Analysis of Summer monsoon rainfall: Sustainability of long-term modelling by machine learning methods
MAUSAM · 28.03.2026 · 77 (2) · 485-502
DOI: 10.54302/mausam.v77i2.6951 ↗Kurz gefasst
Fachartikel in der Zeitschrift MAUSAM, online erschienen am 28.03.2026. Verfasst von Namita Goyal und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Indian monsoon is a distinct meteorological phenomenon owing to its importance in the socio-economical context in the Indian subcontinent. Analysis of available Indian and regional data sets spanning over hundred years gives critical insights into various aspects of rainfall. Two major characteristics namely the trend and break points have been analyzed for Indian summer monsoon rainfall (ISMR) and Kerala summer monsoon rainfall (KSMR). Kerala is selected because of it being the point of onset of monsoon in India. The analysis of trend is investigated with a null hypothesis of there being no trend at the 95% confidence level for both the ISMR and KSMR. Prediction of seasonal rain is vital for overall planning and governance. Machine learning models are statistical models that are proven tools for accurate prediction of meteorological phenomena. However, it is yet to be investigated if the predictions by such models stand the test of time. In the present study we have developed machine learning models to investigate if the predictions reflect the long-term trend and if the predicted seasonal time series of KSMR reflect the same parametric values in the intervals of observed break points. The positions of break points in the observed and predicted time series are analyzed to get a fair estimate of model’s capability to detect the break points independent of training on specific break points. Autocorrelation analysis of the seasonal monsoon is done to find the best predictors. The results are encouraging and demonstrate that trends, break points and the parametric values may be sustained for several decades within statistical acceptability.
Zitieren
APA
Goyal, N., Mahajan, A., & Tripathi, K. C. (2026). Analysis of Summer monsoon rainfall: Sustainability of long-term modelling by machine learning methods. MAUSAM, 77(2), 485-502. https://doi.org/10.54302/mausam.v77i2.6951
Vancouver
Goyal N, Mahajan A, Tripathi KC. Analysis of Summer monsoon rainfall: Sustainability of long-term modelling by machine learning methods. MAUSAM. 2026;77(2):485-502. doi:10.54302/mausam.v77i2.6951
BibTeX
@article{Goyal2026,
author = {Goyal, Namita and Mahajan, Aparna and Tripathi, Krishna Chandra},
title = {Analysis of Summer monsoon rainfall: Sustainability of long-term modelling by machine learning methods},
journal = {MAUSAM},
year = {2026},
volume = {77},
number = {2},
pages = {485--502},
doi = {10.54302/mausam.v77i2.6951}
}
EUVIMED