Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/165229
Title: Pemodelan Karbon Dioksida Negara di ASEAN Menggunakan Regresi Terboboti Geografis dan Temporal
Other Titles: Modelling Carbon Dioxide Emissions in ASEAN Countries Using Geographically and Temporally Weighted Regression
Authors: Djuraidah, Anik
Angraini, Yenni
Pradana, Alfa Nugraha
Naufal, Nabil
Issue Date: 2025
Publisher: IPB University
Abstract: Karbon dioksida (CO2) merupakan komponen utama gas rumah kaca yang paling berkontribusi terhadap pemanasan global akibat aktivitas manusia. Emisi CO2 di kawasan Asia Tenggara (ASEAN) menunjukkan tren peningkatan seiring dengan pertumbuhan ekonomi, urbanisasi, dan ketergantungan terhadap energi fosil. Keragaman kondisi sosial, ekonomi, dan lingkungan antar negara menyebabkan hubungan antara emisi CO2 dan faktor-faktor pemicunya tidak seragam secara geografis maupun temporal. Penelitian ini bertujuan untuk menganalisis dinamika spasial-temporal tersebut menggunakan metode Regresi Terboboti Geografis dan Temporal (RTGT). Data yang digunakan mencakup CO2 density (EDGAR), NDVI (MODIS), curah hujan (ERA5), indikator ekonomi & demografi (World Bank), serta energi (EMBER) untuk periode 2018–2023. Hasil analisis menunjukkan bahwa sebelum 2020, pertumbuhan ekonomi dan konsumsi listrik menjadi faktor utama peningkatan emisi. Namun setelah pandemi COVID-19, faktor lingkungan dan energi seperti NDVI dan ketergantungan bahan bakar fosil mulai berperan lebih besar di beberapa negara. Model RTGT terbaik menggunakan fungsi kernel bisquare dan lebar jendela adaptif, dengan nilai R² sebesar 0,99. Pergeseran pendorong emisi dan adanya perbedaan pola antar negara ini membuktikan bahwa setiap wilayah memiliki profil risiko emisi yang unik dan dinamis. Temuan ini menegaskan perlunya kebijakan mitigasi yang adaptif terhadap dinamika spasial dan temporal, serta pentingnya kolaborasi regional untuk mengatasi tantangan emisi di ASEAN.
Carbon dioxide (CO2) is the main greenhouse gas and the largest contributor to global warming driven by human activities. In the Southeast Asian (ASEAN) region, CO2 emissions exhibit an increasing trend, driven by economic growth, urbanization, and reliance on fossil fuels. The diverse social, economic, and environmental conditions across countries lead to a spatially and temporally heterogeneous relationship between CO2 emissions and their determinants. This research analyzes these spatio-temporal dynamics using the Geographically and Temporally Weighted Regression (GTWR) method. The data used include CO2 density from EDGAR, NDVI from MODIS, precipitation from ERA5, economic and demographic indicators from the World Bank, and energy indicators from EMBER for the period 2018–2023. The analysis reveals that pre-2020, economic growth and electricity consumption were the primary drivers of increased emissions. However, post-pandemic, environmental and energy factors such as NDVI and fossil fuel dependency became more significant in several countries. The optimal GTWR model, which utilized a bisquare kernel function and an adaptive bandwidth, achieved an R² value of 0,99. This shift in emission drivers and the differing patterns among countries demonstrate that each region possesses a unique and dynamic emission risk profile. These findings underscore the need for mitigation policies adaptive to spatio-temporal dynamics, as well as the importance of regional collaboration to address emission challenges in ASEAN.
URI: http://repository.ipb.ac.id/handle/123456789/165229
Appears in Collections:UT - Statistics and Data Sciences

Files in This Item:
File Description SizeFormat 
cover_G1401211008_5b5284daffb94dc981350b0862cfdcde.pdfCover626.75 kBAdobe PDFView/Open
fulltext_G1401211008_246202bbddd3465dbb2356b86b05e50f.pdf
  Restricted Access
Fulltext2.34 MBAdobe PDFView/Open
lampiran_G1401211008_1f5edbbef92249aab1955485e6671e32.pdf
  Restricted Access
Lampiran294.6 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.