Remote Sensing Method for Flood Management System

International Journal of Modern Research in Engineering & Management (IJMREM)
||Volume|| 1||Issue|| 4 ||Pages|| 30-34 ||April 2018|| ISSN: 2581-4540
www.ijmrem.com IJMREM Page 30
Remote Sensing Method for Flood Management System
1,
Md. Ahbabur Rahman, 2,
Md. Nasir Uddin, 3,
Syeda Abida Khatun
1,2,
Sustainable Energy Management (FEM), Prince of Songkhla University (PSU), Hat-Yai, Songkhla, Thailand.
3,
North East University Bangladesh (Sylhet), Bangladesh Faculty of Humanity & Social Science
----------------------------------------------------ABSTRACT------------------------------------------------------
Flood occurred when heavy and continuous rainfall exceeding the absorptive capacity of soil and the flow
capacity of rivers, streams, and coastal areas. Land areas that are most subjected to floods are areas situated
adjacent to rivers and streams, that are known as floodplain and therefore considered as “flood-prone”. These
areas are hazardous to development activities if the vulnerability of those activities exceeds an acceptable level.
The main objectives of this study are; to identify floodplains and other susceptible areas, and to assess the
extent of disaster impact in the study area which is located at Kota Tinggi, Johor, Malaysia. This area
experienced an unprecedented flood during December of 2006 to January of 2007.Questions such as how often
and how long the floodplain will be covered by water, and at what time of year flooding can be expected need to
be answered. Thus, an understanding of the dynamic nature of floodplains is greatly required. Multi-temporal
Radarsat-1images, Landsat ETM+ image, topographical maps and land use maps were used in this study for
the purpose of delineating the flood extend before, during and after the flood event. DEM acquired from
topographic map is used to derive flood depth. The final outputs of this study are flood extent and flood depth
maps where both of these maps show the impact of the flood to environment, lives and properties. This map is
also important and can be applied to develop a comprehensive relief effort immediately after flooding.
KEYWORDS: Remote Sensing, Flood Mapping, Malaysia, Flood-prone.
---------------------------------------------------------------------------------------------------------------------------------------
Date of Submission: Date, 17 October 2017 Date of Accepted: 22 October
2018
---------------------------------------------------------------------------------------------------------------------------------------
I. INTRODUCTION
The geographic of Malaysia is outside the Pacific Rim of fire and freely from any severe ravages and
destruction caused by natural disasters such as earthquake, typhoons and volcanic eruptions; nevertheless the
country is subjected to monsoon floods, landslides and severe haze episodes.[1] Flood is the most frequent
disaster occurred in Malaysia shown in Figure 1.0.In Malaysia, flood can be divided into two categories:
monsoon flood and flash flood[1]. Flood occurred when heavy and continuous rainfall exceeding the absorptive
capacity of soil and the flow capacity of rivers, streams, and coastal areas. Land areas that are most subjected to
floods are areas situated adjacent to rivers and streams, that are known as floodplain and therefore considered as
“flood -prone”[2]
Figure 1.0: Types of natural disaster in Malaysia.

Remote Sensing Method for Flood Management…
II. STUDY AREA & DATA USED
Kota Tinggi, Johor, Malaysia was chosen as study area in this study. During December of 2006 and January of
2007, an extreme rainfall hits at Kota Tinggi causing an unprecedented flood has been occurred. According to
(DREF Bulletin, 2007; Yau, 2007),[3] the extreme of flood event of the year 2006/2007 at Johor is considered
an effect of global warming. Total flood damages are estimated at RM 1.26 billion, with repercussions on the
lives, livelihoods and businesses of the affected (source: government of Malaysia).[3] Figure 2.0 shows the
location of study area.
The data used in this study are listed at below:
Figure 2.0: Study area
III. IDENTIFICATION OF WATER AREAS
Spectral signatures of water are distinct from other land use like vegetable and soil shown below

