Cover Image

Hybrid approach on multi- spatiotemporal data framework towards analysis of long-lead upstream flood: a case of Niger State, Nigeria

Floods have become a global concern because of the vast economic and ecological havoc that ensue. Thus, a flood risk mitigation strategy is used to reduce flood-related consequences by a long-lead identification of its occurrence. A wide range of causative factors, including the adoption of hybrid m...

Full description

Saved in:
Bibliographic Details
Main Author: Ndanusa, Ahmed Babalaji
Format: Thesis
Language:English
English
Published: 2019
Subjects:
HA Statistics
Online Access:https://etd.uum.edu.my/8120/1/s901197_01.pdf
https://etd.uum.edu.my/8120/2/s901197_02.pdf
https://etd.uum.edu.my/8120/
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.uum.etd.8120
record_format eprints
spelling my.uum.etd.81202022-05-09T07:14:11Z https://etd.uum.edu.my/8120/ Hybrid approach on multi- spatiotemporal data framework towards analysis of long-lead upstream flood: a case of Niger State, Nigeria Ndanusa, Ahmed Babalaji HA Statistics Floods have become a global concern because of the vast economic and ecological havoc that ensue. Thus, a flood risk mitigation strategy is used to reduce flood-related consequences by a long-lead identification of its occurrence. A wide range of causative factors, including the adoption of hybrid multi-spatiotemporal data framework is considered in implementing the strategy. Besides the structural or homogenous non-structural factors, the adoption of various Information Systems-based tools are also required to accurately analyse the multiple natural causative factors. Essentially, this was needed to address the inaccurate flood vulnerability classifications and short time of flood prediction. Thus, this study proposes a framework named: Hybrid Multi-spatiotemporal data Framework for Long-lead Upstream Flood Analysis (HyM-SLUFA) to provide a new dimension on flood vulnerability studies by uncovering the influence of multiple factors derived from topography, hydrology, vegetal and precipitation features towards regional flood vulnerability classification and long-lead analysis. In developing the proposed framework, the spatial images were geometrically and radiometrically corrected with the aid of Quantum Geographic Information System (QGIS). The temporal data were cleaned by means of winsorization methods using STATA statistical tool. The hybrid segment of the framework classifies flood vulnerability and performs long-lead analysis. The classification and analysis were conducted using the corrected spatial images to acquire better understanding on the interaction between the extracted features and rainfall in inducing flood as well as producing various regional flood vulnerabilities within the study area. Additionally, with the aid of regression technique, precipitation and water level data were used to perform long-lead flood analysis to provide a foresight of any potential flooding event in order to take proactive measures. As to confirm the reliability and validity of the proposed framework, an accuracy assessment was conducted on the outputs of the data. This study found the influence of various Flood Causative Factors (FCFs) used in the developed HyM-SLUFA framework, by revealing the spatial disparity indicating that the slope of a region shows a more accurate level of flood vulnerability compared to other FCFs, which generally causes severe upstream floods when there is low volume of precipitation within regions of low slope degree. Theoretically, the HyM-SLUFA will serve as a guide that can be adopted or adapted for similar studies. Especially, by considering the trend of precipitation and the pattern of flood vulnerability classifications depicted by various FCFs. These classifications will determine the kind(s) of policies that will be implemented in town planning, and the Flood Inducible Precipitation Volumes can provide a foresight of any potential flooding event in order to take practical proactive measures by the local authority. 2019 Thesis NonPeerReviewed text en https://etd.uum.edu.my/8120/1/s901197_01.pdf text en https://etd.uum.edu.my/8120/2/s901197_02.pdf Ndanusa, Ahmed Babalaji (2019) Hybrid approach on multi- spatiotemporal data framework towards analysis of long-lead upstream flood: a case of Niger State, Nigeria. Doctoral thesis, Universiti Utara Malaysia.
institution Universiti Utara Malaysia
building UUM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Utara Malaysia
