Luthfi, Ahmad (2026) Peningkatan Akurasi Estimasi Curah Hujan Secara Spasial Menggunakan Metode Interpolasi Kriging With External Drifts dan Simulasi Debit Inflow di Waduk Sutami. Masters thesis, Institut Teknologi Sepuluh Nopember Surabaya.
![[thumbnail of 6016232031-Master_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
6016232031-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (1MB) | Request a copy |
Abstract
Di tengah dampak perubahan iklim yang semakin nyata, pengelolaan sumber daya air yang presisi sangat penting untuk menjamin keberlanjutan irigasi, pembangkit listrik tenaga air, dan pasokan air untuk domestik. Pada Bendungan Sutami di DAS Brantas Hulu, Jawa Timur, prakiraan debit masuk yang akurat menjadi krusial untuk mengantisipasi variabilitas hidrologi. Penelitian ini mengembangkan model hidrologi berbasis prakiraan curah hujan dasarian (10 hari) yang dikonversi menjadi debit dengan menggunakan regresi ridge. Sebanyak enam metode interpolasi spasial dibandingkan dengan data dari 108 raster dasarian (Januari I 2021–Desember III 2023), yaitu: Inverse Distance Weighting (IDW), Ordinary Kriging, Universal Kriging, Kriging with External Drift (CHIRP dan CHIRP+DEM), serta CHIRP. Jeda waktu 0–11 dasarian dipilih berdasarkan korelasi positif yang tertinggi. Regresi ridge optimal terjadi pada λ = 0,40. Validasi silang 50:50 menunjukkan bahwa IDW unggul dengan MAE terendah dan korelasi tertinggi. Validasi independen pada 97 pasang data (April III 2021–Desember III 2023) menguatkan temuan tersebut. Model IDW akhir (λ = 0,35–0,45) menghasilkan pola debit yang sangat mirip dengan observasi, dengan MAE 8,9–9,0 m³/detik dan korelasi 0,959. Penelitian ini menegaskan curah hujan dasarian yang diinterpolasi menggunakan IDW dan dimasukkan dalam persamaan regresi ridge merupakan prediktor debit masuk yang sangat andal di Bendungan Sutami. IDW terbukti paling mampu menangkap variabilitas curah hujan lokal pada topografi yang kompleks. Meskipun performanya kuat, dataset yang pendek (2021–2023) membatasi generalisasi pada kondisi ekstrem. Validasi jangka panjang dan kejadian ekstrem tetap diperlukan sebelum penerapan operasional penuh. Kerangka ini tetap menawarkan solusi ilmiah dan efisien untuk prakiraan debit real-time serta pengelolaan air yang resilien terhadap perubahan iklim
=================================================================================================================================
Amid the escalating impacts of climate change, precise water resource management is vital for sustaining irrigation, hydropower generation, and domestic water supply. At Sutami Dam, the Upper Brantas River Basin, East Java, accurate inflow discharge forecasting is crucial for managing hydrological variability. This study developed a hydrological forecasting model that converts dasarian (10-day) rainfall predictions into inflow discharge by using ridge regression. Six spatial interpolation techniques were evaluated using 108 dasarian rainfall rasters (January I 2021 to December III 2023): Inverse Distance Weighting (IDW), Ordinary Kriging (OK), Universal Kriging (UK), Kriging with External Drift (KED-CHIRP and KED-CHIRP+DEM), and raw CHIRP satellite estimates. Time lags of 0–11 decades wereanalyzed, selecting those with the strongest positive correlation to observed data discharge. Ridge regression achieved optimal performance at λ = 0.40. 50:50 Cross-validation revealed IDW as the best performer, yielding the lowest Mean Absolute Error (MAE) and the highest correlation coefficient. Independent validation 97 data pairs (April III 2021–December III 2023) confirmed IDW’s superiority. Final models (λ range 0.35–0.45) closely matched observed discharge patterns. delivering an MAE of 8.9–9.0 m³/s and a correlation coefficient of 0.959. The study concludes that dasarian rainfall interpolated via IDW and processed through Ridge Regression is a highly reliable predictor of Sutami Dam inflow. IDW excels at capturing local rainfall variability in complex topography. Although performance is robust, the short record (2021–2023) limits confidence regarding extreme events. Extended data would improve the model.validation over longer periods and extreme conditions is recommended prior to full operational use. Nevertheless, the framework provides a scientifically robust, computationally efficient solution for real-time inflow forecasting and climate-resilient waduk management.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Curah hujan dasarian, Regresi ridge, Interpolasi spasial, Validasi silang, Bendungan Sutami, Decadal rainfall, Ridge regression, Spatial interpolation, Cross-validation, and Sutami Reservoir |
| Subjects: | G Geography. Anthropology. Recreation > G Geography (General) > G70.5.I4 Remote sensing S Agriculture > S Agriculture (General) > S600.7.R35 Rain and rainfall |
| Divisions: | Faculty of Civil Engineering and Planning > Geomatics Engineering > 29101-(S2) Master Thesis |
| Depositing User: | Ahmad Luthfi |
| Date Deposited: | 26 Jan 2026 06:07 |
| Last Modified: | 26 Jan 2026 06:07 |
| URI: | http://repository.its.ac.id/id/eprint/130354 |
Actions (login required)
 |
View Item |