RMSE values < 0>Tanjung station, which is classified as an estuarine morphology, where the model's accuracy declined. This indicates the model's overall reliability across various coastal morphologies in Indonesia. Meanwhile, the FES2014 model demonstrated good accuracy in narrow channels and estuarine morphologies. The regional BIG model remained consistent across nearly all morphological types, as evidenced by the absence of significant anomalies."> RMSE values < 0>Tanjung station, which is classified as an estuarine morphology, where the model's accuracy declined. This indicates the model's overall reliability across various coastal morphologies in Indonesia. Meanwhile, the FES2014 model demonstrated good accuracy in narrow channels and estuarine morphologies. The regional BIG model remained consistent across nearly all morphological types, as evidenced by the absence of significant anomalies.">
Laporkan Masalah

Analisis Perbandingan Model Pasang Surut Global (FES2014 dan TPXO10-Atlas) dan Model Regional BIG terhadap Data Pengamatan pada Berbagai Tipe Morfologi Pantai

Amelia Virgiola Damayanti, Ir. Abdul Basith, S.T., M.Si., Ph.D.

2025 | Skripsi | TEKNIK GEODESI

Keterbatasan ketersediaan data pasang surut di Indonesia menjadi tantangan dalam memahami dinamika air laut, khususnya fenomena pasang surut secara akurat. Dengan panjang garis pantai mencapai 108.000 km, jumlah 290 stasiun pasut saat ini dinilai tidak mencukupi, karena satu stasiun pasut akan merepresentasikan 372 km garis pantai. Hal tersebut semakin kompleks, karena keragaman morfologi pantai di Indonesia. Untuk mengatasi keterbatasan tersebut, pemanfaatan model pasut, baik regional maupun global dapat menjadi alternatif dalam memperoleh data pasut yang akurat. Model global seperti FES dan TPXO yang sering digunakan dalam studi perairan dangkal karena menunjukkan akurasi yang tinggi. Sementara itu, model regional BIG yang dikembangkan dari satelit multi misi serta data pengamatan lokal, dinilai mampu menangkap variasi lokal dengan lebih baik. Namun, akurasi model TPXO versi terbaru yaitu TPXO10-Atlas di perairan Indonesia belum diketahui, sedangkan model FES2014 telah digunakan pada studi sebelumnya belum memperhatikan keberagaman morfologi di Indonesia. Oleh karena itu, penelitian ini bertujuan untuk membandingkan kesesuaian model pasut regional BIG, FES2014 dan TPXO10-Atlas dalam memprediksi pasang surut di wilayah studi.

Data yang digunakan dalam penelitian ini terdiri dari model regional BIG, model FES2014, dan model TPXO10-Atlas. Periode pasut yang digunakan yaitu pada 7 April – 21 April 2025. Sebelas stasiun pasut dipilih untuk merepresentasikan keberagaman morfologi pantai Indonesia, seperti selat, teluk, tanjung, alur sempit, pantai terbuka, dan muara sungai. Seluruh data dianalisis dalam periode 15 hari dari 7 April – 21 April 2025 untuk menangkap dinamika pasut di lokasi studi. Data pengamatan pasut diolah dengan mendeteksi gap, spike, dan discontinue, lalu difilter menjadi interval per jam. Model regional BIG menyediakan prediksi elevasi secara langsung. Sementara itu, model pasut global FES2014 dan TPXO10-Atlas memerlukan interpolasi spasial dengan metode Inverse Distance Weighting (IDW) untuk memperkirakan nilai amplitudo dan fase di 11 titik sesuai stasiun yang dipilih. Setelah interpolasi, prediksi elevasi pasut dilakukan dengan metode Least Square pada software MATLAB. Kesesuaian masing – masing model pasut terhadap data pengamatan dinilai menggunakan Root Mean Square Error (RMSE) dan koefisien korelasi. 

