Abstract

Introduction: Building performance assessments frequently rely on historical weather files that do not represent future climatic conditions, introducing uncertainty and risk for long-lived buildings and low-carbon design decisions. This paper presents a transparent and replicable method to generate future EnergyPlus Weather (EPW) files by morphing a baseline EPW file, derived from typical meteorological year (TMY) datasets representative of recent historical climate conditions, using monthly climate anomalies derived from a Global Climate Model (GCM).

Materials and methods: The proposed workflow integrates an automated routine that extracts climate anomalies for any geographic location through inverse-distance weighting of the four nearest GCM grid points, followed by variable-specific shift and stretch transformations and psychrometric post-processing to preserve internal consistency among EPW parameters. The method is demonstrated for two climatically and geographically contrasting locations—Exeter (UK) and Bahía Blanca (Argentina)—using a 2080 time horizon (multi-decadal average centred on the 2080s), allowing evaluation across different hemispheres and climate regimes.

Results: Generated time series are benchmarked against available future weather datasets for the selected regions (PROMETHEUS, Meteonorm, and Future Weather Generator). For dry-bulb temperature, the proposed approach shows strong agreement in temporal behaviour with reference data, with Pearson correlation coefficients ranging from 0.744 to 0.799 (p < 0.05). For Exeter, the method reproduces the expected warming signal, with annual maximum dry-bulb temperatures increasing from approximately 27.0 °C to 31.6 °C and minimum dry-bulb temperatures shifting from −4.4 °C to −2.5 °C relative to the baseline.

Conclusions: The results demonstrate that morphing coupled with globally available climate projections can provide practical, location-agnostic future EPW files suitable for early-stage design and sensitivity analyses, while highlighting the importance of baseline weather data quality and climate-model resolution.