View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Forestry and Environment
      • UT - Forest Management
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Forestry and Environment
      • UT - Forest Management
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Deteksi Perubahan Hutan Mangrove Menggunakan Algoritma Decission Tree Di Kabupaten Deli Serdang Dan Sekitarnya Provinsi Sumatera Utara

      Thumbnail
      View/Open
      Cover (640.5Kb)
      Fulltext (6.962Mb)
      Lampiran (399.1Kb)
      Date
      2025
      Author
      Kamaluddin, Azka Fikri
      Rahaju, Sri
      Jaya, I Nengah Surati
      Metadata
      Show full item record
      Abstract
      Hutan mangrove di Kabupaten Deli Serdang dan sekitarnya mengalami tekanan akibat alih fungsi lahan, sehingga memerlukan metode deteksi perubahan yang akurat. Penelitian ini bertujuan membangun model deteksi perubahan tutupan lahan hutan mangrove antara tahun 2017 dan 2024 menggunakan algoritma decision tree. Kajian ini membandingkan tiga kombinasi peubah: (I) hanya peubah spektral, (II) hanya peubah sosio-geo-biofisik, dan (III) integrasi keduanya untuk menemukan model paling optimal. Hasil penelitian menunjukkan bahwa integrasi peubah spektral dan sosio-geo-biofisik secara signifikan meningkatkan kinerja model. Model yang hanya berbasis data citra (Kombinasi I) mencapai akurasi keseluruhan 87.5%, sedangkan model terintegrasi (Kombinasi III) mencapai 93.9% dengan akurasi kappa 91.5%. Peubah paling berpengaruh pada model terbaik adalah besar (magnitude) dan arah (direction) perubahan indeks tanah serta data substrat. Model ini berhasil mengidentifikasi 797.35 ha deforestasi mangrove, 1197.56 ha pertumbuhan kembali (regrowth), dan 6473.15 ha kawasan mangrove yang tidak berubah.
       
      Mangrove forests in Deli Serdang Regency and its surroundings are under pressure from land use conversion, requiring accurate change detection methods. This research aimed to build a mangrove forest land cover change detection model between 2017 and 2024 using a decision tree algorithm. This study compared three variable combinations: (I) only spectral variables, (II) only socio-geo-biophysical variables, and (III) an integration of both to find the most optimal model. The results indicate a significant performance improvement in the model that integrates spectral and socio-geo-biophysical variables. The model based solely on image data (Combination I) produced an overall accuracy of 87.5%, whereas the integrated model (Combination III) achieved an overall accuracy of 93.9% and a kappa accuracy of 91.5%. The most influential variables in the best model included the magnitude and direction of the soil index changes and substrate data. This model successfully identified 797.35 ha of mangrove deforestation, 1197.56 ha of mangrove regrowth, and 6473.15 ha of unchanged mangrove area.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/171512
      Collections
      • UT - Forest Management [3197]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository