Author(s): Mengzhu Chen; Hongjuan Han; Simone Fatichi

Linked Author(s): Mengzhu Chen

Keywords: Climate change stochastic downscaling uncertainty partition Singapore

Abstract: Climate change projections at fine temporal resolution and local spatial scales are crucial for impact models and studies. Downscaling of climate projections not only allows obtaining local projections of changes in climate variables but also can offer valuable insights into different sources of uncertainty. In this study, we used a stochastic downscaling methodology that combines an hourly weather generator (AWE-GEN) and a Bayesian framework to produce an ensemble of hourly meteorological time series, representing possible future climate scenarios for Singapore. This approach enables the quantification of three main sources of uncertainty in climate change projections: climate model, scenario, and internal climate variability uncertainty, which can provide information for decision makers and consultants to prepare adaptation strategies and adjust design criteria. Furthermore, the climate change scenarios generated by the stochastic downscaling methodology in this study were compared with the scenarios obtained in the Singapore's Third National Climate Change Study using the SINGV-RCM dynamical downscaling approach.

DOI:

Year: 2025