Search

Climate zones play an important role in promoting climate responsive building design and implementing climate-specific prescriptions in national building standards and regulations. The existing studies on climate zoning are subject to several limitations, i.e. the incapability of distinguishing microclimates and the lack of consideration of climate change. In this research, we propose a two-tiered ensemble clustering method for the identification of granular climate zones using the projections of future climate. The first tier identifies primary climate zones using a combination of climatic features and geographical locations, whereas the second tier identifies distinct local variations within each primary climate zone using the temperature related features. The proposed ensemble clustering model is applied to the UK to create a mapping of granular climate zones for future proofing building design. The method identified 14 distinct primary zones and distinguished microclimates at a range of scales from large urban areas, such as the Greater London Area, to national parks, such as Dartmoor and the Pennines. The identified mapping resolves two major obstacles in the creation and usage of weather data for building performance assessment in the UK, i.e. the lack of guidance for selecting weather files, and the absence of scientific rationale for representing the UK climate using the current 14 locations. • A hierarchical ensemble clustering method was proposed to uncover spatial patterns of climate. • The latest climate data from the UK Climate Projection 2018 were employed. • Representative climate zones were created to enable climate-responsive building design. • Microclimates such as large urban areas and national parks were also identified. • The climate zones enable more accurate consideration of local climate for building design. [ABSTRACT FROM AUTHOR]