CHAPTER 1

INTRODUCTION

1.1GENERAL

Emergencies occur at unpredictable places at unexpected time. During an emergency, it is

critical to have the right data at the right time, displayed logically, to respond and take

appropriate action. Emergency workers must gain access to a number of department and

their unique data. Most emergencies do not allow time to gather these resources. This

results in emergency responders having to guess, estimate, or make decisions with or

without adequate information. This costs time, money, and in some cases, lives. To prevent

the hazard from becoming a disaster, proper emergency management plans are needed.

Emergency management is the continuous process which deals with the mitigation,

preparedness, response, and recovery phases of disasters. It is possible to minimize the

potential risk by developing early disaster warning strategies, preparing and implementing

evacuation plans and by proper rehabilitation. The main objective of emergency

management is to bring down the extent of hazard by well-defined planning, response and

recovery procedures (Bhanumurthy et al., 2015).

For years, GIS has been an important tool for improving situational awareness,

especially in emergency preparedness and response. It requires the disaster-specific data

along with base layers for data interpretation and analysis. Thematic and infrastructure

1
layers are used for relief operations. The ability of GIS to collect, sort and analyze the

information of the affected areas and presenting the information in the form of interactive

maps has become important source for taking effective decisions for any stakeholder.

Web GIS provides a mechanism to centralize and display critical information

regarding an emergency. Web GIS unlocks the power of GIS to a wider audience who need

not have any technical knowledge. Information conveyed through a Web GIS platform is

easy to interpret than static maps. Moreover, preparing and distributing static maps and

reports for emergency management is a tiresome process. Web GIS can be used to provide

information regarding the disaster and data required for disaster management to all the

stake holders easily. It enables them to take necessary measures and be prepared for dealing

with the emergency situation when it occurs. Data modification and addition done by the

concerned department on the web GIS platform will be available to stake holders without

any delay.

Jairaj and Nair (2017) have conducted a dam break study and emergency action

plan preparation for the three dams in Pampa river basin. In that work an effort has been

taken to estimate the severity of flooding and inundation areas likely to be affected by

Kakki, Pampa and Anathode Dams in an emergency condition. They determined the area

that will be inundated, the depth and velocity of the flood waters and the length of time that

the area will be inundated. The disaster specific data required for the present study are

taken from that report. In the present study, a Web GIS enabled Emergency Information

System (EIS) was developed for some selected Panchayaths in Pampa river basin.

2
1.2 OBJECTIVES

The specific objectives of this study are:

1) To rank the temporary shelters available in the study area based on their site suitability.

2) To prepare an evacuation plan in the event of a dam break flood for study area.

3) To develop a Web GIS based user interface to convey the required information to stake

holders.

During unexpected emergency situations, where there are no existing shelters,

evacuation is done to local facilities which serve as temporary shelters. The temporary

shelters should satisfy certain criteria to be deemed suitable. Prioritizing the shelters based

on their suitability prior to an emergency save a lot of time during evacuation and it helps

the authorities to attend to the various needs of evacuees during their temporary settlement.

A detailed evacuation plan has to be developed prior to the emergency to avoid confusion

during actual evacuation. The emergency information system developed in this work deals

with shelter suitability classification and preparation of evacuation plan for the flood

inundated area.

1.3 SUMMARY

In this project a model of the Emergency information system is prepared for selected

panchayaths in Pampa River basin. This prototype mainly deals with the evacuation plan

for a dam break flood inundation. Emergency Action Plan for Dam break hazard is not

deliberated in this study. The necessity and objectives of the study are explained in chapter

1. A brief summary of the reviewed literature are given in chapter 2. Chapter 3 explains the

3
theoretical background of this study. Methodology of this study and data used are presented

in chapter 4. Results of the study and brief discussion on them are included in chapter 5.

Concluding points of this study are summarized in last chapter.

4
 CHAPTER 2

LITERATURE REVIEW

2.1 GENERAL

This chapter gives an account of literature relevant to the topic of the present study. Web

GIS is used in situations where the work has to be made available to a wider audience who

have no software expertise. Web GIS enabled emergency information system is used in real

life situations all over the world. There are different ways to develop a web GIS interface

and some of them are discussed in the following literature.

