Assessment of [GIS] spatial interpolation methods in estimating rainfall missing data / Norazimah Hani Ismail
Rainfall is an important data to identify the complete rainfall record at the gauging station. There is an incompleted rainfall data due to various factors such absence of the observer and the instrument failures. Thus, to fill the gaps of missing observation in data, several techniques were used...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2018
|
| Subjects: | |
| Online Access: | https://ir.uitm.edu.my/id/eprint/22447/1/TD_NORAZIMAH%20HANI%20ISMAIL%20AP%20R%2018_5.PDF |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Rainfall is an important data to identify the complete rainfall record at the gauging
station. There is an incompleted rainfall data due to various factors such absence of
the observer and the instrument failures. Thus, to fill the gaps of missing observation
in data, several techniques were used to predict the missing rainfall data. The aim of
this study is to assess GIS spatial interpolation methods in estimating rainfall
missing data using Inverse Distance Weighted (IDW), Thiessen Polygon and Kriging
in Northern region of Malaysia. Next, the objectives of this study are to generate
rainfall spatial interpolation data based on IDW, Thiessen Polygon and Kriging as
well as to assess the accuracy of estimated rainfall values for each spatial
interpolation methods. The research study area focuses only in the Northern Region
of Peninsular Malaysia which is Pulau Pinang, Kedah, Perak and Perils. In this
study, 15 out of 143 rainfall stations with completed rainfall data were estimated with
monthly basis. The most suitable method in accuracy for each methods were
compared based on Root Mean Square Error (RMSE). Overall the best RMSE is
found in IDW on January is (16.691) following by the worst RMSE in Thiessen
Polygon on November is (2233.526). However, the RMSE for Kriging is the most
consistent by annually. The finding of this study shows that Kriging is the most
accurate GIS spatial interpolation method in estimating rainfall missing data. Thus,
Kriging Interpolation is possible to be used to improve the conventional methods of
estimating rainfall missing data. |
|---|