World Database and Access Portal Tools (WUDAPT) Level 0 method announced a workflow of mapping Local Climate Zones (LCZs). However, the low accuracy of LCZ classifications in Level 0 especially for the built-up areas caused by recognition of classes and operator bias is becoming an obstacle for further study in WUDAPT Level 1 and 2. Since the landscape in Japan is complicated, the recognition of classes and operator bias may exist for delineating training areas. This article argues an extended workflow of WUDAPT for mapping LCZs with pre-set recognition of classes and parameter analysis. The building coverage ratio (BCR), building height (BH), pervious surface fraction (PSF) were intersected with LCZ map for analysis and expound of the pre-set recognition of LCZ classes. Given the universality of WUDAPT workflow, a satellite method for deriving building data based on free available data sources was proposed. Contributing to WUDAPT level 1 and 2, a LCZ classification of Sendai, as a representative of Japanese large cities, was selected. The study will provide not only an improved methodology of development LCZ data, but also a new urban morphological dataset and its corresponding parameters for mesoscale climate modelling and simulations in Japan.

世界数据库和访问门户工具（WUDAPT）的0级方法宣布了绘制本地气候区（LCZ）的工作流。但是，由于等级识别和操作员偏见而导致的0级LCZ分类的准确性较低，尤其是对于建筑物区域，这正成为WUDAPT 1级和2级进一步研究的障碍。由于日本的情况复杂，因此识别类别和操作员的偏见可能会划定培训区域。本文讨论了WUDAPT的扩展工作流，该工作流用于通过预先识别类和参数分析来映射LCZ。将建筑物覆盖率（BCR），建筑物高度（BH），透水率（PSF）与LCZ图相交，以进行分析和阐述对LCZ类的预设识别。鉴于WUDAPT工作流程的普遍性，提出了一种基于免费可用数据源的建筑数据推导卫星方法。为了达到WUDAPT 1级和2级标准，选择了仙台的LCZ分类来代表日本大城市。该研究不仅将为开发LCZ数据提供一种改进的方法，而且还将为日本的中尺度气候建模和模拟提供一个新的城市形态数据集及其相应参数。