Prediction of Missing Streamflow Data using Principle of Information Entropy
Abstract: Incomplete (missing)
of streamflow data often occurs. This can be caused by a not continous data
recording or poor storage. In this study, missing consecutive streamflow data
are predicted using the principle of information entropy. Predictions are performed
using the complete monthly streamflow information from the nearby river. Data
on average monthly streamflow used as a simulation sample are taken from
observation stations Katulampa, Batubeulah, and Genteng, which are the Ciliwung
Cisadane river areas upstream. The simulated prediction of missing streamflow
data in 2002 and 2003 at Katulampa Station are based on information from
Genteng Station, and Batubeulah Station. The mean absolute error (MAE) average
obtained was 0,20 and 0,21 in 2002 and the MAE average in 2003 was 0,12 and
0,16. Based on the value of the error and pattern of filled gaps, this method
has the potential to be developed further.
Author: Santosa, B., Legono, D,
Suharyanto
Journal Code: jptsipilgg140025
