Tropical cyclone Cempaka entered South Java sea from 25th November 2017 until 27th November 2017 and caused strong winds and high waves. Delft 3D Modeling Simulation was used to investigate the effect of tropical cyclone Cempaka development on sea level rise in coastal areas and high waves in the South Java sea. The FNL wind data with a resolution of 0.25° × 0.25 ° and GEBCO bathymetry data with a grid resolution of 30 seconds was used as input to the Delft 3D model. The results of the study showed that there was an increase of significant wave height during the tropical cyclone Cempaka period with significant wave values from 1 to 3.5 meters on 27th November 2017. While for the value of sea level height in the coastal area showed results there was an increase during cyclone Cempaka period with anomalies values to 0.4. However, at the center of the cyclone had a low anomaly value that reached -0.2. From the results of the correlation test and RMSE test on sea level Delft 3D output obtained a correlation value of 0.85 and error 0.43 in the Sadeng, Yogyakarta. while the correlation test and RMSE test for the significant wave height parameter obtained a correlation value -0.198 and error value 0.168.

<p class="AbstractEnglish"><strong>Abstract:</strong> Numerical weather predictions are currently being developed to address the need for high resolution rainfall forecasting. However, numerical weather forecasts in Indonesia are still problematic in terms of the accuracy of numerical models. Several previous studies have shown that modeling accuracy is strongly influenced by errors in the initial condition data. This study examines efforts from the research and development of the Weather Forecast and Forecast (WRF) model of preliminary data using a satellite beam assimilation procedure for forecasting rainfall in the Ambon region for two different case studies in 2018. Six experimental models are carried out by assimilation of sensors AMSU-A and MHS satellites use the WRFDA 3DVar system. This research was conducted by increasing the assimilation analysis on the initial data model, analyzing the model skills in the dichotomy of rainfall predictions, rainfall criteria, spatial rainfall, and time series of rainfall accumulation compared to BMKG rainfall observation data. The results showed that the DA AMSU-A and MHS experiments correctly modified the initial condition data of the model. Meanwhile, the results of dichotomous verification revealed that the DA observation experiment had the highest skill score forecast compared to other assimilation. but more experiments are needed in the northern Sumatra area to provide more significant results.</p><p class="KeywordsEngish"><strong>Abstrak:</strong> Prediksi cuaca numerik saat ini terus dikembangkan untuk mengatasi kebutuhan akan ramalan curah hujan resolusi tinggi. Namun, ramalan cuaca numerik di Indonesia masih bermasalah dalam hal akurasi model numerik. Beberapa penelitian sebelumnya menunjukkan bahwa akurasi pemodelan sangat dipengaruhi oleh kesalahan dalam data kondisi awal. Penelitian ini mengkaji upaya-upaya dari penelitian dan pengembangan model Prakiraan Cuaca dan Prakiraan (WRF) data awal menggunakan prosedur asimilasi pancaran satelit untuk prakiraan curah hujan di wilayah Ambon untuk dua studi kasus pada musim yang berbeda selama 2018. Enam model eksperimental dijalankan dengan asimilasi sensor satelit AMSU-A dan MHS menggunakan WRFDA sistem 3DVar. Penelitian ini dilakukan dengan analisis peningkatan asimilasi pada model data awal, analisis keterampilan model pada dikotomi prediksi curah hujan, kriteria curah hujan, curah hujan spasial, dan time series akumulasi hujan dibandingkan dengan data pengamatan curah hujan BMKG. Hasil penelitian menunjukkan bahwa eksperimen DA AMSU-A dan MHS memodifikasi data kondisi awal model dengan benar. Sementara itu, hasil verifikasi dikotomis mengungkapkan bahwa eksperimen DA observasi memiliki skor ketrampilan prakiraan tertinggi dibandingkan dengan asimilasi lainnya. namun diperlukan lagi percobaan di daerah Sumatra utara untuk memberikan hasil yang lebih signifikan.</p>