The Impact of Temperature Variation on Urban Residential Electricity Consumption

Under the influence of climate change, both the frequency and duration of extreme heat events have been increasing. Previous energy consumption studies have consistently identified a relationship between outdoor air temperature and residential energy use. In particular, under high-temperature conditions, increased demand for air conditioning often leads to a rise in household electricity consumption. Therefore, this study focuses on Taipei City and Tainan City, examining the impact of temperature on residential electricity consumption at the village scale, with the aim of capturing the characteristics of energy use variation under high-temperature conditions.

This study integrates the 2-km gridded historical climate reconstruction dataset (TReAD) provided by the TCCIP program of the National Science and Technology Center for Disaster Reduction, along with village-level monthly residential electricity consumption data published by Taiwan Power Company. A GIS-based area-weighting method was first applied to harmonize spatial scales. Considering the characteristics of residential electricity data in Taiwan, data calibration and grouped regression analyses were then conducted. A threshold of 23°C was adopted to represent the onset of cooling demand, allowing comparison between low- and high-temperature groups.

The results indicate that, due to Taiwan’s relatively mild winter climate and limited heating demand, electricity consumption is less sensitive to temperature variations in the low-temperature range. For example, in Tainan City, within the low-temperature group (<23°C), temperature explains only about 5.08% of the variance in residential electricity consumption (R² = 0.0508), with an estimated elasticity of approximately 0.8–1.2% increase in electricity use per 1°C rise. However, once temperatures exceed the cooling threshold, electricity consumption becomes highly responsive to temperature changes. In the high-temperature group (>23°C), a strong positive correlation is observed (R² = 0.932), indicating that temperature accounts for over 93% of the variation in electricity consumption.

Further estimation using elasticity analysis suggests that, within the high-temperature range, a 1°C increase in temperature leads to an approximate 7.57% rise in monthly electricity consumption per household. In practical terms, assuming an average monthly household electricity consumption of 400 kWh during summer, this corresponds to an additional 30–35 kWh per month for each 1°C increase.

These findings highlight that high-temperature environments not only elevate thermal discomfort for residents but also significantly increase energy demand. This, in turn, may impose additional financial burdens on households—particularly for vulnerable groups—thereby exacerbating issues of climate inequality.