Impact of global warming on the sensible heat load in a detached house in Tokyo in the 2030s

1. INTRODUCTION

 Global warming and urban heat islands may promote an increase in energy consumption for air-conditioning of buildings in summer periods. Recently, designers of buildings have often used energy simulations in their design process to estimate energy consumption. In these simulations, regional climate and weather data are commonly used that are based on current or past conditions. However, most buildings are used for several decades, over which the climate will gradually change. Therefore, it is more appropriate to use weather data representative of the near future in which the designed buildings will actually be used. Our research group is now developing near-future weather data using dynamic downscaling from a global climate model (GCM) to a regional climate model (RCM). In this study, we assessed the impact of climate change on the sensible heat loads in a detached house in Tokyo in summer in the 2030s using a building energy simulation with weather data obtained from the downscaling simulation.

2. METHODS

 Initially, we analyzed climate change in the Kanto region of Japan from the present to the 2030s using a GCM and an RCM. Although the GCM can predict long-term global warming, the spatial resolution of the data is too coarse to use as weather data for the energy simulation. Therefore, we employed a dynamic downscaling method in which the GCM data is used in the RCM as initial and boundary conditions and the GCM data are downscaled based on models of physical processes (Yuqing et al., 2004). In this study, we employed the Model for Interdisciplinary Research on Climate (MIROC; Nozawa et al., 2007) as the GCM and Weather Research and Forecasting (WRF) as the RCM (Lo et al., 2008).

 The target area for this study was the Kanto region, including Tokyo and surrounding areas. In the WRF simulation, we conducted nested regional climate modeling at four levels. The first and fourth levels had horizontal spatial resolutions of 54 km and 1 km, respectively. The MIROC data were used as boundary conditions for the first-level domain in the WRF simulations. We obtained weather data (temperature, humidity, solar radiation, wind speed, etc.) for August over 5 yr both for the present (2006-2010) and the 2030s (2031-2035) based on the simulated results of the fourth-level domain.

 We then analyzed the impact of climate change on the energy consumption of a detached house using building energy simulation (Urano, 2009). We employed a standard house model located in Tokyo as the target building. Figure 1 illustrates plans for a typical house used for environmental studies of architecture in Japan. The software TRNSYS (University of Wisconsin, Madison, WI, USA) was used for the energy simulations. We analyzed the energy consumption of the house shown in Fig. 1 in August in the present and in the 2030s using weather data acquired from the climate simulation. We assumed a cooling system with units in a living and dining room with a kitchen, a bedroom, and a child's room (A); these rooms had areas of 29.8, 13.3, and 10.8 m², respectively. The cooling equipment was assumed to be activated when the air temperature in these rooms rose above 26 °C.

3. RESULTS

 Figure 2 shows the frequency distributions for temperature at a height of 2 m in Tokyo (at the Otemachi weather observation station) in August of the present and the 2030s. The frequency of high temperatures was predicted to slightly increase in the 2030s compared with the present. Although the distribution varies each year, the mean temperature over 5 yr was predicted to increase by 1.12 °C from the present to the 2030s (Table 1).

 Table 1 also summarizes the average sensible heat loads for cooling of the house. The sensible heat load is the energy that needs to be removed from the house by the cooling system to maintain the desired air temperature. Because the mean outdoor temperature and solar radiation for a given month varies each year, the sensible heat loads predicted for August in some years in the 2030s were lower than those in the present. However, based on the 5-yr mean, the sensible heat load for August in the 2030s was predicted to be 10% higher than that for the present (2.56x10³ and 2.32x10³ MJ/mon, respectively).

4. CONCLUSIONS

 We assessed the impact of climate change on the energy consumption of a detached house in Tokyo, Japan in the 2030s. We carried out climate simulations using the downscaling technique with global and regional climate models to derive weather data for August for 5 yr in the present (2006-2010) and in the 2030s (2031-2035). Using this weather data, we conducted energy simulations of a detached house in Tokyo to evaluate the sensible heat loads for cooling. Based on a 5-yr mean in August, the outdoor temperature was predicted to increase by 1.12 °C (from 26.24 to 27.36 °C) from the present to the 2030s. As a result, the sensible heat load for the house was predicted to increase by 10% under the study conditions.