Rampersad et al. (2026) A Bayesian workflow for multi-station IDF curve development in data-limited island settings: application to Grenada

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Theoretical and Applied Climatology
• Year: 2026
• Date: 2026-09-12
• Authors: Aaron Jerome Rampersad, Christianne Marie-Claire Faith Zakour
• DOI: 10.1007/s00704-026-06558-4

Research Groups

• National Water Resources Laboratory, University of the West Indies (Mona Campus)
• Department of Geography and Geology, University of the West Indies (Cave Hill Campus)

Short Summary

This study develops a Bayesian workflow for multi-station intensity-duration-frequency (IDF) curves and spatially distributed daily rainfall extremes in Grenada, addressing data limitations and uncertainty associated with short records. The framework provides a probabilistic basis for rainfall extreme estimation in similar island settings.

Objective

• Investigate the feasibility of developing site-specific multi-station IDF curves in Grenada using Bayesian inference and machine learning techniques.

Study Configuration

• Spatial Scale: Island-wide, focusing on Grenada's topographic regions.
• Temporal Scale: Daily rainfall extremes over a 10-year period (2005-2014).

Methodology and Data

• Models used:
  • Ordinary kriging with non-stationary spatial correlation models (SCMs)
  • Bayesian generalized extreme value (GEV) formulation
  • Bayesian generalized Pareto (GPD) formulation
• Data sources: Daily rainfall observations from 10 stations across Grenada, supplemented by satellite and reanalysis data.

Main Results

• The study reveals that daily rainfall dependence is primarily controlled by separation distance, with elevation exerting a secondary influence moderated by temporal non-stationarity.
• Ordinary kriging using Bayesian-inferred spatial correlation models outperformed deterministic, geostatistical, and machine-learning-based alternatives in imputing missing daily rainfall values.

Contributions

This study contributes to the development of transferable probabilistic frameworks for rainfall extreme estimation in data-limited island settings, addressing the limitations of single-station IDF curves and conventional semivariogram-based approaches.

Funding

• This research was supported by the Caribbean Community Climate Change Centre (CCCCC) under the project "Climate Resilience and Adaptation in Small Island Developing States" (CRASIDS).
• Additional funding was provided by the University of the West Indies Research and Publications Fund.

Citation

@article{Rampersad2026Bayesian,
  author = {Rampersad, Aaron Jerome and Zakour, Christianne Marie-Claire Faith},
  title = {A Bayesian workflow for multi-station IDF curve development in data-limited island settings: application to Grenada},
  journal = {Theoretical and Applied Climatology},
  year = {2026},
  doi = {10.1007/s00704-026-06558-4},
  url = {https://doi.org/10.1007/s00704-026-06558-4}
}