DISASTER MANAGEMENT

Modern technology is improving in many areas, particularly in science and

technology, yet these advancements also present numerous obstacles. Although we
can copy and artificially adapt things using technology, development is less
significant than people and other living things. The likelihood of disasters is also
increased by technological advancements. Depending on a number of variables, a
disaster may be either natural or man-made, or it may be an accident. Climate
change brought on by human activity, including soil, water, and air pollution, is the
main contributing element. The use of Geographic Information System (GIS)
technology is crucial in the fight against such catastrophes. The seamless integration
of this technology at all phases of the disaster management cycle, from mitigation to
recovery, is made possible by the combination of geographic information systems
with remote sensing and photogrammetry technologies.
Geographic Information Systems (GIS) are data processing systems that
store, alter, analyse, integrate, and display data. Data from thousands of sources,
including the population's age, gender, and statistics on utility services as well as
information on the state's whole landmass, can be used to predict disasters.
Natural catastrophes are risky occurrences brought on by earth's natural
processes. Floods, earthquakes, hurricanes, cyclones, volcanic eruptions, tsunamis,
and other geological processes are examples of natural disasters. Disasters that are
anthropogenic or man-made are brought on by human activity. Disasters caused by
human activity have a negative impact on people, other living things, and ultimately
ecosystems. All forms of pollution, nuclear disasters, chemical disasters, biological
disasters, terrorist attacks, and other unintentional calamities are examples of man-
made catastrophes.
In Malaysia, landslides are frequent natural disasters that happen on hillsides
and steep slopes. When government agencies and rescue teams exhibit vulnerability
maps and notify the Risk time based on climates, it will raise society's awareness
because historical data can be obtained using GIS technology.
Through the use of geographical linkages, GIS also combines qualitative and
quantitative data. Other top-notch capabilities like the query builder, overlay analysis,
raster & vector analysis, and user interfaces enable the query-based analysis to be
performed in different thematic levels.
The development of several theme maps, such as elevation, slope, aspect,
and hill shade using terrain analyzer, which is particularly helpful in predicting
landslides and avalanches, is an example of spatial analysis.
A vast database is needed to solve common problems like mapping
landslides. Users can more clearly comprehend and visualise issues by conducting
further analysis on GIS results.
1.1 Types of studies that can be conducted related to the application

The slope, aspect, geology, structures, land use/land cover, road network,
slope curvature, drainage density, lineament density, and many other pieces of
information are needed to determine the risk of landslides. These data, which are
kept in a GIS database for analysis, are gathered from field observations, digital
data, remote sensing, and already-existing maps.
The socio-economic statistics, such as population, economic situation,
educational level, level of disaster awareness, etc., that collectively indicate their
coping capacity are what are needed for vulnerability mapping. The risk map,
however, takes into account both social and technological factors, making it useful
for prioritising the areas for measures to reduce landslide risk. Each bigger landslide
that has the potential to disrupt infrastructure, settlements, or agricultural areas
should be thoroughly researched because each one should be seen as being unique
in nature.

