EARTHQUAKE PREDICTION BY USING GIS

1
EARTHQUAKE PREDICTION BY USING GIS

• Prepared by
• Mansoor Jehangir

-A CRP 514 Term Project-
-Submitted to Dr. Baqer Al-Ramadan-
2
INTRODUCTION
Natural hazards and catastrophes are recurring
phenomenon which affects one or the other part of
the world every now and then. Of all such hazards
the most devastating are the earthquakes.

• Occurrence of earthquakes is very uncertain.
• It can neither be predicted nor forecast.
• ? Causes immense destruction and loss of life.

3
INTRODUCTION

• What is an earthquake and what causes them to
  happen?
• ?Earthquakes occur from the deformation of outer,
  brittle portions of "tectonic plates", the
  earth's outermost layer of crust and upper
  mantle.
• ?Due to the heating and cooling of the rock below
  these plates, causes the adjacently overlying
  plates to move, and, under great stresses,
  deform.
• ?The rates of plate movements range from about 2
  to 12 centimeters per year.

• What is plate tectonics?
• The earth's crust (the outer layer of the planet)
  is made up of several pieces, called plates.
  Plate tectonics is the continual slow movement of
  the tectonic plates.
• What is a fault?
• A fault is a fracture or zone of fractures
  between two blocks of rock.
• Faults allow the blocks to move relative to each
  other.

4
INTRODUCTION
Figure on the left shows the world divided into
different tectonic plates and figure on the right
describes the three plate boundaries
5
Earthquake Prediction

• It may never be possible to predict the exact
  time and location of the earthquake.
• However, when predicting where the next big
  earthquake might occur, seismologists look
  closely at areas where major earthquakes have
  occurred in the past.
• There are several different methods used for
  earthquake prediction
• Statistical probability
• Physical measurements
• Geochemical observations
• Observations of animal behavior

6
Earthquake Prediction

• Statistical Method
• ?By collecting adequate historical data,
  predictions can be made as where and when might a
  big earthquake strike.
• ?Predictions based on statistical analysis of
  historical earthquake data can only be given for
  the long or intermediate time frames.

• Example
• If there have been four magnitude 7.0 or greater
  earthquakes in an area in a 200 year period, then
  this frequency is one in 50 years i.e. if the
  last data used for evaluation was of year 2000,
  then the prediction for earthquake will be on
  2050.
• This term project involves the use of this method
  to predict a future earthquake activity in Japan.

7
PROJECT DETAILS

• Skills Involved
• Navigating through and effectively using a GIS
  software.
• Changing map symbology.
• Manipulating tabular information, including
  selecting attributes, querying
• tables and sorting tables.
• Querying map data.
• Downloading and formatting data from the
  Internet for use in GIS.
• Changing map projections.
• Creating new information from existing data.

8
PROJECT DETAILS
METHODOLOGY
Task 1

• First task was to download data from the site.
• Layers include faults, plate lines, countries
  and world.

View of base map with all features on ?
9
PROJECT DETAILS
METHODOLOGY
Task 2

• This task includes downloading data from the
  internet and exploring it .
• Data was downloaded from the USGS site, which
  provides many different kinds of Earthquakes
  data, involving historical data

10
PROJECT DETAILS
METHODOLOGY
Task 2

• Editing data
• ArcView requires a comma deliminated text file or
  a dbase file to generate points.
• It also expects west longitude and south latitude
  to be negative numbers, with no directional
  designations (N, S, E, or W).
• For plotting the Earthquake data, I have used a
  third party software called ET Geowizard.

• ET Geowizard
• It is a third party software through which data
  from text files, comma delaminated files, or
  dbase files can be generated into Arcview format
  and features like points, lines and polygons can
  be created.
• There are many other useful functions available
  in the software.

