PREDICTION OF VOLCANIC ASH DISPERSION IN INDONESIA WITH FALL3D

• Main Author: RACHMANDIENTA HERMAWAN (nim : 12810020), ANGGHREA
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/25591
• Institution: Institut Teknologi Bandung

<p align="justify">Volcanic ash is destructive. Prediction distribution of volcanic ash is needed to prevent losses in the aviation world. Previous study using PUFF model (model trajectory) as a predictive model of volcanic ash dispersion. The weakness of this model lies in not accounting for the vertical velocity in the model. Subsequent research confirmed the role of the vertical velocity with FALL3D volcanic ash dispersion model. However, this study wasdoing simulation and the role of the <br /> <br /> vertical speed is only seen through the input data. The purpose of this study is to predict the different cases spatially and temporally and reviewing the role of the <br /> <br /> vertical velocity in the prediction of volcanic ash dispersion using FALL3D model. FALL3D model input is meteorological data from WRF output with 9 kmresolution. Output FALL3D model verified qualitatively and quantitatively with the MTSAT satellite data. In this study was also conducted trial predictions of volcanic ash dispersion in Indonesia using FALL3D model on 13 September 2014 with WRF model input resolution of 27 km. Overall FALL3D models can predict the direction of the spread of volcanic ash fairly well (qualitatively) although quantitatively less accurate but not so bad. The role of vertical velocity assessed by comparing the PUFF output and <br /> <br /> FALL3Doutput. FALL3D able to predict the direction of the distribution well despite the complex weather conditions, and also quantitatively accurate enough. This is because the FALL3D model calculates vertical speed in the equation. The trial predictions results demonstrate that the model prediction FALL3D quantitatively accurate. FALL3D models running for ~ 13 minutes for each of the <br /> <br /> volcanoes with WRF model input resolution of 27 km.<p align="justify">