Figure 3.0: Typical spectral reflectance’s curve for vegetation, soil and water.
In the Figure 3, water has no reflectance at the approximately wavelength of 0.7 µm and beyond. In the order
words, all infrared wavelengths are suitable for delineating water from other land use features. Baumann (1999)
and Wang et. al. (2002) have successfully use infrared bands in Landsat TM to identify flood-inundated area.
[1]
In calm weather, flood water acts as a specular reflector (or forward scatterer) of the SAR signal, resulting in
very low backscatter values for flooded areas that appear as a dark area in SAR image. Mason et. al. (2009)
stated that wind or rain may cause roughening of the water surface. [4]Therefore, backscatter from the water
may rise to similar or greater levels than the surrounding land. A further complicating factor may be the
presence of emergent vegetation or buildings at the flood edge, leading to substantial increase in backscatter due
to multiple reflections. These factors tend to reduce the accuracy of SAR-derived flood extent maps. [5]
IV. METHODOLOGY
The general idea is that RS provide spatial and temporal data input required by the distributed hydrological
models in order to simulate runoffs and thus floods. RS data in some studies have also been utilized to calibrate
and improve on the performance of distributed hydrological models. RS also provides an option of accessing
information from otherwise physically inaccessible areas. This Figure provide a summary of how RS can be
used in flood management. Flood damage assessment is Knowledge of damage inflicted by flood is required by
the authorities and Insurance companies in order to effect compensation as well as to have an estimate of the
cost of reconstruction.
Fig. 4: Flow chart showing steps on how RS data can be processed for use in flood damage assessment.
V. PRE-PROCESSING FOR RADARSAT IMAGE
1. All pre-processing steps for RADARSAT image include geometric correction, convert digital number (DN)
into backscattering coefficient, rescale 16-bit RADARSAT image into 8 bits image, and image filtering. All
these pre-processing are implementing using PCI Geomatica software. [6]
2. Since Landsat image downloaded at USGS website have been geometrical corrected. Therefore,
RADARSAT image is registered to Landsat image. More ground control points (GCPs) extracted from
Landsat image to rectify raw RADARSAT image to improve the accuracy of geometric correction. [7]
3. Image filtering is used to reduce speckle noise in SAR imagery. Eight different adaptive speckle are process
on SAR imagery and then using the equation 1 to determine the best adaptive speckle.[8] Standard
deviation to mean (STM), to determine the ability to reduce speckle noise of a homogeneous land cover
segment. 𝑆𝑇𝑀 =
𝑆𝐷
𝑀
VI. EXTRACTING THE FLOOD EXTENT FROM SAR IMAGE
Extracting flood extend from SAR image by simple image processing technique: Threshold implies dividing the
histogram into two or more parts. There is a clear distinction between water and surrounding objects on radar
image with backscatter value. Therefore, flooded area is very easily delineated by using thresholding technique.
[2, 8]

