Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Eduvest - Journal of Universal Studies Jurnal Penginderaan Jauh dan Pengolahan Data Citra Digital
Parwati Sofan, Parwati
Unknown Affiliation
Aerospace Engineering Agriculture, Biological Sciences & Forestry Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Health Professions Neuroscience Social Sciences

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Optimizing Machine Learning for Daily Rainfall Prediction in Bogor: A Statistical Downscaling Approach Intan Arassah, Fradha; Sadik, Kusman; Sartono, Bagus; Sofan, Parwati
Eduvest - Journal of Universal Studies Vol. 5 No. 6 (2025): Eduvest - Journal of Universal Studies
Publisher : Green Publisher Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59188/eduvest.v5i6.51307

Abstract

This study explores the use of machine learning models as a statistical downscaling technique to predict daily rainfall in Bogor, Indonesia. The general circulation model (GCM) is a leading tool for climate prediction, and this research applied a two-stage machine learning model to improve its predictions. The main objectives were to evaluate different GCM domains and handle missing data using two imputation approaches. The first stage involved constructing datasets with varying methods for addressing missing values, followed by the application of a support vector classification (SVC) model to classify rainy and non-rainy days. In the second stage, a recurrent neural network (RNN) model was developed to predict daily rainfall amounts. The results revealed that using random forest imputation for missing data enhanced model accuracy and reduced the root mean square error (RMSE). Among the different GCM domains, the 5 km resolution GCM data was the most accurate when compared to local station climatology. The SVC model, using a radial basis function kernel, achieved an impressive classification accuracy of 98.5%, while the RNN model achieved an RMSE of 16.19. These findings are valuable for improving rainfall predictions and can provide effective data-driven recommendations for disaster mitigation efforts in the region.
MONITORING OF DROUGHT-VULNERABLE AREA IN JAVA ISLAND, INDONESIA USING SATELLITE REMOTE-SENSING DATA Roswintiarti, Orbita; Sofan, Parwati; Anggraini, Nanin
Jurnal Penginderaan Jauh dan Pengolahan Data Citra Digital Vol. 8 No. 1 (2011)
Publisher : Institut Teknologi Sepuluh Nopember

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30536/inderaja.v8i1.3248

Abstract

The impact of climatic variability and climate change is of great importance in Indonesia. Monitoring this impact, furthermore, is essential to the preparedness of the regions in dealing with drought-vulnerable conditions. In this study, satellite remote sensing data were used for monitoring drought in Java island, Indonesia. Monthly rainfall data from Tropical Rainfall Measuring Mission (TRMM) data were used to derive the Standardized Precipitation Index (SPI). The Moderate Resolution Imaging Spectroradiometer (MODIS) onboard the Terra and Aqua satellites was used for calculating the Enhanced Vegetative Index (EVI) and Land Surface Temperature (LST). EVI and LST were then converted to the Vegetation Condition Index (VCI) and the Temperature Condition Index (TCI), which are useful indices for the estimation of vegetation moisture and thermal conditions, respectively. Vegetation Health Index (VHI) was calculated using the VCI and TCI to represent the overall vegetation health. The analysis was carried out during the El Niño/Southern Oscillation (ENSO) of June to August 2009. From the SPI analysis, it is found that since June 2009 the conditions of mild drought (-1.0 < SPI < 0) have developed in almost all parts of Java island due to rainfall deficiency. The VCI maps show that the vegetative stress (VCI < 36) as a result of the vegetation moisture condition has gradually developed in the East Java province in June 2009. Meanwhile, from the TCI maps it is found that the vegetative stress (TCI < 36) due to the thermal condition of vegetation was built up in the West Java province in June 2009. Hence, the overall vegetative health in Java island obtained from the VHI maps shows that the moderate vegetative drought (VHI < 36) started to develop in July 2009.
Co-Authors Bagus Sartono Intan Arassah, Fradha Kusman Sadik Nanin Anggraini, Nanin Orbita Roswintiarti, Orbita