Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/107382Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Nurdiati, Sri | - |
| dc.contributor.advisor | Julianto, Mochamad Tito | - |
| dc.contributor.author | Rohimahastuti, Fadillah | - |
| dc.date.accessioned | 2021-07-13T12:00:57Z | - |
| dc.date.available | 2021-07-13T12:00:57Z | - |
| dc.date.issued | 2021 | - |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/107382 | - |
| dc.description.abstract | Fenomena global seperti El Nino Southern Oscillation (ENSO) dan Indian Ocean Dipole (IOD) dapat meningkatkan risiko kebakaran hutan dan lahan (Karhutla). Salah satu indikator yang dapat digunakan untuk mengidentifikasi Karhutla adalah hotspot. Penelitian ini bertujuan mengembangkan model regresi indikator global terhadap pola bersama sebaran hotspot dan prediksi indikator iklim dan menganalisis pengaruh fenomena ENSO dan IOD terhadap hotspot di Kalimantan dengan menggunakan data bulanan dari 2001-2020. Metode optimasi digunakan untuk memaksimalkan koefisien determinasi model, sedangkan Principal Component Regression (PCR) digunakan untuk membuat model regresi. Model prediksi yang dihasilkan dapat digunakan untuk memprediksi hotspot pada bulan berikutnya menggunakan informasi masing-masing indikator dalam 2 bulan terakhir. Validasi model menggunakan 3 tahun data sebagai testing set dan 17 tahun sebagai training set menghasilkan model regresi yang memberikan koefisien determinasi lebih dari 0,5 pada data testing sebesar 31,58%. Kontribusi tertinggi terhadap parahnya musim kemarau dan jumlah hotspot di Kalimantan dipengaruhi oleh fenomena ENSO yang disusul dengan IOD. | id |
| dc.description.abstract | Global phenomena such as the El Nino Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) can increase the risk of forest and land fires (Karhutla). One indicator that can be used to identify forest and land fires is a hotspot. This study aims to develop a global indicator regression model on the joint pattern of hotspot distribution and climate indicator prediction and analyze ENSO and IOD phenomena on hotspots in Kalimantan using monthly data from 2001-2020. The optimization method is employed to maximize the coefficient of model determination and the Principal Component Regression (PCR) is employed to create a regression model. The result of this study can be applied to predict hotspots in the following month using the information of each indicator in the last 2 months. The model validation using 3 years of data as a testing set and 17 years as a training set produces a regression model that provides a coefficient of determination of more than 0,5 on the testing data of 31,58%. The highest contribution to the severity of the dry season and the number of hotspots in Kalimantan is influenced by the ENSO phenomenon followed by IOD. | id |
| dc.language.iso | id | id |
| dc.publisher | IPB University | id |
| dc.title | Karakterisasi Model Regresi Indikator Global terhadap Pola Bersama Sebaran Hotspot dan Prediksi Indikator Iklim di Kalimantan | id |
| dc.title.alternative | Characterization of Global Indicator Regression Model on The Joint Pattern of Hotspot Distribution and Climate Indicator Prediction in Kalimantan | id |
| dc.type | Undergraduate Thesis | id |
| dc.subject.keyword | ENSO | id |
| dc.subject.keyword | Hotspot | id |
| dc.subject.keyword | IOD | id |
| dc.subject.keyword | PCR | id |
| dc.subject.keyword | Rainfall | id |
| Appears in Collections: | UT - Mathematics | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| Cover, Lembar Pernyataan, Abstrak, Lembar Pengesahan, Prakata, Daftar Isi.pdf Restricted Access | Cover | 2.07 MB | Adobe PDF | View/Open |
| G54170002_Fadillah Rohimahastuti.pdf Restricted Access | Fullteks | 8.55 MB | Adobe PDF | View/Open |
| Lampiran.pdf Restricted Access | Lampiran | 2.05 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.