Hasil pengolahan data pasut menunjukkan bahwa seluruh model menunjukkan kesesuaian yang baik di hampir 6 tipe morfologi pantai yang dianalisis. Hal tersebut ditunjukkan dengan rata – rata RMSE < 0>nilai korelasi yang mendekati 1 di hampir 11 stasiun. Namun, terdapat pengecualian ditemukan pada model FES2014 di stasiun Lembar, yang menunjukkan korelasi rendah yaitu, < 0>menandakan keterbatasan model dalam merepresentasikan dinamika air di teluk sempit. Model TPXO10-Atlas terbukti paling unggul di hampir seluruh tipe morfologi pantai dengan RMSE < 0>Pengecualian terjadi pada stasiun Kuala Tanjung yang diklasifikasikan sebagai morfologi muara sungai, di mana akurasinya menurun. Hal ini menunjukkan keandalaannya di berbagai morfologi pantai di Indonesia. Sementara itu, model FES2014 menunjukkan akurasi yang baik pada tipe morfologi alur sempit dan muara sungai. Model regional BIG konsisten di hampir seluruh tipe morfologi, ditandai dengan tidak adanya anomali yang signifikan.

The limited availability of precise tidal observation data in Indonesia poses a major constraint on accurately modeling ocean dynamics, particularly tidal phenomena. With a coastline stretching over 108,00 km, only 290 tide stations are currently available, representing approximately 372 km of coastline per station, which is deemed insufficient. This issue is further complicated by Indonesia’s diverse coastal morphology, making the provision of representative tidal data difficult. To address this limitation, the use of tidal models—both regional and global—offers an alternative for obtaining accurate tidal information. Global tidal models (e.g., FES and TPXO) are frequently used in coastal studies due to their high accuracy. Meanwhile, the regional BIG model, developed from multimission satellite data and local observations, is considered more capable of capturing local variations. However, the accuracy of the latest TPXO version, TPXO10-Atlas, remains unevaluated in Indonesian waters. Moreover, while FES2014 has been used previously, its performance against Indonesia’s diverse coastal morphology has not been systematically assessed. Therefore, this study aims to compare the performance of the regional BIG model, FES2014, and TPXO10-Atlas in predicting tides within the study area.

The data used in this study include the regional BIG model, FES2014, and TPXO10-Atlas. The analyzed tidal period spanned from April 7 to April 21, 2025. Eleven tide stations were selected to represent the diversity of Indonesia’s coastal morphology, including straits, bays, capes, narrow channels, open coasts, and river estuaries. All data were analyzed over 15 days to capture tidal dynamics at the study sites effectively. Observational tidal data underwent quality control, including the detection of gaps, spikes, and discontinuities, and were subsequently filtered into hourly intervals. The regional BIG model provides direct elevation predictions, whereas the global models FES2014 and TPXO10-Atlas required spatial interpolation using the Inverse Distance Weighting (IDW) method to estimate amplitude and phase. Following interpolation, tidal elevation predictions were generated using the Least Squares method in MATLAB. The accuracy of each tidal model was evaluated using Root Mean Square Error (RMSE) and correlation coefficients.

The results indicate good agreement among all models across the six analyzed coastal morphology types. This is supported by average RMSE values < 0 class="cf0">RMSE values < 0>Tanjung station, which is classified as an estuarine morphology, where the model's accuracy declined. This indicates the model's overall reliability across various coastal morphologies in Indonesia. Meanwhile, the FES2014 model demonstrated good accuracy in narrow channels and estuarine morphologies. The regional BIG model remained consistent across nearly all morphological types, as evidenced by the absence of significant anomalies.

Kata Kunci : Pasang Surut, TPXO10-Atlas, FES2014, Model Regional BIG, Evaluasi Akurasi

  1. S1-2025-481987-abstract.pdf  
  2. S1-2025-481987-bibliography.pdf  
  3. S1-2025-481987-tableofcontent.pdf  
  4. S1-2025-481987-title.pdf