Pemodelan Kekeringan di Wilayah Nusa Tenggara Timur (NTT) dengan Metode Statistical Downscaling Pra-Pemroses PCA

Yuliatin, Ika Lulus (2017) Pemodelan Kekeringan di Wilayah Nusa Tenggara Timur (NTT) dengan Metode Statistical Downscaling Pra-Pemroses PCA. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
1313100078-Undergraduate_Theses.pdf - Published Version

Download (5MB) | Preview

Abstract

Kekeringan merupakan salah satu bencana alam yang terjadi secara perlahan-lahan namun membawa dampak sangat luas dan bersifat lintas sektor. Menurut Indonesia Food Security Monitoring Bulletin, 10 Kabupaten di Provinsi NTT menjadi prioritas utama dalam klasifikasi yang mengalami dampak kekeringan di Indonesia. Oleh karena itu penelitian ini mengkaji risiko kekeringan di NTT berdasarkan nilai Standardized Precipitation Index (SPI) pada skala waktu satu, dua , dan tiga bulanan menggunakan metode Statistical Downscaling reduksi Principal Component Analysis. Penelitian dilakukan pada data curah hujan harian di NTT dan rata-rata geopotensial bulanan ketinggian 500, 850, 875, 900, 975, dan 1000 hPa. Hasil analisis yang dilakukan, nilai SPI tiga bulanan mampu memberikan hasil paling baik dalam meramalkan kekeringan dan SPI satu bulanan nilai akurasinya paling kecil dalam meramalkan kekeringan. Berdasarkan data out sample selama satu tahun, seluruh peramalan kekeringan memberikan hasil yang sama dengan observasi kecuali bulan Januari dan September. Ketinggian geopotensial 900 hPa merupakan ketinggian paling baik dalam meramalkan kekeringan. ================================================================= Drought is one of the natural disasters that occur slowly but has wide impact and cross-sectoral. According to Indonesia Food Security Monitoring Bulletin, 10 districts in NTT Province are the top priority of the drought affected classification in Indonesia. Therefore, this study examines drought risk in NTT based on the value of Standardized Precipitation Index (SPI) on time scale of one, two, and three monthly using Statistical Downscaling pre-processing Principal Component Component Analysis method. The research was conducted on daily rainfall data in NTT and monthly geopotential average height of 500, 850, 875, 900, 975, and 1000 hPa. The result of the analysis, SPI value of three monthly can give the best result in predicting drought and SPI one monthly has least accuracy in predicting drought. Based on out-sample data for one year, all drought forecasting gave similar results to observations except in January and September. The height of geopotensial 900 hPa is the best height in predicting drought.

Item Type: Thesis (Undergraduate)
Additional Information: RSSt 519.535 4 Yul p
Uncontrolled Keywords: Kekeringan, geopotensial, Principal Component Analysis, Statistical Downscaling, Standardized Precipitation Index
Subjects: G Geography. Anthropology. Recreation > GA Mathematical geography. Cartography
Q Science > QA Mathematics > QA278.5 Principal components analysis. Factor analysis.
Divisions: Faculty of Mathematics and Science > Statistics > (S1) Undergraduate Theses
Depositing User: Yuliatin Ika Lulus
Date Deposited: 12 Dec 2017 04:07
Last Modified: 05 Mar 2019 06:50
URI: http://repository.its.ac.id/id/eprint/48037

Actions (login required)

View Item View Item