2.2 REVIEW OF LITERATURE

Hodgson and Kar (2008) developed a GIS-Based Model to determine site suitability of

Emergency evacuation shelters for hurricane vulnerability in Florida. The purpose of their

study was to rank the existing and candidate shelters (schools, colleges, churches and

community centers) available in the state based on their site suitability. A Geographic

Information System-based suitability model integrating Weighted Linear Combination

(WLC) with a Pass/Fail screening technique was implemented. Factor weights determined

by a focus group were implemented for existing/candidate shelters to help in emergency

service planning. Proximity to evacuation routes, health centers, hazard zone and nature of

population distribution are the parameters considered. Candidate shelter, based on its

physical suitability score are classified as low, medium and high.

5
Karnatak et al. (2014) developed an internet GIS domain which could identify and map

potential biodiversity rich areas in India. The spatial data generated during the study is

organized using PostgreSQL and POSTGIS. The spatial queries and analysis with vector

data is achieved through SQL queries on POSTGIS and WFS-T operations. The vector data

stored in PostgreSQL is accessed through POSTGIS and presented as WMS and WFS

using GeoServer. The development of geoprocessing engine for raster data analysis in Web

GIS environment is using python, PostGIS and PHP. These GIS viewers are developed

using OpenLayers API and PHP. The rich GUI is developed using GeoEXT.

Singh and Singh (2014) This paper presents the integration of open-source geospatial tools

and web technology to visualize and interact with spatial data using web browser. This

paper is aimed to provide a method for building a new web-based spatial visualization

prototype by using a combination of open-source geospatial packages and Microsoft .NET

platform. The city of Taj, Agra in India is selected as a pilot area to create the prototype

spatial mapping system. All the layers of study area are stored in PostgreSQL database.

MapServer is used as the web map server. MapServer can be used in two ways, with the

CGI interface or with the MapScript. This prototype system adopts a server side

architecture where most of the processing takes place on the server side.

Agarwal and Gupta (2014) compared two Web GIS frameworks. In first Web GIS

framework, WAMP server, ALOV, Quantum GIS and MySQL have been used while in

second Web GIS framework, Apache Tomcat server, GeoServer, Quantum GIS,

PostgreSQL and PostGIS have been used. The objective of the paper is to compare the Web

GIS framework based on WAMP server with the one based on Apache Tomcat server. Both
6
frameworks are developed using Thin-Client architecture because of the added advantages

like ease in updating data, negligible resource requirement at client side etc. In the

framework based on WAMP, presentation tier uses PHP, logical tier consists of WAMP

server and ALOV while data tier composed of MySQL. In framework based on Apache

Tomcat, presentation tier used servlet and JSP. Its logical tier consists of Apache Tomcat

server and GeoServer while data tier composed of PostgreSQL and PostGIS. Both the

frameworks are compared based on the supported operating system, development platform,

GIS Server and database support. From the comparison it is clear that the framework based

on Apache Tomcat is more preferable than the framework based on WAMP.

Hwang et al. (2015) developed a Web-based system for shelter and evacuation path

selection. It identifies suitable shelters and provides the shortest evacuation path to each

shelter for civilians who are within the danger zone. Shelters are identified by considering

proximity to health care facilities and to major road, and coverage of each shelter in danger

zone. Candidate shelters include existing schools, colleges, churches, and community

centers. Score for each shelter is found by weighted overlay by assigning factor ratings and

weightage to each parameter. Evacuation path model is prepared using network analysis by

taking time and distance as constraints. Shortest route is calculated from each evacuation

point to shelter facility. Then Web and mobile applications to generate evacuation shelters

and paths based on OWS (Open Geospatial Consortium Web services) architecture is

developed.

Bhanumurthy et al. (2015) developed a framework for addressing emergency

management by integrating various multidisciplinary technologies. Using geospatial

7
technologies such as Remote Sensing, GIS and GPS platform providing near/real time

information is developed. Major challenges in the emergency response such as damage

assessment, situation monitoring, evacuation planning, information dissemination, and

coordination of resource and logistic planning are considered in their work. Shelters are

located based on the incident locations and optimum routing. The network analysis tool is

used to carry out a shortest path from a source to destination using geospatial routing

algorithm.

