Model Output Statistics Untuk Prakiraan Cuaca Jangka Pendek Menggunakan Statistically Inspired Modification of Partial Least Square

Kurniasari, Vira Oktavia (2017) Model Output Statistics Untuk Prakiraan Cuaca Jangka Pendek Menggunakan Statistically Inspired Modification of Partial Least Square. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 1313100007-Undergraduate_Theses.pdf]
Preview
Text
1313100007-Undergraduate_Theses.pdf - Published Version

Download (4MB) | Preview

Abstract

Perubahan cuaca yang ekstrim seringkali memberikan dampak buruk bagi kehidupan manusia. Dampak buruk tersebut dikarenakan kurangnya informasi prakiraan cuaca yang cepat dan tepat. BMKG telah melakukan kajian pemodelan cuaca dengan data Numerical Weather Project (NWP). Namun, NWP masih menghasilkan ramalan yang bias. Untuk mengatasinya, dilakukan post processing dengan Model Output Statistics (MOS) menggunakan Statistically Inspired Modification of Partial Least Square (SIMPLS). Sebelumnya, dilakukan reduksi dimensi menggunakan Independent Component Analysis (ICA). Observasi cuaca yang digunakan sebagai variabel respon adalah suhu maksimum (Tmax), suhu minimum (Tmin), dan kelembapan (RH), sedangkan parameter NWP digunakan sebagai variabel prediktor. Berdasarkan validasi model dengan kriteria RMSEP, disimpulkan bahwa RMSEP untuk Tmax di Stasiun Juanda berkriteria baik, sedangkan di Stasiun Soekarno Hatta dan Ngurah Rai berkriteria sedang. RMSEP untuk Tmin berkriteria baik di Stasiun Juanda dan Soekarno Hatta. Namun, RMSEP Tmin di Stasiun Ngurah Rai berkriteria buruk. Nilai RMSEP untuk RH berkriteria baik di tiga stasiun. Nilai %IM atau koreksi bias NWP untuk prakiraan Tmax, Tmin, dan RH berkisar antara 26% - 92%.

=========================================================================

The extreme change of weather often gives negative impacts on human life. These negative impacts are due to the lack of information about the weather forecast in a quick and accurate way. BMKG has attempted a study of short-term weather modelling with data of Numerical Weather Project (NWP). However, NWP still produces a bias prediction. To overcome this, post processing is needed with a Model Output Statistics (MOS) using Statistically Inspired Modification of Partial Least Square (SIMPLS). Previously, there is a dimensional reduction using Independent Component Analysis (ICA). The weather observation that is used as the response variable is the maximum temperature (Tmax), the minimum temperature (Tmin), and humidity (RH), while the NWP parameter is used as the predictor variable. Based on the model validation with the criteria of RMSEP, it is concluded that the criterion of RMSEP for Tmax at Juanda Station are in good, while at Soekarno Hatta and Ngurah Rai Station are in average. The criterion of RMSEP for Tmin is good at Juanda and Soekarno Hatta Station, but poor at Ngurah Rai Station. RMSEP RH is good at the three stations. The value of %IM or the bias correction of NWP for the forecast of Tmax, Tmin, and RH ranged from 26% - 92%.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: ICA, Kelembapan, MOS, NWP, SIMPLS, Suhu
Subjects: Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Q Science > QA Mathematics > QA278.5 Principal components analysis. Factor analysis. Correspondence analysis (Statistics)
Divisions: Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: Vira Oktavia Kurniasari
Date Deposited: 15 Jan 2018 07:18
Last Modified: 05 Mar 2019 04:00
URI: http://repository.its.ac.id/id/eprint/48524

Actions (login required)

View Item View Item