#TITLE_ALTERNATIVE#

• Main Author: HERMAWAN (NIM 15107005); Pembimbing: Dr. rer. nat. Poerbandono dan Dr. techn. Dudy Darmawan Wi, RENDY
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/14422
• Institution: Institut Teknologi Bandung

Tidal prediction is an activity to get the data sea level in the future. Tidal predictions can be used to obtain information such as tidal datum. Tidal predictions made using previous tidal observational data, where the observational data that varies in length and contain errors such as spike and gap. These circumstances lead to the prediction data that obtained is not close to the reality in the field. Tidal predictions can be obtained by various methods such as the least squares method. This method of conditioning the value of squared errors that occur is minimum. Tidal predictions are made of the tidal components derived from the method of least squares. The error in the prediction of tides can be done by comparing the predictions with observational data at the same observation time. It also can be seen from the error amplitude of tidal constituent components. Tidal Datum also be made of specific tidal components depending on its type. Quality of the tidal predictions can be seen from the value of the variance between the prediction data and the observational data where the tolerance limit variance is 0.008856 m. The length of the observed data have greatly effect to the variance, the shorter the greater the variance were obtained. On missing data prediction quality of the data depends on the amount of gap data, both the length of the data and the number of occurrences. As for the spike, the amount of data that is allowed to jump is 0.5% of the overall data. The resulting tidal datum is independent from the length of observation and of the error but effected by the value tidal component that obtained.