Gohain et al. (2015) mapped the Infrastructure information of North Eastern Hill

University (NEHU) campus and integrated this information to web based GIS Platform.

They have used Google Earth Imagery to create layers using Quantum GIS Software,

GeoServer as web GIS server, PostgreSQL/PostGIS as Object Oriented Relational

Database management system (ORDBMS). Interface is designed using different

technologies like Web designing language HTML and scripting language JavaScript, CSS,

and OpenLayers API. Data serving is done using WMS/WFS protocol. These Layers are

then overlays on top of Google maps. The web Based interface designed consists of

different components which include overlays option for addition of spatial layers and Base

maps option for flipping of different Google Maps.

Bandari et al. (2016) generated a wasteland information system for the entire state of

Uttarakhand using GIS and Remote Sensing techniques. Free open source Web GIS

systems like GeoServer which is a web GIS server and PostgreSQL/PostGIS as an

ORDBMS is used in this project. User interface is developed using HTML, CSS,

8
Javascript, OpenLayers API and PHP. District wise wasteland information and overall state

information can be accessed based on user needs.

2.3 SUMMARY

From these reviews it is clear that, GIS has emerged as one of the proven assets in the

emergency management. Above discussed literature provides different multidisciplinary

technologies for addressing emergency management. Free open source Web GIS systems

are used in most works and it proves that such systems are sufficient to handle any study in

the field of Web GIS. It is found that client side architecture is preferred to server side

architecture when dealing with moderate works. Scripting language used to develop

interface differ from literature to literature. None of the previous work had explored the

possibility of using XAMPP as a web server along with GeoServer. Here OpenGeo Suite

bundle is used instead of installing separate software's. GeoExplorer, which is a map

composer package in OpenGeo Suite, is also not used in previous literature's to publish the

maps in GUI (graphical user interface).

9
 CHAPTER 3

THEORETICAL BACKGROUND

3.1 GENERAL

In the present study a Web GIS enabled emergency information system is developed for

dam break flood in selected panchayaths of Pampa river basin. An effort is made to classify

temporary shelters in the study area based on their site suitability. Analytic Hierarchy

Process (AHP) is used for assigning weights to each parameter used for shelter selection

and site suitability is found through Weighted Linear Combination (WLC) method.

Evacuation routing is done using Network Analysis. Web GIS interface is developed using

GeoServer as the GIS server, XAMPP as web server and PostgreSQL as the database.

Emergency management system, methods used for shelter classification, Network Analysis

etc. are explained in this chapter. Brief details about different software used in this project

are also given.

3.2 EMERGENCY MANAGEMENT SYSTEM

Emergency management involves preparing for, responding to and recovering from a

disaster or emergency. Emergency planning should aim to prevent emergencies from

occurring, and failing that, it should develop a good action plan to result and effect of

emergencies.

Typical emergency management involves four phases:

10
i. Mitigation – It includes actions taken to prevent hazards from developing into

disasters. Mitigation activities can be structural or non-structural.

ii. Preparedness – It includes plans and preparations to be implemented in response and

recovery phase. Preparedness activities takes place before an emergency occur.

iii. Response – Response phase includes actions taken during the time of disaster. This

includes providing the evacuation planning, medical assistance, identifying the relief

shelters, dispatching the relief material and required food items etc.

iv. Recovery – Recovery actions takes place after an emergency. It includes providing

assistance to affected people, rebuilding damaged property, providing financial

assistance and re-employment to the survivors etc.

Although considered of low risk, incidents with dams may cause significant damage

both directly and indirectly. Catastrophic flooding occurs when a dam fails and the

impounded water escapes through the breach to cause death and destruction in the

downstream valley. The magnitude of the flow greatly exceeds all previous floods and the

response time available for warning the populace is much shorter than for precipitation-

runoff floods occurred in that valley (Lariyah et al., 2013). Another distinguishing

characteristic of dam break flood is the extremely short time for arrival of flood wave.

Velocity of flood wave is considerable high in the downstream area, so affected population

has to be evacuated to safe places before the arrival of flood wave.

