EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Entropy-Fractal Complexity Modelling for Sub-Seasonal Precipitation Forecasting: A Rolling-Window Machine Learning Framework

Vinita Sangwan, Surbhi Sharma, Rashmi Bhardwaj

Boletim da Sociedade Paranaense de Matemática · 03.07.2026 · 44 (7) · 1-19

DOI: 10.5269/bspm.82142 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Boletim da Sociedade Paranaense de Matemática, online erschienen am 03.07.2026. Verfasst von Vinita Sangwan und 2 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Atmospheric processes are nonlinear and multi-scaled variability of the atmosphere due to combined thermodynamic and hydrological processes. The traditional predictors of meteorology are not usually effective in capturing structural complexity in climate time series. This paper will use the joint entropy-fractal model to describe atmospheric variability and improve the forecasting of precipitation in ten Indian states based on reanalysis data that is available on the NASA POWER (Prediction of Worldwide Energy Resources) data access portal, which is managed by NASA. To measure stochasticity and the multi-scale complexity of time series with multiple parameters, the Shannon entropy and Higuchi fractal dimension were used to analyze the weekly multi-parameter meteorological observations. These complexity measures derived were combined into a regression model based on supervised machine learning to predict the precipitation in the coming month. Findings indicate that entropy fractal descriptors are very effective in predictive performance and produce $R^2=0.9099$ and $\mathrm{RMSE}=0.0857$ (normalized scale). The suggested methodology offers an integrated statistical-geometric description of atmospheric variability and forms complexity-based forecasting an avenue to enhanced sub-seasonal weather prediction.

Zitieren

APA

Sangwan, V., Sharma, S., & Bhardwaj, R. (2026). Entropy-Fractal Complexity Modelling for Sub-Seasonal Precipitation Forecasting: A Rolling-Window Machine Learning Framework. Boletim da Sociedade Paranaense de Matemática, 44(7), 1-19. https://doi.org/10.5269/bspm.82142

Vancouver

Sangwan V, Sharma S, Bhardwaj R. Entropy-Fractal Complexity Modelling for Sub-Seasonal Precipitation Forecasting: A Rolling-Window Machine Learning Framework. Boletim da Sociedade Paranaense de Matemática. 2026;44(7):1-19. doi:10.5269/bspm.82142

BibTeX
@article{Sangwan2026,
  author = {Sangwan, Vinita and Sharma, Surbhi and Bhardwaj, Rashmi},
  title = {Entropy-Fractal Complexity Modelling for Sub-Seasonal Precipitation Forecasting: A Rolling-Window Machine Learning Framework},
  journal = {Boletim da Sociedade Paranaense de Matemática},
  year = {2026},
  volume = {44},
  number = {7},
  pages = {1--19},
  doi = {10.5269/bspm.82142}
}
Kostenlos starten →