Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Scientific Nexus
Khanifah Afifi
Atmospheric Science and Earth System Research Group, Faculty of Science, Institut Teknologi Sumatera, Indonesia
Astronomy Chemistry Earth & Planetary Sciences Materials Science & Nanotechnology Medicine & Pharmacology

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Spatio-temporal pattern analysis and prediction of extreme rainfall frequency in the Lampung region using random forest Ary Ramlan Simangunsong; Wirid Birastri; Ridlo Wahyudi Wibowo; Khanifah Afifi; Ririn Andriyani; Nindhita Pratiwi
Scientific Nexus Vol. 1 No. 2 (2025): Scientific Nexus
Publisher : Fakultas Sains Institut Teknologi Sumatera

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35472/scinexus.2533

Abstract

This study aims to analyze the spatio-temporal patterns and predict the frequency of extreme rainfall (FE) during the December-January-February-March (DJFM) season in the Lampung region. The frequency of extreme events (FE) was determined using the 90th percentile threshold via the Cumulative Distribution Function (CDF). Principal Component Analysis (PCA) was applied to extract the spatio-temporal principal components (PCs) of extreme events. These PCs were then correlated with large-scale atmospheric parameters, including Sea Surface Temperature (SST), zonal and meridional winds at 850 hPa (U850/V850), and the ONI and DMI climate indices. Areas with high correlations were extracted as predictors for the Random Forest model. The model was trained and tested over the period from January 1982 to December 2020 and evaluated using Root Mean Square Error and Pearson correlation (r). Wind patterns and SST showed a significant relationship with FE, while ONI and DMI were weakly correlated. Evaluation of the RF model indicated a tendency for the model to overestimate low FE and underestimate high FE, although the model was able to represent the general patterns of FE. This approach demonstrates potential for developing extreme rainfall frequency predictions based on large-scale atmospheric data, which can contribute to hydrometeorological disaster mitigation in Lampung.
Co-Authors Ary Ramlan Simangunsong Nindhita Pratiwi Ridlo Wahyudi Wibowo Ririn Andriyani Wirid Birastri