Estimasi Return Level pada Pemodelan Spatial Extreme Value Kecepatan Arus Laut Bali dengan Pendekatan Max-Stable Process Model Smith dan Brown-Resnick

Sanjaya, Nyoman Gede Trisna (2023) Estimasi Return Level pada Pemodelan Spatial Extreme Value Kecepatan Arus Laut Bali dengan Pendekatan Max-Stable Process Model Smith dan Brown-Resnick. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06311940000054-Undergraduate_Thesis.pdf] Text
06311940000054-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2025.

Download (4MB) | Request a copy

Abstract

Bali merupakan destinasi pariwisata terpopuler kedua di dunia pada 2023. Salah satu pariwisata terbaik adalah keindahan pesisir pantainya. Walaupun menjadi destinasi pariwisata terbaik, namun tidak jarang terjadi bencana di wilayah pesisir pantai Bali. Karakter pantai di Bali Selatan memiliki tingkat abrasi yang tinggi, serta Bali bagian Timur sekitar Gianyar juga perlu diwaspadai karena terjangan gelombang kuat. Salah satu faktor penting terjadinya bencana pesisir dari perairan seperti banjir rob dan abrasi adalah arus laut. Arus laut merupakan salah satu faktor oseanografi yang cepat lambatnya dipengaruhi kedalaman arus, kecepatan angin, dan pasang surut air laut. Analisis spasial kecepatan arus laut dilakukan pada empat lokasi perairan di Laut Bali. Metode yang digunakan adalah Spatial Extreme Value pendekatan Max-Stable Process model Smith dan Brown-Resnick. Data yang digunakan adalah data harian periode 2 Maret 2017 - 30 Desember 2020. Pemilihan data ekstrem menggunakan Block Maxima dengan blok 14 harian, sehingga terdapat 100 blok untuk tiap lokasi perairan. Proporsi data training dan testing adalah 80:20. Data training mengikuti distribusi Generalized Extreme Value dan tidak memiliki tren pola (stasioner). Data training kemudian ditransformasi ke unit margin Frechet. Hasil pengukuran koefisien ekstremal berkisar di antara 1,18604 - 1,59485 menunjukan dependensi antarlokasi cukup kuat. Model trend surface terbaik adalah model yang hanya memiliki faktor koordinat longitude pada parameter location dan latitude pada parameter scale. Setelah estimasi parameter spasial model Smith dan Brown-Resnick selanjutnya adalah validasi model dengan RMSE dan MAPE. Nilai RMSE dan MAPE model Smith sebesar 0,15503 dan 7,75076%, sedangkan model Brown-Resnick sebesar 0,29576 dan 14,12131%. Nilai return level periode yang sama setelah data testing tergolong arus kuat dan berturut-turut sebesar 1,20586 m/s, 1,63592 m/s, 1,51322 m/s, dan 2,13233 m/s untuk Perairan Serangan, Gianyar, Nusa Dua, dan Nusa Lembongan. Informasi return level ini diharapkan menjadi pertimbangan yang dapat digunakan oleh instansi terkait seperti Balai Pengelolaan Sumberdaya Pesisir dan Laut (BPSPL) dan Badan Penanggulangan Bencana Daerah (BPBD) Provinsi Bali sebagai upaya mitigasi bencana pesisir agar lebih efektif, efisien, dan tepat sasaran.
================================================================================================================================
Bali is the world's second most popular tourist destination in 2023. One of the best tourisms is the beauty of its coasts. It is not uncommon for disasters to occur in the coastal areas of Bali. The beach in South Bali has a high level of abrasion, and around Gianyar also needs to be watched out for because of the crashing waves. One important factor in the occurrence of coastal disasters from waters such as tidal flooding and abrasion is ocean currents. Sea currents are one of the oceanographic factors which are quickly and slowly influenced by current depth, wind speed, and sea tides. Spatial analysis of sea currents velocity was carried out at four water locations in the Bali Sea. The method used is the Spatial Extreme Value with Max-Stable Process model Smith and Brown-Resnick. The data used is daily data for the period March 2, 2017 to December 30, 2020. Extreme data selection with Block Maxima uses 14 daily blocks, so there are 100 blocks for each water location. The proportion of training and testing data is 80:20. The training data follows the Generalized Extreme Value distribution and has no pattern trend (stationary), and then transformed to unit of margin Frechet. The measurement results of the extreme coefficients between 1.18604 to 1.59485 show a fairly strong dependence between locations. The best trend surface model is a model that only has longitude coordinates on the location parameter and latitude on the parameter scale. After estimation of the spatial parameters of the Smith and Brown-Resnick models, the next step is model validation with RMSE and MAPE. The RMSE and MAPE values for the Smith model are 0.15503 and 7.75076%, while the Brown-Resnick model are 0.29576 and 14.12131%. Return level values for the same period after testing data are classified as strong currents and are respectively 1.20586 m/s, 1.63592 m/s, 1.51322 m/s and 2.13233 m/s for Serangan, Gianyar, Nusa Dua, and Nusa Lembongan Waters. This return level information is expected to be a consideration that can be used by related institute such as the Coastal and Marine Resources Management Institution (BPSPL) and the Bali Province Regional Disaster Management Institution (BPBD) as an effort to manage coastal disasters to make them more effective, efficient, and on target.

Item Type: Thesis (Other)
Uncontrolled Keywords: Kecepatan Arus Laut, Max-Stable Process, Model Brown-Resnick, Model Smith, Return Level, Brown-Resnick Model, Return Level, Sea Currents Velocity, Smith Model
Subjects: G Geography. Anthropology. Recreation > GC Oceanography > GC89 Sea Level
H Social Sciences > HA Statistics > HA30.3 Time-series analysis
H Social Sciences > HA Statistics > HA30.6 Spatial analysis
H Social Sciences > HA Statistics > HA31.7 Estimation
Q Science > QA Mathematics > QA614.58 Catastrophes
T Technology > TC Hydraulic engineering. Ocean engineering > TC147 Ocean wave power.
T Technology > TC Hydraulic engineering. Ocean engineering > TC203.5 Coastal engineering
T Technology > TC Hydraulic engineering. Ocean engineering > TC343+ Reclamation of land from the sea
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User: Nyoman Gede Trisna Sanjaya
Date Deposited: 23 Jun 2023 05:47
Last Modified: 23 Jun 2023 05:53
URI: http://repository.its.ac.id/id/eprint/98179

Actions (login required)

View Item View Item