ESTIMATION OF PROBABLE MAXIMUM PRECIPITATION (PMP) USING SATELLITE DATA

• Main Author: RAHMAH AGRIAMAH (NIM : 12813021), DEWI
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/21692
• Institution: Institut Teknologi Bandung

Probable Maximum Precipitation (PMP) is commonly used to estimate Probable Maximum Flood (PMF) as a reference for flood management planning, to determine temporary retention areas and spillways design on dams. The objective of this research is to estimate PMP using satellite data, especially Tropical Rainfall Measuring Mission (TRMM) as one of the solution to the lack of observation data, and to know the representativeness of satellite data used. TRMM data used is the original TRMM and corrected TRMM. Corrected TRMM was obtained from the corrected TRMM data to observation data using the correction bias method of the region average areas. Prior to application, the method of correction bias of the region average areas was tested first using a variety of testing and training station schemes. The average area correction bias was found capable to decrease the root mean square error (RMSE) value to 29-38% of the RMSE of the original TRMM. Thus the corrected TRMM data used for the PMP estimation using the Hershfield’s statistical method. The validity of Hershfield's statistical methods in the western part of Java Island was examined by dividing the observational data into three decades. The test results show that Hershfield's statistical method is consistently used for different data conditions. After the correction bias testing and validation of the Hershfield’s statistical method, a PMP distribution map was created for durations 3, 6, 9, 12, 18, 24, 36 and 48 hours in the western part of Java Island. PMP distribution maps are important for estimating reliable and consistent PMP for each location in the western part of Java Island.