11
PROJECT DETAILS
METHODOLOGY
Task 3

• Labeling and symbolizing features
• Colors, fonts, size and positions were assigned
  to the features so that the map can be viewed
  without any difficulty.
• Symbolizing Features
• ? Plate boundary layer was symbolized by
  categorical attributes, using different colors
  for different types.
• Earthquake data layers (points) were also
  categorically symbolized for making the map more
  informative.
• Classifying the earthquake values by magnitude
  (natural brinks) presents a clear picture on the
  map, as to which value of the earthquake is
  larger at any particular location.

12
PROJECT DETAILS
METHODOLOGY
Task 3
13
PROJECT DETAILS
METHODOLOGY
Task 4

• Analyzing Data
• ?Eight different earthquake data layers were
  added.
• ?Layers include earthquake data of the year 1997,
  2000, 2001, 2002, 2003 and 2004.
• ?Two other data sets were included to perform
  historical analyses in the most earthquake prone
  area of the world, Japan.

14
PROJECT DETAILS
METHODOLOGY
Task 4

• What do we notice about the distribution of these
  earthquakes?
• ?By examining the map, we can say that most of
  the earthquakes occur near plate boundaries.
• ?By running a query we can also find out the
  percentage of earthquakes within 100 miles or 200
  of the plate boundaries.
• ?The result of the query shows that about 87 of
  the earthquakes strike within the 200 miles of
  the plate boundary.

15
PROJECT DETAILS
METHODOLOGY
Task 4

• What type of plate boundary is associated with
  earthquakes with a magnitude of 6.0 or greater?
• ?Using features by attributes and running a query
  can point us to the plate boundary which is
  mostly associated with earthquakes having larger
  magnitude.
• ?Therefore the result of the query clearly shows
  that big earthquakes occur most frequently at
  convergent plate boundaries.

16
PROJECT DETAILS
METHODOLOGY
Task 4
?Similar query is performed as before, to find
the percentage of earthquakes striking within 100
and 200 miles of convergent boundaries. ?Results
in the table below shows the number of
earthquakes occurring within 100 and 200 miles of
convergent boundaries.
17
PROJECT DETAILS
METHODOLOGY
Task 4

• Why is Japan, so earthquake prone area?

?Display units were set to decimal minutes
seconds, but for our query, we need to change it
to miles, as shown in figure. ?Next task will be
to zoom to Japan. ?The zoomed frame is captured
as a bookmark for future reference.
18
PROJECT DETAILS
METHODOLOGY
Task 4
Why is Japan, so earthquake prone area?
?Query is run by which convergent boundaries and
country Japan is selected. ?The figure clearly
shows that Japan lies near most active boundary
i.e. Convergent boundaries. ?Using the
measurement tool, we can find the distance from
the plate boundary to the closest large
earthquake. ?Display shows the figure of 174
miles, which proves why Japan is most earthquake
prone country of the world. Since it is already
proved that most EQs occur within 200 miles from
the convergent boundary.
19
PROJECT DETAILS
METHODOLOGY
Task 4
?Next and final task is to predict when the next
big earthquake will strike Japan in
future. ?Using the statistical method a query is
run to find the occurrence of earthquake
(magnitude 6.5 or greater) in the country
Japan. ?This data is from year 1556 till 2003.
20
PROJECT DETAILS
METHODOLOGY
Task 4
?By the query results shown in the previous
slide, the results can be calculated as ?Since
the data is from year 1556 till 2003, means a
(2003 1556) 447 years data. ?In 447 years, 9
big earthquakes have struck Japan. ?With a
frequency of 1 in 50 years (447 / 9 50).
Therefore we can say that in the year 2053
(200350), there is a fair chance of big
earthquake striking the country Japan again.
21
CONCLUSION

• Scientists can not predict the exact location,
  time, or magnitude of an earthquake.
• But by statistical method we can identify
  approximate time, as to when it is more likely
  for a big earthquake to occur sooner rather than
  late.
• All the detail analysis of the data clearly
  shows why Japan is most earthquake prone area of
  the world, and
• By statistical method of prediction of
  earthquake it is mostly likely that in the year
  2053, next big earthquake will strike Japan.

22
THANKYOU
Questions Comments