Article Per Year (5 Year)
p-Index From 2021 - 2026
0.778
P-Index
This Author published in this journals
All Journal MANAJEMEN HUTAN TROPIKA Journal of Tropical Forest Management Jurnal Sylva Lestari Indenesian Journal of Tropical Biology Indenesian Journal of Tropical Biology
Purnama, Miftahul Irsyadi
Unknown Affiliation
Agriculture, Biological Sciences & Forestry Biochemistry, Genetics & Molecular Biology Chemical Engineering, Chemistry & Bioengineering Earth & Planetary Sciences Energy Environmental Science Health Professions Immunology & microbiology Materials Science & Nanotechnology Medicine & Pharmacology Social Sciences

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Integrating Multi-Variable Driving Factors to Improve Land Use & Land Cover Classification Accuracy using Machine Learning Approaches: A Case Study from Lombok Island Purnama, Miftahul Irsyadi; Çoban, Hüseyin Oğuz
Jurnal Manajemen Hutan Tropika Vol. 31 No. 2 (2025)
Publisher : Institut Pertanian Bogor (IPB University)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.7226/jtfm.31.2.123

Abstract

Accurate classification of land cover is essential for effective land management and environmental monitoring. This study aimed to enhance land cover classification for Lombok Island using advanced machine learning algorithms. The models employed include Random Forest, Gradient Boosting, Decision Tree, and Naive Bayes, integrating a wide range of variables, such as Landsat satellite imagery, spectral indices, physiographic, climatic, and socio-economic data. Among these, Random Forest demonstrated the highest model accuracy at 82%, followed by Gradient Boosting at 80%, Decision Tree at 73%, and Naïve Bayes at 61%. In field validation assessments, comparing the predictions of these machine learning models with ground truth data, Random Forest was the most reliable, achieving an overall accuracy of 88%. This superior performance is largely due to the multi-variable approach, which allows the model to mitigate issues like cloud cover in satellite images. The key variables that significantly influenced the land cover classification on Lombok Island include proximity to settlements, temperature, and distance to roads. These results provide essential insights for land management strategies, enabling policymakers and stakeholders to make informed decisions on sustainable development, urban planning, and environmental conservation in rapidly changing landscapes.
Three Decades of Forest Biomass Estimation in Southeast Asia: A Systematic Review of Field, Remote Sensing, and Machine Learning Approaches (1995–2025) Latifah, Sitti; Gandaseca, Seca; Afifi, Mansur; Prasetyo, Andrie Ridzki; Purnama, Miftahul Irsyadi; Kertalam, Lalu Rizky Aji; Pratama, Roni Putra
Jurnal Sylva Lestari Vol. 13 No. 3 (2025): September
Publisher : Department of Forestry, Faculty of Agriculture, University of Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/jsl.v13i3.1162

Abstract

Aboveground biomass plays a pivotal role in estimating tropical forest carbon stocks, particularly in Southeast Asia, a region rich in biodiversity but threatened by deforestation and land-use change. This systematic review analyzes 71 peer-reviewed studies published between 1995 and 2025, selected from an initial pool of 8,509 articles. The review aims to evaluate methodological developments and performance across three major approaches: field-based and allometric models, remote sensing including Unmanned Aerial Vehicle (UAV) platforms, and Machine Learning (ML) with data fusion, within key tropical forest countries: Indonesia, Malaysia, and Vietnam. These countries were selected due to their high forest cover, rapid land-use change, and central roles in the implementation of Reducing Emissions from Deforestation and Forest Degradation (REDD+). Field-based models, particularly those calibrated locally, consistently produced high accuracy, with R² values generally ranging from 0.80 to 0.96. Remote sensing techniques, particularly the integration of airborne LiDAR and optical–SAR, demonstrated strong predictive performance (R² > 0.85) and relatively low Root Mean Square Error (RMSE), typically below 30 Mg/ha. ML approaches such as Random Forest, Support Vector Machines, and LightGBM also achieved competitive results, with R² typically between 0.75 and 0.85 and RMSE below 40 Mg/ha when trained on high-quality input data. Mangrove and dipterocarp forests emerged as the most frequently studied ecosystems. While methodological innovations are evident, notable gaps remain in model harmonization and representation of ecosystem diversity. The review recommends integrating species-specific allometric models with remote sensing and machine learning pipelines, supported by open-access datasets, to enhance national forest monitoring systems and REDD+ readiness across Southeast Asia. Keywords: aboveground biomass, allometric, biomass estimation, carbon stock, South East Asia
Modelling Current and Future Mangrove Distribution under RCP 8.5 Climate Scenario: A Machine Learning Approach on Lombok Island, Indonesia Purnama, Miftahul Irsyadi; Azizah, Lutfia; Grendis, Nuraqila Waida Bintang; Zulkurniawan, Muhamad; ÇOBAN, Hüseyin Oğuz
Indonesian Journal of Tropical Biology Vol. 2 No. 1 (2026): April (In Progrees)
Publisher : Yayasan Siti Widhatul Faeha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.65622/ijtb.v2i1.235

