Vernanda, Devita Prima (2019) Statistical Downscaling Output General Circulation Model Dengan Pendekatan Gaussian Copula Marginal Regression Untuk Prediksi Curah Hujan dan Banjir Genangan di Kota Surabaya. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211540000115_Undergraduate_Thesis.pdf Download (19MB) | Preview |
Abstract
Kota Surabaya merupakan kota metropolitan kedua di Indonesia dengan karakteristik wilayah berupa dataran rendah sehingga menyebabkan kota Surabaya rawan terkena bencana banjir. Banjir dapat disebabkan oleh dua hal, yakni penyebab alami dan tindakan manusia dimana salah satunya disebabkan oleh curah hujan yang tinggi. Kota Surabaya memiliki enam pos hujan pengamatan curah hujan yang terletak di Gunungsari, Wonokromo, Wonorejo, Keputih, Kedungcowek dan Gubeng. Karakteristik curah hujan pada enam pos hujan didapatkan pada bulan Mei hingga bulan Oktober curah hujan rendah sedangkan bulan November curah hujan tinggi. Pemodelan dilakukan dengan menggunakan data asli, data tanpa nilai nol di masing – masing curah hujan dan menggunakan lag ARIMA pada grid 1x1 dan grid 3x3 dengan Gaussian Copula Marginal Regression (GCMR). Hasil pemodelan terbaik untuk curah hujan diperoleh dengan tanpa melibatkan nilai 0 (tidak hujan) dengan prediktor seluruh variabel output General Circulation Model (GCM). Model terbaik dipilih berdasarkan dari nilai RMSE terendah dari masing – masing pos hujan. Diantara enam pos hujan yang berada di kota Surabaya, hanya pos hujan Kedungcowek yang tidak memiliki potensi untuk banjir.
=================================================================================================================================
Surabaya is the second metropolitan city in Indonesia with the characteristics in the form of lowlands, which causing Surabaya prone to floods. Floods can be caused by two things, natural causes and human actions. one of the causes is a high rainfall. Surabaya has six rainfall observation posts located in Gunungsari, Wonokromo, Wonorejo, Keputih, Kedungcowek and Gubeng. The characteristis of rainfall at six rainfall posts were obtained in May until October rainfall was low while in November rainfall was high. Modelling has been done with original data, data without zero values and using lag ARIMA for grid 1x1 and grid 3x3 using Gaussian Copula Marginal Regression. The best model for rainfall is obtained without involving zero values (no rain) with predictor from all of General Circulation Model (GCM) output variables. The best model is selected based on the lowest RMSE value. The analysis is done separately for each rain post. Among the six rain posts in the Surabaya, only Kedung Cowek rain post doesn’t have the potential to flood.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSSt 519.536 Ver s-1 2019 |
| Uncontrolled Keywords: | Flood, Gaussian Copula Marginal Regression, GCM, PCA, Rainfall, Statistical Downscaling |
| Subjects: | H Social Sciences > HA Statistics |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | DEVITA PRIMA VERNANDA |
| Date Deposited: | 04 Mar 2024 03:10 |
| Last Modified: | 04 Mar 2024 03:10 |
| URI: | http://repository.its.ac.id/id/eprint/64220 |
Actions (login required)
 |
View Item |