A detailed Emergency Action Plan (EAP) includes actions to be implemented and

procedures to be followed during the four phases of emergency management. It also

provides information regarding the chain of command of authorities in dealing with the

11
emergency. Action Plan addresses the type of actions expected of different employees and

when and where mock drills are done. EAP should specify resource mobilization of

machinery and materials, management of evacuation camps, address and contact number of

conveyance vehicles etc. The emergency information system developed in this work deals

with shelter suitability classification and preparation of evacuation plan for the flood

inundated area.

During a disaster the responsibility of evacuation lies with the District Disaster

Management Agency (DDMA) of the respective districts. Police department, Fire and

rescue department, District Magistrate/Collector, Grama Panchayath etc. will co-operate

together to minimize the effect of disaster. Information is passed to people by Grama

Panchayath or Municipalities through elected representatives at ward level.

3.2.1 Evacuation Shelters

Shelters are facilities in which rescues can find health assistance, food, and safety. The

location of these facilities depends on the location of other facilities (e.g., hospitals, fire

stations), the road network and the evolution of the flood (water depths and speeds). Shelter

is a critical determinant for survival in the initial stages of a disaster. Beyond survival,

shelter is necessary to provide security and personal safety, protection from the climate and

enhanced resistance to ill health and disease.

Locational suitability of shelters includes the following factors:

1. Should be far away from hazard area.

2. Should be nearer to health care facilities.

12
3. Should be located in a built up area so that water, electricity and proper sewage

disposal methods are available.

4. Should be located in an easily accessible area.

5. Should be able to provide coverage to maximum evacuation area etc.

The location of temporary shelter also should consider the proximity of transport

facilities for the delivery of humanitarian assistance, proximity to local primary storage and

to food distribution sites (Nappi and Souza, 2015). There are many other social, economic

and cultural factors which determine the shelter selection for temporary settlement during

an evacuation.

3.3 ANALYTIC HIERARCHY PROCESS (AHP)

Analytic Hierarchy Process (AHP) was developed by Thomas Saaty to calculate the needed

weighing factors with the help of preference matrix. In the first step, pair-wise comparisons

were carried out for all factors based on their importance level. In the construction of a

pair-wise comparison matrix, each factor was rated against every other factor by assigning

a relative dominant value between 1 and 9 to the cell concerned. A normalized matrix was

computed by dividing each of the columns by the corresponding sum. As the last step, the

average values of each row were computed and these were used as weights for the

suitability classification of shelters. In order to check whether the matrix judgments were

generated randomly, an index of consistency known as a consistency ratio (CR) was found

out. Generally, a consistency ratio (CR) of the value of 0.10 or less was considered

relevant.

13
3.4 WEIGHTED LINEAR COMBINATION (WLC)

WLC is a ranking method most commonly used in GIS-based studies such as site

suitability, site selection, and resource evaluation analysis. ArcGIS software uses a

‘weighted overlay’ tool to perform WLC analysis. It combines multiple raster inputs,

representing multiple factors, of different weights or relative importance to create an

integrated analysis. It is one of common methodologies used for site selection in general.

Score for each site is found using equation 3.1.

𝑆𝑐𝑜𝑟𝑒 = ∑𝑛𝑗=1(𝐹𝑅𝐽 ∗ 𝑊𝑗 ) (3.1)

Where

Score = suitability score for the location.

FRj= Factor Rating for intervals

j,n =number of parameters

Wj= weight assigned to factor j.

3.5 NETWORK ALALYSIS

A network is a system of linear features that has the appropriate attributes for the flow of

objects. A network is typically topology based: lines (arcs or links) meet at intersections

(junctions or nodes), lines cannot have gaps and lines have directions. Links are the basic

geometric features of a network. A link refers to a road segment defined by two end points.

A junction refers to a street intersection. A road system is a familiar network. Putting

together a road network involves three tasks: gather the linear features of network build the

necessary topology for the network and attribute the network features.