content_source UUM Electronic Theses
url_provider http://etd.uum.edu.my/
language English
English
topic HA Statistics
spellingShingle HA Statistics
Ndanusa, Ahmed Babalaji
Hybrid approach on multi- spatiotemporal data framework towards analysis of long-lead upstream flood: a case of Niger State, Nigeria
description Floods have become a global concern because of the vast economic and ecological havoc that ensue. Thus, a flood risk mitigation strategy is used to reduce flood-related consequences by a long-lead identification of its occurrence. A wide range of causative factors, including the adoption of hybrid multi-spatiotemporal data framework is considered in implementing the strategy. Besides the structural or homogenous non-structural factors, the adoption of various Information Systems-based tools are also required to accurately analyse the multiple natural causative factors. Essentially, this was needed to address the inaccurate flood vulnerability classifications and short time of flood prediction. Thus, this study proposes a framework named: Hybrid Multi-spatiotemporal data Framework for Long-lead Upstream Flood Analysis (HyM-SLUFA) to provide a new dimension on flood vulnerability studies by uncovering the influence of multiple factors derived from topography, hydrology, vegetal and precipitation features towards regional flood vulnerability classification and long-lead analysis. In developing the proposed framework, the spatial images were geometrically and radiometrically corrected with the aid of Quantum Geographic Information System (QGIS). The temporal data were cleaned by means of winsorization methods using STATA statistical tool. The hybrid segment of the framework classifies flood vulnerability and performs long-lead analysis. The classification and analysis were conducted using the corrected spatial images to acquire better understanding on the interaction between the extracted features and rainfall in inducing flood as well as producing various regional flood vulnerabilities within the study area. Additionally, with the aid of regression technique, precipitation and water level data were used to perform long-lead flood analysis to provide a foresight of any potential flooding event in order to take proactive measures. As to confirm the reliability and validity of the proposed framework, an accuracy assessment was conducted on the outputs of the data. This study found the influence of various Flood Causative Factors (FCFs) used in the developed HyM-SLUFA framework, by revealing the spatial disparity indicating that the slope of a region shows a more accurate level of flood vulnerability compared to other FCFs, which generally causes severe upstream floods when there is low volume of precipitation within regions of low slope degree. Theoretically, the HyM-SLUFA will serve as a guide that can be adopted or adapted for similar studies. Especially, by considering the trend of precipitation and the pattern of flood vulnerability classifications depicted by various FCFs. These classifications will determine the kind(s) of policies that will be implemented in town planning, and the Flood Inducible Precipitation Volumes can provide a foresight of any potential flooding event in order to take practical proactive measures by the local authority.
format Thesis
author Ndanusa, Ahmed Babalaji
author_facet Ndanusa, Ahmed Babalaji
author_sort Ndanusa, Ahmed Babalaji
title Hybrid approach on multi- spatiotemporal data framework towards analysis of long-lead upstream flood: a case of Niger State, Nigeria
title_short Hybrid approach on multi- spatiotemporal data framework towards analysis of long-lead upstream flood: a case of Niger State, Nigeria
title_full Hybrid approach on multi- spatiotemporal data framework towards analysis of long-lead upstream flood: a case of Niger State, Nigeria
title_fullStr Hybrid approach on multi- spatiotemporal data framework towards analysis of long-lead upstream flood: a case of Niger State, Nigeria
title_full_unstemmed Hybrid approach on multi- spatiotemporal data framework towards analysis of long-lead upstream flood: a case of Niger State, Nigeria
title_sort hybrid approach on multi- spatiotemporal data framework towards analysis of long-lead upstream flood: a case of niger state, nigeria
publishDate 2019
url https://etd.uum.edu.my/8120/1/s901197_01.pdf
https://etd.uum.edu.my/8120/2/s901197_02.pdf
https://etd.uum.edu.my/8120/
_version_ 1732947714440691712
score 13.211869

Similar Items