Bayesian Stochastic Rainfall Generator (BSRG): Intelligent and Digital tool for Rainfall Forecasting through Bayesian Causal Modelling

Jose-Luis Molina (), Carmen Patino-Alonso () and Fernando Espejo ()
Additional contact information
Jose-Luis Molina: Salamanca University, High Polytechnic School of Engineering Avila
Carmen Patino-Alonso: Salamanca University
Fernando Espejo: Salamanca University, High Polytechnic School of Engineering Avila

Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA), 2025, vol. 39, issue 8, No 22, 4033-4050

Abstract: Abstract Rainfall is probably the most uncertain natural variable. This research is mainly aimed to design an intelligent data-driven tool called Bayesian Stochastic Rainfall Generator (BSRG) for improving the rainfall forecast. There are several previous stochastic generators for rainfall data generation. However, there is not any designed through a Bayesian stochastic approach. The utilization of rainfall data from the Muñogalindo rain gauge in Ávila, Spain, is discussed in this study, offering insights into the region’s Mediterranean continental climate. The innovative methodology known as Hyetoclust, serves as the foundation for the subsequent development of the BSRG. Findings reveal that cluster-specific characteristics significantly influence rainfall patterns, expressed as probabilistic distributions, under both normal and extreme regimes. In normal conditions, Cluster 3 exhibits the most sensitive behaviour of Rainfall Intensity to the optimization (Max and Min) of Event Duration compared to Clusters 1 and 2. This highlights the heterogeneous nature of rainfall patterns and emphasizes the necessity of considering cluster-specific traits in modelling and forecasting. Conversely, under extreme rainfall conditions, clusters exhibit varied responses. Clusters 1 and 3 tend to have similar effectively impacts under optimization (Max and Min) scenarios, while Cluster 2 displays a more complex behavior. In general terms, cluster 2 is the least sensitive to a minimization of time duration. This emphasizes the nuanced nature, expressed as probabilistic distributions, of hydrological responses and reinforces the importance of cluster-specific analysis in optimizing strategies during extreme regimes.

Keywords: Stochastic Rainfall Generator; Bayesian Causal Modelling; Rainfall Forecasting; Artificial Intelligence (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s11269-025-04143-4 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:waterr:v:39:y:2025:i:8:d:10.1007_s11269-025-04143-4

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/11269

DOI: 10.1007/s11269-025-04143-4

Access Statistics for this article

Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA) is currently edited by G. Tsakiris

More articles in Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA) from Springer, European Water Resources Association (EWRA)
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().