Spatial-temporal analysis of building surface temperatures in Hung Hom

摘要

This thesis presents a study on spatial-temporal analysis of building surface temperatures in Hung Hom. Observations were collected from Aug 2013 to Oct 2013 at a 30-min interval, using iButton sensors (N=20) covering twelve locations in Hung Hom. And thermal images were captured in PolyU from 05 Aug 2013 to 06 Aug 2013. A linear regression model of iButton and thermal records is established to calibrate temperature data. A 3D modeling system is developed based on Visual Studio 2010 development platform, using ArcEngine10.0 component, Microsoft Access 2010 database and C# programming language. The system realizes processing data, spatial analysis, compound query and 3D face temperature rendering and so on. Building materials are found to have a significant effect on building surface temperatures. Buildings with lower albedo materials tend to have higher temperatures and larger thermal conductivity material have significant diurnal variations. For the geographical locations, buildings located at the southeast are cooler than the western. Furthermore, human activity is found to have a strong relationship with building surface temperatures through weekday and weekend comparison.