14
 i. Attributing the Network features

Network attribute include the link impedance, one-way streets, speed limits and turn

impedance. Link impedance is the cost of traversing a link. A simple measure of the cost is

the physical length of the link. But the length may not be a reliable measure of cost,

especially in cities where speed limits and traffic conditions vary significantly along

different streets. A better measure of link impedance is the travel time estimated from the

length and speed limit of a link. A turn is a transition from one street segment to another at

a junction. Turn impedance is the time it takes to complete a turn, which is significant in a

congested street network.

ii. Building Topology

Topology rules define the spatial relationship between features which are permissible. The

rules that are defined for a topology control the relationships between features within a

feature class and between features in different feature class. Topology has two main

advantage, assurance of data quality and enhancement of GIS analysis. The topological

relationships enable us to detect errors such as lines that do not meet correctly or polygons

that are not closed properly. These kinds of errors must be corrected to avoid incomplete

features and to ensure data integrity. For example, a shortest path analysis requires the

roads to meet correctly. If a gap exists on a supposedly continuous road, the analysis will

take a circuitous route to avoid the gap. Geo-database data model has introduced a set of 25

topology rules. Some of these rules apply to two or more datasets. The common type of

topological error occurring in a network data set are :

15
a. Dangle nodes: Dangle nodes occur when the end node on a feature is not connected to

a node on another feature. They often indicate that a feature is not correctly snapped to

other features.

b. Pseudo nodes: pseudo nodes occur when the end node of a feature connects to another

end node with no other nodes present. They often indicate that the two features can be

merged.

c. Multiple end nodes: They occur at intersection where three or more features meet.

3.6 WEB GIS ARCHITECTURE

Web GIS extends a basic Web application by giving it GIS capabilities. The basic

architecture of Web GIS is similar to Web applications but with the addition of GIS

components (due to voluminous amount of data). Web GIS consists of three layers namely

user interface layer, application layer and database layer (Fig. 3.1). A user uses a Web GIS

application through a client, which can be a Web browser, a desktop application or a

mobile application. The client sends a request to the Web server over the Internet via HTTP

(Hyper Text Transfer Protocol). The Web server forwards GIS related requests to the GIS

server. The GIS server retrieves the needed data from the GIS database and processes the

request, which can be to generate a map, conduct a query or perform an analysis. The data,

map or other result is sent by the Web server to the client in a response via HTTP. The

client then displays the result to the user, which completes the request and response cycle

(Fu and Sun, 2011). Web GIS applications can be categorized as either thin client

architecture or thick client architecture depending on how the work load is distributed

between client and server.

16
 Fig. 3.1: Basic architecture of Web GIS (Fu and Sun, 2011).

i. Thin Client

Thin client architecture relies on the server to perform most of the work, leaving the

client to do the least amount. The client simply sends the user’s requests to the server.

The server does the processing, such as generating a map and performing analysis. The

results, typically in HTML format embedded with GIF, PNG, or JPEG images are then

returned to the client and displayed for the user (Fu and Sun, 2011).

ii. Thick Client

Thick client architecture relies on the client rather than the server to perform most

functions. This is usually accomplished with a Web browser plug-in or a native client

application. The client plug-in executes locally on client computers. The thick client

requests the source data (coordinates of vector data) from the server, and then renders maps

and performs analysis on the client side. This architecture is not feasible in situations where

there is only limited technical knowledge at client side (Fu and Sun, 2011).

3.6.1 GeoServer

It is an open source platform that supports the OGC standards like Web Map Service

(WMS), Web Coverage Service (WCS) and Web Feature Service (WFS). It can work with
17
a large range of data formats like Shapefile, Geo Tiff, JPEG etc., and various databases like

ArcSDE, Oracle Spatial, PostGIS etc. It has fully-featured Web administration interface.

OpenLayers is client side software which is widely used with GeoServer (Agarwal and

Gupta, 2014).

3.6.2 PostgreSQL/ PostGIS

PostGIS is spatial database which is built on top of a standard database PostgreSQL.

PostgreSQL provides transaction management, disk storage routines, SQL processing and

planning and other tasks of standard. PostGIS add spatial data types like points, line strings,

polygons, multi-points, multi-line strings, multi-polygons and geometry collections. It

supports both geometry and geography types. It also adds spatial functions like area,

distance, union, difference, buffer, touches, intersects, within, contains, overlaps etc.

(Agarwal and Gupta, 2014).

3.6.3 OpenGeo Suite

In this project OpenGeo suite which is a complete web mapping solution was used instead

of installing stand-alone software’s. OpenGeo suite developed by Boundless includes

PostGIS, GeoServer, OpenLayers, QGIS, GeoExplorer, GeoWebCache etc. Boundless

Suite is a full geospatial software stack that helps you create, store, style, compose and

publish maps and data with ease. Boundless Suite GeoExplorer is a tool for composing

mapping interface.

