River flow prediction and floodplain mapping using Artificial Neural Networks and GIS
This study attempts to predict river flows and to perform floodplain mapping using Artificial Neural networks and GIS in the lower area of Mitchell catchments in Gippslands, Australia. This area is selected because it experiences serious and frequent floods and causes substantial suffering, loss of...
| Main Author: | |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2006
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/276104/1/4%20ISG%202006%20Flood%20prediction.pdf |
| _version_ | 1826050028500680704 |
|---|---|
| author | Santosa, Purnama Budi |
| author_facet | Santosa, Purnama Budi |
| author_sort | Santosa, Purnama Budi |
| collection | UGM |
| description | This study attempts to predict river flows and to perform floodplain mapping using Artificial Neural networks and GIS in the lower area of Mitchell catchments in Gippslands, Australia. This area is selected because it experiences serious and frequent floods and causes substantial suffering, loss of life and economic damage. In order to be able to map the floodplain, sufficient data such as river flows and DEM of the area are needed. Insufficiency river flow data is often one of the problems faced by researchers in modeling flood. Based on this experience, this research tries to evaluate the use of Artificial Neural Networks (ANNs) for predicting river flows to fill the data insufficiency. This was done by applying supervised learning based on training dataset. Sensitivity analysis was employed to investigate the best ANNs model which yields the best results. Once the network has been trained, it then can be used to predict the missing river flow data. On the other hand, Digital elevation models (DEM) of the study site location were created in ANUDEM through the GIS interpolation algorithms of spot heights, contour lines and river networks, which are obtained from topographic maps at different sources and scales. This DEM were then converted into TIN model as the geometric database for hydraulic flood modeling. Once the insufficient river flow data has been completed, and the TIN data have been created, river flood could be modeled based on 100 year return period flood. The results show that sensitivity analysis employed in this study is useful for defining the ANNs model for predicting river flows. The work resulted in floodplain mapping shows spatial distribution of flood extent and the affected landuse. This model is a good way of predicting flood and can be used for flood prevention, risk assessment and flood management that can strengthen the local authorities in risk management. |
| first_indexed | 2024-03-13T23:56:29Z |
| format | Conference or Workshop Item |
| id | oai:generic.eprints.org:276104 |
| institution | Universiti Gadjah Mada |
| language | English |
| last_indexed | 2024-03-13T23:56:29Z |
| publishDate | 2006 |
| record_format | dspace |
| spelling | oai:generic.eprints.org:2761042020-06-16T01:12:55Z https://repository.ugm.ac.id/276104/ River flow prediction and floodplain mapping using Artificial Neural Networks and GIS Santosa, Purnama Budi Geomatic Engineering This study attempts to predict river flows and to perform floodplain mapping using Artificial Neural networks and GIS in the lower area of Mitchell catchments in Gippslands, Australia. This area is selected because it experiences serious and frequent floods and causes substantial suffering, loss of life and economic damage. In order to be able to map the floodplain, sufficient data such as river flows and DEM of the area are needed. Insufficiency river flow data is often one of the problems faced by researchers in modeling flood. Based on this experience, this research tries to evaluate the use of Artificial Neural Networks (ANNs) for predicting river flows to fill the data insufficiency. This was done by applying supervised learning based on training dataset. Sensitivity analysis was employed to investigate the best ANNs model which yields the best results. Once the network has been trained, it then can be used to predict the missing river flow data. On the other hand, Digital elevation models (DEM) of the study site location were created in ANUDEM through the GIS interpolation algorithms of spot heights, contour lines and river networks, which are obtained from topographic maps at different sources and scales. This DEM were then converted into TIN model as the geometric database for hydraulic flood modeling. Once the insufficient river flow data has been completed, and the TIN data have been created, river flood could be modeled based on 100 year return period flood. The results show that sensitivity analysis employed in this study is useful for defining the ANNs model for predicting river flows. The work resulted in floodplain mapping shows spatial distribution of flood extent and the affected landuse. This model is a good way of predicting flood and can be used for flood prevention, risk assessment and flood management that can strengthen the local authorities in risk management. 2006 Conference or Workshop Item PeerReviewed application/pdf en https://repository.ugm.ac.id/276104/1/4%20ISG%202006%20Flood%20prediction.pdf Santosa, Purnama Budi (2006) River flow prediction and floodplain mapping using Artificial Neural Networks and GIS. In: International Symposium & Exhibition on Geoinformation 2006, 2006, Malaysia. |
| spellingShingle | Geomatic Engineering Santosa, Purnama Budi River flow prediction and floodplain mapping using Artificial Neural Networks and GIS |
| title | River flow prediction and floodplain mapping using Artificial Neural Networks and GIS |
| title_full | River flow prediction and floodplain mapping using Artificial Neural Networks and GIS |
| title_fullStr | River flow prediction and floodplain mapping using Artificial Neural Networks and GIS |
| title_full_unstemmed | River flow prediction and floodplain mapping using Artificial Neural Networks and GIS |
| title_short | River flow prediction and floodplain mapping using Artificial Neural Networks and GIS |
| title_sort | river flow prediction and floodplain mapping using artificial neural networks and gis |
| topic | Geomatic Engineering |
| url | https://repository.ugm.ac.id/276104/1/4%20ISG%202006%20Flood%20prediction.pdf |
| work_keys_str_mv | AT santosapurnamabudi riverflowpredictionandfloodplainmappingusingartificialneuralnetworksandgis |