention has been drawn to the quantitative description of the urban thermal patterns and their correlations to fundamental surface descriptors. Recently, a detailed classification scheme of Local Climate Zones (LCZ) was introduced, based on various former typologies, which explicitly defines urban landscapes according to their thermal properties. The scheme aims to be objective (incorporating measurable and testable features relevant to surface thermal climate), inclusive (sufficiently generic in its representation of local landscapes to not inherit regional or cultural biases) and standardized. The individual classes aim to have relatively homogeneous air temperature within the canopy layer and they are defined by fact sheets with both qualitative and quantitative properties, including several features that can be derived from EO data. In this study, advanced remote sensing techniques are applied and quantitative information required for discriminating between LCZ is derived. Parameters like the pervious and impervious surface fraction and the surface albedo were quantified for a case study (the city of Heraklion, Greece), using EO data. Combining ancillary information for the geometry of the city, parameters like the buildings density and mean building height, the canyon aspect ratio and the terrain roughness are estimated. Having estimated all those properties, a methodology was established to outline possible LCZ in the urban fabric with different thermal behaviour. Time series of land surface temperature estimates derived from satellite data are also employed to identify the thermal patterns and their correspondence to the LCZ.

(Stewart and Oke, 2009) have acknowledged the need for a more accurate knowledge of ground description at micro-scale levels in order to improve consistency and accuracy in urban climate reporting. To address this, they propose to use Local Climate Zone (LCZ) as geographical objects allowing a finest description of landscape, whatever its nature, urban or rural. We agree that this could hence enhance a lot the understanding of Urban Heat Islands phenomenon, but we are also convinced that LCZ should include somehow the knowledge about citizens behaviour in their in-doors: do they open windows during winter? Do they use intensively air conditioners? This approach has been funded by the French National Agency for Research, in a project named MaPuce. One part of this project consists to automate the computing of urban morphological indicators as well as estimate households’ behaviour at the micro-scale levels, in order to incorporate those quantitative data in urban micro-climate simulations. The project specifies that data sources should be available everywhere in France, for free, which is compatible with the goal of contributing to World Urban Databases.

In this proposal, we describe a methodology to automatically build LCZ knowledge from French databases available for free to research and academic domain, covering the national territory: topographic data are provided by the French National Agency, (IGN) and socio-economical datasets are downloaded from the French National Institute for Statistical and Economic Studies (INSEE). The originality of this task is to develop a vectorial approach using geometric shapes of the buildings, roads, vegetation and hydrographic objects, but also by using cadastral parcels to delimit new LCZ objects. It mixes various sources, in order to build an enriched and qualified dataset. However this task is difficult because of the various inconsistencies (buildings may overlap, intersect, or appear twice) in the sources. Furthermore, the income and education level, the age, and the size of the residents’ households have to be downscaled at the LCZ level in order to be able to estimate household behavior for energy consumption.

We show in this contribution how this task can be achieved, using programs we have developed under open-source license, and we introduce a zonal object that implements the LCZ specification: the urban islet. It groups as set of buildings separated from each other’s by roads or streets or rivers, grouping a set of contiguous cadastral parcels, and it allows for the automatic computing of an accurate topographic and socio-economic classification fitting the LCZ’s one. A comparison with LCZ specification is made in order to check whether an islet answers fully the needs raised by LCZ (Stewart I. D. and Oke, T. R. 2012). The workflows automatically fixes geometric and topographic errors in sources, mixes them to build islets with their associated land use. In a second step, it computes automatically socio-economic proxies (like number of inhabitants, households, distribution of households income, etc.) at the building level in order to aggregate data at the LCZ level. We make a comparison with results that could have been obtained using a fine grid of 250m by 250 m. Finally, we show how this can be used in micro-urban simulation, feeding models like TEB (Town Energy Balance), (Masson, 2000). One of the perspectives of this work is also to enhance Energy-Climate assessment, by estimating energy consumption of households at this micro-scale level, and publishing this information for urban planning.

References

Masson V., 2000 : A Physically-based scheme for the Urban Energy Budget in atmospheric models. Boundary-Layer Meteorol., 94, 357-397

Stewart, I. and Oke T. 2009 Classifying urban climate field sites by “Local Climate Zones” – The case of Nagano, Japan. 7th International Conference on Urban Climate, 29 June - 3 July 2009, Yokohama, Japan

Stewart I. D. and Oke, T. R. 2012: Local Climate Zones for Urban Temperature Studies. Bull. Amer. Meteor. Soc., 93, 1879–1900.doi: http://dx.doi.org/10.1175/BAMS-D-11-00019.1