Implementing WUDAPT into fast urbanization of mega-regions in China

Urbanization has been rapid in China since the 1970s. According to the latest report from the National Bureau of Statistics, it has been found that the urbanization rate reached 54.77% in 2014. However, the 13^th 5-year plan released by the Central Government of China in Feb. 2016 suggests that the future urbanization mode in China change from mega-cities to mega-regions. It means high density development and urban expansion in uprising cities and regions will be the trend. However, the associated environmental pressures are expected to be exacerbate the already serious urban heat island (UHI) effects and may also cause high public health risks. Thus, there is an urgent need for police makers and planners to understand urban morphology characteristics of Chinese cities and their associated impact on UHI. Due to lack of data accessibility and data harmonization of land cover and land use information, it turns to be a difficult and challenge to them.

The WUDAPT is a global initiative to use free available Landsat images to capture urban forms, function and land cover information by following the Local Climate Zone scheme (Stewart & Oke, 2012) and urban experts can use different WUDAPT products for various climatic and energy models and urban applications. The WUDAPT method developed by Bechtel and etc. al. in 2015 mainly focuses on city-scale. For meeting the challenge in China, an improved method for level 0 product is developed in this study.

In this study, first three major economic development regions including Jing-jing-ji region, the Yangzi River Delta Region and the Pearl River Delta Region are selected as case study. The background information of these three regions has been collected and different types of LCZ and training samples of seventeen climate-based classification categories have been obtained by visual interpretation from Google Earth. Training areas of each LCZ is about 50 polygons.  Second, collected Landsat data of three regions is preprocessed. For each region, seamless mosaic and atmospheric correction are applied to combine the collected Landsat images into one image. Resample the images’ resolution from 30m to 100m. Thirdly, the preprocessed Landsat images and the selected training areas of the whole region are inputted together into the SAGA-GIS software. And then, the LCZ map for each region can be generated. Fourthly, the accuracy assessment is conducted to check the product’s quality. The urbanization and geographical characteristics and problems of three regions are analyzed and discussed. Finally, the LCZ map of municipalities can be extracted from the regional LCZ map if needed. At the same time, the regional map serves as information platform to provide the background information for urban climatic application at the city-level. All the study results will be uploaded on the WUDAPT official open on-line database and made available to policy makers and planners. The study will contribute to the scholarly understanding of urbanization in China. In practice, the developed regional LCZ map serves as information platform to provide the background information for urban climatic application at both city and regional level. The findings will also inform “The environmental performance assessment guide of urban ecological development of China (Trial Version)”.

 (*This abstract is preferred to submit to the special WUDAPT session)