ANALYSIS OF MANGROVE HEALTH PREDICTIONS USING THE RANDOM FOREST REGRESSION METHOD
DOI:
https://doi.org/10.53840/ejpi.v12i5.323Keywords:
Mangrove Health Index; Mangrove Health; Random Forest RegressionAbstract
Mangrove forests are important ecosystems in coastal areas that play a role in maintaining environmental balance and protecting the coast from erosion and abrasion. However, these ecosystems have experienced a decline in quality and quantity due to human activities, such as land conversion for industry and plantations. This study aims to predict the health of mangroves in the Belawan Sicanang area, Medan City, North Sumatra, using the Random Forest Regression (RFR) method. The data used include multispectral images with NDVI, NDRE, and ARVI vegetation indices, as well as Mangrove Health Index (MHI) values. The analysis process was carried out through several stages, namely processing drone image data, oversampling using the SMOTE method, and applying RFR models to predict mangrove health. The prediction results showed that 68.9% of the area was classified as healthy, 22.3% moderate, and 8.8% unhealthy, with good model accuracy indicated by an R-squared (R²) value of 0.757 and a Root Mean Squared Error (RMSE) of 8.515. This study shows that the Random Forest Regression method is effective in predicting and mapping mangrove health conditions with a high level of accuracy. The recommendation of this study is to increase the sample during field data collection so that the research results can be more accurate and good.
Downloads
References
Aspan, H., Rian Putra, R., Sri Wahyuni, E., Pembangunan Panca Budi, U., & Batam, U. (2023). Socialization of the development of Belawan Sicanang Mangrove Forest Tourism by implementing the Dala Information System. ARSY :Research Application to the Community, 4(1), 2023–2024. Retrieved from http://journal.al-matani.com/index.php/arsy,Online
Doni Nurdiansah, & I Wayan Eka Dharmawan. (2021). Structure and Health Conditions of Mangrove Communities on Middleburg-Miossu Island, West Papua. Journal of Tropical Marine Science and Technology, 13(1), 81–96. https://doi.org/10.29244/jitkt.v13i1.34484
Ishak Ariawan1*, A. A. R., Anzani1, L., Minsaris1, L. O. A., & Lukman1. (2021). Identification of Mangrove Species Using the Random Forest Algorithm. Maritime Journal: Indonesian Journal of Maritime, 2(2), 87–96.
Liu, Y., Liu, S., Li, J., Guo, X., Wang, S., & Lu, J. (2019). Estimating biomass of winter oilseed rape using vegetation indices and texture metrics derived from UAV multispectral images. Computers and Electronics in Agriculture, 166(July), 105026. https://doi.org/10.1016/j.compag.2019.105026
Minh, P., Id, H., Tinh, P. H., Son, N. P., & Thuy, T. Van. (2022). Mangrove Health Assessment Using Spatial Metrics And Multi-temporal Remote Sensing Data. Plos One, 17, 1–20. https://doi.org/10.1371/journal.pone.0275928
Özen, F. (2024). Random forest regression for prediction of Covid-19 daily cases and deaths in Turkey. Heliyon, 10(4), 1–19. https://doi.org/10.1016/j.heliyon.2024.e25746
Singgalen, Y. A. (2023). Analysis of Potential Mangrove Ecotourism Area Development Model Based on Hyper Spectral of Remote Sensing and Analytical Hierarchy Process. Journal of Information System Research (JOSH), 4(3), 969–979. https://doi.org/10.47065/josh.v4i3.3385
Suyadi, S., Naroli, I., & Stephen, A. B. (2021). Characteristics of Mangrove Vegetation and Its Utilization: A Case Study in East Seram, Maluku. Journal of Marine and Applied Fisheries (JKPT), 4(1), 37. https://doi.org/10.15578/jkpt.v4i1.9502
Suyuti, H. (2019). Analysis of Mangrove Forest Health Using Hemispherical Photography Method in the Coastal Area of Central Tapanuli Regency, North Sumatra Province. Student Online Journal, 5–10.
Tridawati, A., Armijon, A., Yanto, F., & Novianti, T. C. (2023). Mapping Mangrove Forest Distribution Using Machine Learning Algorithms in the Petengoran Mangrove Forest Area. Journal of Incentive Technology, 17(2), 84–98. https://doi.org/10.36787/jti.v17i2.1101
Zhu, Y., Liu, K., Liu, L., Myint, S. W., Wang, S., Liu, H., & He, Z. (2017). Exploring the potential of world view-2 red-edge band-based vegetation indices for estimation of mangrove leaf area index with machine learning algorithms. Remote Sensing, 9(10). https://doi.org/10.3390/rs9101060
Downloads
Published
Issue
Section
License
Copyright (c) 2025 e-Jurnal Penyelidikan dan Inovasi
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
