Estimation of Mean Recurrence and Residence Times of Drought Events Using an Empirical Bayes Approach for Markov Chain Parameters
Journal of Applied Mathematics · 2026-01 · 2026 (1)
DOI: 10.1155/jama/4032906 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Applied Mathematics, online erschienen am 17.11.2026. Verfasst von Wahidah Sanusi und 3 weiteren Autor:innen. Die Arbeit nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Drought is a prolonged lack of rainfall that can be hazardous to humans and other living organisms. It is difficult to detect the beginning and end of drought events, particularly extreme droughts, because it is a slow‐onset phenomenon without clearly defined thresholds. Therefore, this study employs the Standardized Precipitation Index (SPI) to classify drought categories, focusing on the 1‐month SPI (SPI‐1) timescale. In addition, the Markov chain model was applied to capture the characteristics of SPI‐1 drought events, particularly their mean recurrence and residence times. Transition probabilities, as the parameters of the Markov chain model, are estimated using the empirical Bayes approach. The analysis is based on monthly rainfall data (in mm) collected from 57 rain gauge stations across South Sulawesi Province during the period 1988–2023. The data were obtained from the Water Resources, Human Settlements, Spatial Planning and Development Office of South Sulawesi province, as well as from the Meteorology, Climatology and Geophysics Agency (BMKG) of Makassar. The results indicate that a first‐order homogeneous Markov chain provides an adequate fit to the data under the applied tests. The findings revealed that the Pambasian area in North Luwu Regency experiences severe dry conditions more frequently than other parts of South Sulawesi. In contrast, the Bengo area in Bone Regency is more prone to extreme dry conditions. Overall, South Sulawesi tends to experience predominantly non‐dry (wet) conditions throughout the year. However, certain regions have been affected by severe and extreme drought events. The study shows that the province of South Sulawesi experiences non‐dry conditions with mean recurrence time (return periods) of approximately 1.5 to 2 months, whereas severe dry conditions have return periods of about 10 to 25 months, and extremely dry conditions occur every 50 to 100 months as typical regional values. However, localized areas exhibit longer return periods, exceeding 100 months, reflecting spatial variability across the province. Wet periods generally persist for 2–3 months, whereas dry periods typically last 1–2 months. The findings of this study are expected to provide valuable insights for decision‐makers in formulating strategies to mitigate the impacts of extreme drought events.
Zitieren
APA
Sanusi, W., Sidjara, S., Patahuddin, S., & Danial, M. (2026). Estimation of Mean Recurrence and Residence Times of Drought Events Using an Empirical Bayes Approach for Markov Chain Parameters. Journal of Applied Mathematics, 2026(1). https://doi.org/10.1155/jama/4032906
Vancouver
Sanusi W, Sidjara S, Patahuddin S, Danial M. Estimation of Mean Recurrence and Residence Times of Drought Events Using an Empirical Bayes Approach for Markov Chain Parameters. Journal of Applied Mathematics. 2026;2026(1). doi:10.1155/jama/4032906
BibTeX
@article{Sanusi2026,
author = {Sanusi, Wahidah and Sidjara, Sahlan and Patahuddin, Sudarmin and Danial, Muhammad},
title = {Estimation of Mean Recurrence and Residence Times of Drought Events Using an Empirical Bayes Approach for Markov Chain Parameters},
journal = {Journal of Applied Mathematics},
year = {2026},
volume = {2026},
number = {1},
doi = {10.1155/jama/4032906}
}
EUVIMED