Rising temperatures driven by anthropogenic climate change are expected to substantially alter patterns of energy consumption across Central Europe. This study investigates the historical relationship between surface temperature and electricity demand in the Kisalföld region of Hungary — a lowland agricultural area characterised by high summer temperature variability — and projects how this relationship evolves under IPCC AR6 warming scenarios of 1.5 ◦C, 2 ◦C, and 3 ◦C above pre-industrial levels by 2100. Using ERA5 reanalysis data spanning 1970–2024, upper-air sounding observations from WMO station 12843 (Budapest/Pestszentlőrinc), and Hungarian national grid load records from ENTSO-E (2015– 2024), we train an XGBoost demand model achieving R2 = 0.77 and RMSE = 285 MW on a held-out test set. ERA5 data reveal warming of +0.42 ◦C per decade since 1970, with cooling degree days tripling and heating degree days declining by 20%. Projections suggest that net climate-driven demand may initially fall as heating savings outpace cooling gains, but under the high-emissions scenario (SSP5-8.5) cooling demand overtakes heating savings around 2065, representing a structural transition from a winter-peak to a summer-peak energy system. Additionally, temperature-driven increases in grid transmission losses add up to 100 MW of continuously wasted generation capacity by 2100 under SSP5-8.5 — a compounding infrastructure burden that is largely absent from existing demand projections. An interactive Progressive Web Application accompanying this paper makes all projections publicly accessible.
Zsombor Sztyehlik (Sat,) studied this question.