Abstract

Mangrove ecosystems are essential for coastal protection and biodiversity, yet their distribution is highly influenced by climate variability. This study aims to predict current and future distribution of mangrove habitats on Lombok Island-Indonesia, using environmental predictor variables derived from topographic data and Köppen–Geiger climate classification. Mangrove distribution data were classified into presence and absence categories and integrated with climatic and terrain variables to develop habitat suitability models using five machine learning algorithms: Random Forest (RF), Decision Tree (DT), Naïve Bayes (NB), Artificial Neural Network (ANN), and Support Vector Machine (SVM). Model performance was evaluated using accuracy metrics, and the best-performing model was selected for spatial projection under the Representative Concentration Pathway (RCP) 8.5 scenario for 2050 and 2080. The RF model showed the highest predictive performance. The results indicate a substantial decline in suitable mangrove habitats, decreasing from 12,443 ha under current conditions to 7,255 ha in 2050 and 6,336 ha in 2080, representing a reduction of nearly 50%. This decline is associated with changes in precipitation and temperature regimes that influence hydrological conditions and habitat suitability. The application of machine learning provides a robust spatial approach for predicting mangrove distribution and supports conservation planning and climate-adaptive coastal management.
Modelling Current and Future Mangrove Distribution under RCP 8.5 Climate Scenario: A Machine Learning Approach on Lombok Island, Indonesia Purnama, Miftahul Irsyadi; Azizah, Lutfia; Grendis, Nuraqila Waida Bintang; Zulkurniawan, Muhamad; ÇOBAN, Hüseyin Oğuz
Indonesian Journal of Tropical Biology Vol. 2 No. 1 (2026): April (In Progrees)
Publisher : Yayasan Siti Widhatul Faeha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.65622/ijtb.v2i1.235

Abstract

Mangrove ecosystems are essential for coastal protection and biodiversity, yet their distribution is highly influenced by climate variability. This study aims to predict current and future distribution of mangrove habitats on Lombok Island-Indonesia, using environmental predictor variables derived from topographic data and Köppen–Geiger climate classification. Mangrove distribution data were classified into presence and absence categories and integrated with climatic and terrain variables to develop habitat suitability models using five machine learning algorithms: Random Forest (RF), Decision Tree (DT), Naïve Bayes (NB), Artificial Neural Network (ANN), and Support Vector Machine (SVM). Model performance was evaluated using accuracy metrics, and the best-performing model was selected for spatial projection under the Representative Concentration Pathway (RCP) 8.5 scenario for 2050 and 2080. The RF model showed the highest predictive performance. The results indicate a substantial decline in suitable mangrove habitats, decreasing from 12,443 ha under current conditions to 7,255 ha in 2050 and 6,336 ha in 2080, representing a reduction of nearly 50%. This decline is associated with changes in precipitation and temperature regimes that influence hydrological conditions and habitat suitability. The application of machine learning provides a robust spatial approach for predicting mangrove distribution and supports conservation planning and climate-adaptive coastal management.
Co-Authors Azizah, Lutfia Çoban, Hüseyin Oğuz Grendis, Nuraqila Waida Bintang Kertalam, Lalu Rizky Aji Muhammad Afif Prasetyo, Andrie Ridzki Pratama, Roni Putra Seca Gandaseca Sitti Latifah Zulkurniawan, Muhamad