Analysis of rainfall datasets for major Asian cities (Jakarta)

This reliability study involves the assessment of the accuracy of the APHRODITE and TRMM datasets. The assessment will be based on parameters looking into 3 categories which are intensity, duration and frequency Software were employed to extract data from the product’s website and retrieve data from...

Full description

Saved in:
Bibliographic Details
Main Author: Loke, Jun Wei.
Other Authors: Lo Yat-Man, Edmond
Format: Final Year Project
Language:English
Published: 2013
Subjects:
Online Access:http://hdl.handle.net/10356/52817
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:This reliability study involves the assessment of the accuracy of the APHRODITE and TRMM datasets. The assessment will be based on parameters looking into 3 categories which are intensity, duration and frequency Software were employed to extract data from the product’s website and retrieve data from the binary files. Specific methodology was developed to successfully extract all data. Intensity of rainfall was the least reliable as average monthly rainfall of APHRODITE and TRMM were 50% and 40% to the actual amounts from ground stations respectively. APHRODITE better estimates rainfall data during heavy rainfall season while higher underestimation of data during light rainfall season. As for TRMM, higher underestimation was observed during heavy rainfall season while a lower underestimation was observed during light rainfall season. As for the frequency of wet days, TRMM reflects good accuracy with only 17% overestimation while APHRODITE reflects 70% overestimation. TRMM displays good duration accuracy for the wet spell length with an overestimation of only 10% against the actual rainfall data while APRODITE shows a 100% overestimation. TRMM is highly reliable for the accounts of the frequency and duration of rainy days while APHRODITE tends to overestimate to a larger extent. While expecting datasets to be able to capture peak/extreme values better, results showed low agreement of product datasets capturing extreme events as recorded by the ground gauges. Nevertheless, datasets are useful and effective in determining the frequency of rainy days per month but the monthly rainfall and the exact day that rains are still questionable.