Relationship between annual mean temperature and degree-days

Abstract

Degree-days are a versatile climatic indicator and used for many applications in the design and operation of energy efficient buildings – from the estimation of energy consumption and carbon emissions due to space heating and cooling to the energy and environmental monitoring of buildings. This research is aimed at developing an equation for calculating degree-days from low-resolution temperature data by exploring the relationship between degree-days and annual mean temperature of 5511 locations around the world, using multiple non-linear regression. Results suggest a very strong relationship between annual mean temperature and degree-days. Incorporating standard deviation (SD) of monthly mean temperature and latitude increases the accuracy of prediction (R2>.99), demonstrating the strength of the location-agnostic relationship in predicting degree-days from two temperature parameters: annual mean and SD of monthly mean. Research findings can be used to calculate degree-days of locations, for which daily temperature data may not be available. The equation can also be used to calculate degree-days from low-resolution global circulation model (GCM) projections of increasing temperature, for investigating the impact of climate change on building heating and cooling energy demand at global scale without the need to create synthetic weather series through morphing or downscaling.