1.2 Possible data that can be used and the source of data

Landslide is the disaster we select for this disaster management. The state of
Selangor is where this study was carried out. Since we know that landslides must
harm that state, one of our study's goals is to identify regions that are susceptible to
them.
As an appropriate pilot area to assess the frequency and distribution of
landslides, the eastern region of Selangor state was chosen. This region has seen
significant landslide damage as a result of heavy rainfall. The Malaysian peninsula's
southwest coast is where the Selangor region is situated. State of Perak, Pahang,
and Negeri Sembilian, Melaka, form its northern and eastern and southern
boundaries, respectively. One of the 13 states that make up the Federation of
Malaysia is Selangor. There are roughly 3 million people living there.
 For probabilistic landslide hazard analysis, precise location detection of
landslides is crucial. The utilisation of remote sensing data, such as satellite photos
and aerial photographs, is beneficial to obtaining comprehensive and affordable
knowledge on landslides. Aerial images taken at a scale of 1:25,000 were used in
this investigation to locate the landslides. The landslides were found through picture
interpretation, and the locations were confirmed by fieldwork. These images were
taken between 1981 and 2000. Breaks in the forest canopy, bare soil, or other
geomorphic features like head and side scarps, flow tracks, and soil and debris
deposits below a scar were used to identify recent landslides. A total of 327
landslides were mapped in a mapped area of 182,406,000 km2 in order to compile a
database to evaluate the surface area and quantity of landslides in the study area.
The approach utilised in this work and many others is predicated on the
fundamental tenet that the probability of a slope stability issue is equivalent to the
actual frequency of landslides. Lineament, land cover, and vegetation index values
were extracted from Landsat TM satellite images, and topographical and lithological
databases were built. Analyses of meteorological data were used to obtain the
distribution of precipitation. Then, in ARC/INFO, the calculated and extracted factors
were transformed into a 1010 m grid. Frequency ratio and logistic regression models
were used to investigate the geographical correlations between the site of the
landslide and each component associated with it. The relationship was used as a
rating for each factor in the overlay analysis using the frequency ratio. Based on the
associations, a formula predicting the likelihood of a landslide occurring was
developed using logistic regression. The landslide hazard index was created using
this formula, and the index was then used to map the landslide hazard. Finally,
success rates were calculated after the maps were cross-checked and compared
using known landslide locations.
A spatial database that takes into account landslide-related characteristics
was created in order to use the probabilistic technique. In Malaysia, this statistics are
accessible in hard copy or as digital maps. Table 1 displays the spatial database that
was created. The likelihood was determined by taking into account ten elements that
were retrieved from the created geographic database. In a GIS setting, they were
converted into a vector-type spatial database. Additionally, data on landslide-related
elements was taken from the database. From the topographic database, a digital
elevation model (DEM) was first produced. A DEM with a resolution of 10 m was
created using contour and survey-based points with elevation values from the
1:25,000-scale topographic maps.
 Slope angle, slope aspect, and slope curvature were computed using the
DEM. Positive curvatures represent convexity, zero curvatures represent flat
surfaces, and negative curvatures represent concavity. The Environmental Systems
Research Institute (ESRI) routine in Arc View was used to create the curvature map.
Additionally, the topography database was used to compute the separation from
drainage. In 1-m intervals, the drainage buffer was determined. The lithology was
taken from the geology database, and the distance from the lineament was
determined. A geological map at a scale of 1:250,000 was used to create the
lithology map. In 1-m intervals, the lineament buffer was determined. A soil map at a
scale of 1:250,000 was used to create the soil map.
A LANDSAT TM image was used to classify land cover data using an
unsupervised classification algorithm and topographic map. The land cover map was
divided into six categories for land-cover mapping: dense forest area, barren land,
agriculture, rubber, residential area (concrete), sparse forest area, and residential
area (non-concrete). Finally, LANDSAT TM satellite pictures were used to create the
NDVI map. The equation NDVI=(IRR)/(IR+R) was used to obtain the NDVI value,
where IR stands for the infrared region of the electromagnetic spectrum and R
stands for the red portion. Areas of vegetation in an image are indicated by the NDVI
value. Utilizing data from meteorological stations across the entire Peninsula of
Malaysia during the previous 20 years, precipitation data was interpolated.
The logistic regression and frequency ratio models were applied to the factors
using a raster grid of 1010 m cells. The area grid measured 14,140 rows by 12,900
columns, or 182,406,000 rows and columns. 327 cells had landslides.
The training area's frequency ratio data for each element were added to
determine the Landslide Hazard Index (HSI). The relative risk of a landslide
occurring is shown by the landslide hazard value. As a result, the risk of a landslide
occurring increases with increasing value and decreases with decreasing value.
(HSI: Landslide Hazard Index; Fr: rating of each factor’s type or range)The landslide hazard
map was made using the HSI values and for interpretation is shown.