View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Computer Science
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Computer Science
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Model Rekonstruksi Data Temporal Tinggi Muka Air Lahan Gambut

      Thumbnail
      View/Open
      Cover (596.1Kb)
      Fulltext (5.623Mb)
      Lampiran (152.1Kb)
      Date
      2023
      Author
      Hakim, Abdul
      Buono, Agus
      Sitanggang, Imas Sukaesih
      Metadata
      Show full item record
      Abstract
      Lahan gambut adalah akumulasi tumbuhan mati yang terdekomposisi penuh selama lebih dari ribuan tahun dan dapat menyimpan karbon, memproduksi oksigen serta mengatur tingkat air. Tinggi muka air (TMA) menjadi indikator penting dalam peringatan kebakaran lahan dan hutan (karhutla). Sistem Pemantau Air Lahan Gambut (Sipalaga) sudah menjadi sistem yang digunakan untuk merekam TMA kelembaban gambut, dan tingkat curah hujan. Penelitian ini menggunakan data Sipalaga berstasiun di Rimba Panjang. Namun data yang digunakan masih mengandungi banyak missing value yang harus ditangani. Penelitian ini berfokus untuk membandingkan metode rekonstruksi data TMA lahan gambut menggunakan Last Observation Carried Forward, Next Observation Carried Forward, Interpolation, Weighted Moving Average, Kalman filter, K-nearest Neighbor Imputation, dan Simple Averaging Ensemble. Data yang sudah direkonstruksi dievaluasi menggunakan Long-Short Term Memory. Hasil terbaik yang diperoleh adalah menggunakan Stacked LSTM dengan unit 64 dan 128, dengan metode Spline Interpolation menghasilkan MSE 0.00012, MAE 0.00822 dan R2 0.95316. Hasil penelitian ini diharapkan dapat membantu prediksi TMA lahan gambut sebagai peringatan dini karhutla di Indonesia.
       
      Peatlands are accumulations of dead plants that have fully decomposed over thousands of years that can store carbon, produce oxygen and regulate water levels. The groundwater level (GWL) is an important indicator for forest fires. The Sistem Pemantau Air Lahan Gambut (Sipalaga) has become a system used to record GWL, peatland moisture, and rainfall levels at peatland locations. This study will use data from Sipalaga stationed at Rimba Panjang. However, the data used still contains many missing values that must be addressed. This research will focus on comparing methods of reconstructing peatlands’ GWA data using Last Observation Carried Forward, Next Observation Carried Forward, Weighted Moving Average, Interpolation, Kalman filter, K-nearest Neighbor Imputation and Simple Averaging Ensemble. Reconstructed data will be evaluated with Long-Short Term Memory predictions. The best model made in this study is using Stacked LSTM with 64 and 128 units using Spline Interpolation resulting MSE 0.00012, MAE 0.00822 dan R2 0.95316. The results of this study are expected to help predict the TMA of peatlands as an early warning of forest and land fires in Indonesia.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/122082
      Collections
      • UT - Computer Science [2482]

      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