Nonlinear black-box models for short-term forecasting of air temperature in the town of Palermo

Abstract

Weather data are crucial to correctly design buildings and their heating and cooling systems and to assess their energy performances. In the intensely urbanized towns the effect of climatic parameters is further emphasized by the “ urban heat island” (UHI) phenomenon, known as the increase in the air temperature of urban areas, compared to the one measured in the extra-urban areas. The analysis of the heat island needs detailed local climate data which can be collected only by a dedicated weather monitoring system. The Department of Energy and Environmental Researches of the University of Palermo (Italy) has built up a weather monitoring system that works 24 hours per day and makes data available in real-time at the web site: www.dream.unipa.it/meteo. The data collected by the system have been used to implement a set of nonlinear black-box models aiming to obtain short-term forecasts of the air temperature and map them over the monitored area. By using the data recorded during the 2008 summer, the daily profiles of the hourly average temperature have been plotted for each weather station of the monitoring system, thus clearly highlighting the temperature differences between the urban and extra-urban area and the average intensity of the UHI of Palermo.