Variations in heating and cooling degree‐days in the south‐eastern USA, 1960–1989

Temporal trends and spatial variability of heating and cooling degree‐days are examined for a portion of the south‐eastern USA for the period 1960–1989. Temporal variability is examined through simple time‐series analyses, regression analyses, and a climate departure index (CDI). Maps of 30‐, 15‐, 10‐, and 5‐year climatic normals for degree‐days are presented. The CDI indicates that heating degree‐day (HDD) variability was high during the 1970s with a sudden drop to low variability during the 1980s. Simple regressions show weak downward trends for actual HDD values across the region during the study period. However, the HDD decline was more pronounced through the 1980s, coinciding with the period of low year‐to‐year variability. Isoline maps show that most sites recorded distinctly fewer HDD in the most recent 5‐year period compared with longer term averages. Although the long‐term trend for cooling degree‐days (CDD) is slightly positive, there is considerable interyearly and interdecadal variability. Unlike HDD, there were no sustained periods with consistently high or low variability for CDD. Mapped patterns of climatic normals show that CDD values peaked across the study area during the most recent 5‐year period. Mapped patterns for both HDD and CDD can be influenced by data for large urban sites. Repeating the analyses without the seven largest urban centres resulted in less anomalous spatial patterns but almost no shift in study area mean‐degree‐day or CDI values and their temporal patterns. Atlanta emerged as the most distinct anomaly for the mapped patterns. From an applied viewpoint, the most significant finding occurred for the heating season with the combination of a sustained downward trend in HDD and comparatively low variability around the long mean (low CDI values) during the last 13 years of the study period. This period represents a span where the climate‐driven demand for space heating was decreasing and the heating‐season generating capacities for power generation utilities were not challenged.