According to Sanyal and Lu (2004), change detection is a powerful technique to detect flooded area in SAR
imagery. It is generally performed by acquiring two imageries taken before and after the flood or during the
flood. [9]Normalization have to be processed upon multiple images in order to all images have same
atmospheric condition and improve result of change. One of the advantage of using change detection technique
upon multiple images is we can study spatial and temporal changing of the flood extend.
Extract flood extend from SAR image by visual interpretation usually takes a lot of time, labor and cost. The
method of automatically extracting flood extends from RADARSAT image with the support of LNADSAT TM
has been developed by Yang and Zhou, (2000).[10] The shade and water body from TM imagery were fused
with water body from RADARSAT image with aim at extracting flood extent from RADARSAT imagery. The
flood extent can be accurately, semi-automatically extracted from RADARSAT SAR by the method. [11]
Fig. 5: Flowchart showing the generation of flood-extent maps from Satellite Radar (SAR) image. [5]
VII. RESULT ANALYSIS
Every flood extent map generate by each method is validated by existing flood map produced by Department of
Irrigation and Drainage (DID). Flooded area converts into vector format to overlay on landuse map and DEM
data to produce flood extend and flood depth maps respectively to indentify flood-prone area and study the
impact of the flood event to environment, lives and properties.
VIII. CONCLUSION
From the preceding three sections of this review, quite a number of conclusions can be made. The first one is
that flood hazard is the most widely distributed and devastating disaster, more so in this era of climate change.
Secondly, knowledge of flood types is critical in flood prediction and flood damage assessment and here it is
worth noting that although flash floods have not been very common compared to other types like river floods,
they have caused more deaths and property damage than many other flood types. Flood management is a three-
phase procedure that includes pre-flood, during flood and post-flood activities. These three phases can further be
subdivided into flood prediction, flood prevention and mitigation, flood risk identification and mapping and
flood damage assessment. The lesson we learn from here is that flood management is very diverse and it
requires multidisciplinary involvement. As an example, flood prediction, mapping and damage assessment
require disciplines of hydrology, soil science and geography. Flood prevention, mitigation and flood damage
assessment require efforts from government, insurance companies, professionals and above all the general
public.
REFERENCES
[1] P. Baumann, "Flood analysis: 1993 Mississippi flood," URL: http://rscc. umn.
edu/rscc/Volume4/baumann/baumann. html, 1999.

[2] P. Brivio, R. Colombo, M. Maggi, and R. Tomasoni, "Integration of remote sensing data and GIS for
accurate mapping of flooded areas," International Journal of Remote Sensing, vol. 23, pp. 429-441,
2002.
[3] T. T. Huey, A. L. Ibrahim, M. S. Saayon, and M. Z. A. Rahman, "Remote Sensing Methods for
Mapping Flood-Prone Areas," Asian Association of Remote Sensing, 2010.
[4] T. T. Huey, A. L. Ibrahim, and M. S. Saayon, "Prof., Dr., Eng. Mr., Ms.): Mr."
[5] 歐靚芸, "結合聚類法與類神經網路發展颱風淹水預警系統," 臺灣大學土木工程學研究所學位論
文, pp. 1-91, 2012.
[6] P. W. M. Souza Filho and W. R. Paradella, "Use of RADARSAT-1 fine mode and Landsat-5 TM
selective principal component analysis for geomorphological mapping in a macrotidal mangrove coast
in the Amazon Region," Canadian Journal of Remote Sensing, vol. 31, pp. 214-224, 2005.
[7] T. Toutin, "Review article: Geometric processing of remote sensing images: models, algorithms and
methods," International Journal of Remote Sensing, vol. 25, pp. 1893-1924, 2004.
[8] A. Lopes, E. Nezry, R. Touzi, and H. Laur, "Structure detection and statistical adaptive speckle
filtering in SAR images," International Journal of Remote Sensing, vol. 14, pp. 1735-1758, 1993.
[9] J. Sanyal and X. Lu, "Application of remote sensing in flood management with special reference to
monsoon Asia: a review," Natural Hazards, vol. 33, pp. 283-301, 2004.
[10] C. Zhou, J. Luo, C. Yang, B. Li, and S. Wang, "Flood monitoring using multi-temporal AVHRR and
RADARSAT imagery," Photogrammetric engineering and remote sensing, vol. 66, pp. 633-638, 2000.
[11] A. Demirkesen, F. Evrendilek, S. Berberoglu, and S. Kilic, "Coastal flood risk analysis using Landsat-7
ETM+ imagery and SRTM DEM: A case study of Izmir, Turkey," Environmental monitoring and
assessment, vol. 131, pp. 293-300, 2007.

Remote Sensing Method for Flood Management System

More Related Content

What's hot (20)

Similar to Remote Sensing Method for Flood Management System (20)

More from IJMREMJournal (20)

Recently uploaded (20)

Remote Sensing Method for Flood Management System