Studying the altered urban environment is important because of the high number of the involved inhabitants. In urban areas surface cover and geometry differ from the rural surfaces, and the water and energy balances are modified. Since the thermal environment and the airflow conditions are modified, these modifications affect the energy consumption and wind energy potential in urban areas. The evaluation of the urban surface geometry and its parameters are not straightforward, and still a rapidly developing field of research. In this paper new software methods for the calculations of urban surface parameters are presented. These software tools enable us to evaluate the building and tree-crown databases from spectral and elevation data, and to use these two datasets to calculate SVF and roughness parameters. These software tools were applied for study areas in several cities in Hungary, in order to test the methods and to gather information about the variance of the urban surface parameters in different sized urban areas. The outcome of this software method (building dataset, tree crowns, SVF, roughness parameters) is widely used in urban climate studies, thus a new and time effective method can help several urban climate studies, where these parameters are needed to describe the urban surface.

Classification of land cover in urban areas can play an important role in urban planning decisions and in characterizing urban materials properties such as reflectance. Taking into account the large offer of new and future remote sensing sensors with different spectral and spatial characteristics, it is important to compare their classification performances in urban area. To this aim, this work simulates from airborne data the at sensor images acquired by three space borne instruments in the Visible Near Infrared (0.4 µm – 1.0 µm) (Pléiades, SENTINEL-2 and HYPXIM) and in the Shortwave Infrared (1.0 µm-2.5µm) (SENTINEL-2 and HYPXIM) spectral ranges. Five classification maps with 8 land cover classes over the city of Toulouse (France) are generated with a Support Vector Machine rule. Correct values of accuracy are obtained in all cases (kappa coefficient higher than 0.65 and overall accuracy better than 70 %). Results show that SWIR data is necessary to discriminate between classes and the coarse spatial resolution is compensated by the spctral richness of the hyperspectral images.

Urban areas already contain 54% of the global population, with projections indicating an additional 2.5 billion urban inhabitants by 2050. To project urban growth scenarios, we need to understand the complex dynamics of urbanization. Urban scaling theory states that properties of urban infrastructure and socioeconomic production are scale-invariant functions of population, and that scale transformations can be applied across spatial and temporal scales. Recent work by Luis Bettencourt and colleagues have provided evidence of urban scaling using a range of urban properties such as total impervious surface area or total volume of roads in a city as a function of city population size. How well do projections of future urban areas correspond with the urban scaling relations observed in the early part of the 21st century? Here we use the percent developed imperviousness (%ISA) data (30 m) from the USGS (US Geological Survey) National Land Cover Database for 2001, 2006, and 2011. We characterize the relationship of urbanized area to population size for 212 metropolitan and micropolitan statistical areas located in the central United States. Next we compare the scaling factors derived from the 30 meter resolution %ISA dataset to the scaling factors found using the coarser spatial resolution (1 km) global urban extent dataset created by Jackson et al. (2010). We then evaluate the scaling relationships in two different urban growth projection datasets, each of which uses four of the IPCC Special Report on Emissions scenarios to guide projections. The first dataset was produced by the USGS for the National Assessment of Ecosystem Carbon and Greenhouse Gas Fluxes at a spatial resolution of 250 meters and is provided annually from 2006-2100. The second dataset was produced by the US Environmental Protection Agency for the Integrated Climate and Land Use Scenarios (ICLUS) at a spatial resolution of 1 km provided decennially from 2010-2100. We characterize the relationship of urbanized area as a function of population size using the two urban growth projection datasets for the years 2030, 2050, 2070 and 2100, and also analyze the changes in urbanized area and population. We are interested in the viability and efficacy of using urban scaling theory to project urban growth for modeling purposes.

Mesoscale modeling is an excellent tool in analyzing urban atmospheric interactions within the ABL; phenomena such as the urban heat island and local heavy rainfall can be investigated. Prevailing challenges to this approach are urban representation from available land use databases and its parameterization. On the other hand, global satellites have been advancing rapidly in the past decades. Taking advantage of the information obtained from satellites, land use database can be created as one approach to fulfill the requirements of detailed urban surface representation. Our focus is on the potential of the new Landsat 8 mission, recently launched in February, 2013. In global land use database construction, validation of available land use remains a challenge especially in developing countries. We aim to (i) conduct land use classification from Landsat 8 imagery in creating urban surfaces representatives in 4 urban classes: commercial, high-density urban, low-density urban, and other urban areas; (ii) validate the land use classification by using ancillary data of building footprint; and (iii) evaluate the correlation between urban classes with available plane area indices (PAI). The study areas are Tokyo Metropolitan Area, Japan and Greater Jakarta, Indonesia. Semi-automatic land use classification plugin available in QGIS was used; the plugin also includes the Top of Atmospheric Reflectance (TOA) method for Digital Number (DN) conversion to reflectance and Dark Object Subtraction 1 (DOS1) for atmospheric correction. We define five major classes: water, vegetation, bare soil, agriculture, and urban with detailed minor classes. Region growing algorithm is utilized in creating supervised training data for hyper-spectral signature collection. Spectral Angle Mapping (SAM) classification using constructed hyper-spectral signature files is conducted on multitemporal imageries of both study areas to define land use classes mentioned above. Post-processing was then conducted to remove clouds and shadows. Validation of our derived classification for Tokyo Metropolitan Area was done using high resolution digital map provided by Geospatial Information Authority of Japan. Google Earth and other open source map ancillary data containing built-up area information was considered for the validation in Jakarta. Results showed that the Landsat 8-derived land use classification for urban built-up areas obtained 96% accuracy and each urban class has high agreement with corresponding PAI parameter.

The urban climate is one the key variables to be considered in sustainable city design facing climate change and global warming. In order to improve thermal comfort, energetic efficiency and air quality, bioclimatic guidelines, based on Urban Climatic Maps (UCMaps), are important and recognized tools to attain a reliable urban management. Among several features included in UCMaps, urban density is one of the most important variables to consider, due to its effect of urban volumes on radiative and energy balances and on natural ventilation. However, this key factor to urban climate analysis is very often unavailable or not suitable for urban climate studies and its determination becomes a very time consuming task and difficult to carry out.

In order to get urban density maps, a methodology developed with a GIS environment is presented in this paper. Buildings geometry and volumetric parameters will be automatically computed as indicators of urban density and morphology.

With a GIS, the urban area is divided into cells (the size of the cells will be discussed in detail in the communication). For the densities calculation, each cell will take into account the height of buildings, the surface area and volume, the width of the streets and the area exposed to the prevailing wind (it is also possible to consider other directions). The aerodynamic roughness (z0) inside each cell will be also considered in the automation process.

This methodology will allow us to speed up the process of creating bioclimatic indicators that can be use in applied urban climate studies. The research presented here are the result of a collaborative research between CEG/Zephyrus research team (University of Lisbon) and the Municipality of Cascais (Portugal), that were applied in the Bioclimatological assessment of the Municipal Master Plan.