18
3.6.4 XAMPP

It is a free and open source cross-platform web server solution stack package developed by

Apache Friends, consisting mainly of the Apache HTTP server, MariaDB database, and

interpreters for scripts written in the PHP and Perl programming languages. In XAMPP

framework, PHP codes are used for database connection. XAMPP stands for Cross-

Platform (X), Apache (A), MariaDB (M), PHP (P) and Perl (P). It is a simple, lightweight

Apache distribution that makes it extremely easy for developers to create a local web server

for testing and deployment purposes.

3.7 SUMMARY

Preparation of Emergency information system is a part of preparedness phase of emergency

management. It is to be used in the response phase of emergency to reduce the effect of

disaster. Emergency managers could refer this Web GIS interface for all kind of

information regarding the hazard. Developing the interface using XAMPP as the web

server and OpenGeo Suite as the GIS package has never been attempted before. This

project checks the suitability of these software packages for the development of Web GIS

interface.

19
 CHAPTER 4

MATERIALS AND METHODS

4.1 GENERAL

The main emphasis of this study is to develop a Web GIS enabled Emergency Information

System for dam break flood hazard. Some panchayaths in Pampa River valley which is in

danger of inundation from dam break flood (breach of Pampa, Kakki or Anathode dam) is

selected for the study. A detailed evacuation plan to evacuate affected population from

inundated area to safe shelters is illustrated here. Procedure for shelter suitability

classification and evacuation route selection is also included.

4.2 STUDY AREA

The Pampa river basin extends over an area of 2,235 square kilometers (863 sq. mi) with

the entire catchment area within Kerala state. The basin is bounded on the east by Western

Ghats and on the west by Arabian Sea. The disaster considered in this study is Dam break

flood inundation. There are three dams namely Pampa, Kakki and Anathode dams in

Pampa river basin. The Pampa dam is constructed in Pampa River forming Pampa reservoir

and Kakki and Anathode dams are constructed in Kakki River forming Kakki-Anathode

reservoir.

More than sixty panchayath will be inundated if any of the three dams in Pampa

river basin break [Jairaj and Nair (2017)]. In this study, six panchayath in Pampa River

Basin are selected for the preparation of Emergency Information System.

20
Fig 4.1 Map showing the location of study area

Fig 4.2 Study area

21
 Location of study area is shown in Fig. 4.1. Panchayaths considered in this study are

Aranmula, Eraviperoor, Thiruvalla, Kuttoor and Thiruvanvandoor of Pathanamthitta district

and Chengannur Municipality of Alappuzha district as shown in Fig. 4.2. Study area spans

from 76° 32' E to 76° 42' E longitude and 9° 18' N to 9° 24' N latitude. Of the selected six

panchayaths, Kuttoor is fully inundated and rest is partially inundated. This particular study

area is selected for this project as it includes both fully inundated and partially inundated

panchayaths.

4.3 DATA COLLECTION

The data required for the preparation of an emergency information system for Pampa dam

break flood includes two types of information, disaster specific information and

information regarding the infrastructural facilities in the study area. Disaster specific

information include inundation extent map, time of arrival of flood wave, inundation depth

map etc. Infrastructural data are required for developing an evacuation plan and that

includes:

a. Hospital facilities

b. Road network

c. Educational institutions

d. Police and fire stations

e. Government offices

f. Places of worship etc.

Jairaj and Nair (2017) had conducted a dam break study and emergency action plan

preparation for the three dams in Pampa river basin. All the required information regarding

22
dam break flood hazard like, inundation map, time of arrival of flood wave, details about

the facilities in the study area etc. are collected from their project.

4.4 METHODOLOGY

In the present study, an Emergency Information System (EIS) was developed with the

purpose of publishing and maintaining information related to the emergency situation

resulting from dam break flood in selected panchayaths in Pampa river basin. Development

of EIS constitutes of four phases and it is shown in Fig. 4.3. First phase deals with the

prioritization of shelters in the study area based on their site suitability. In second phase,

population to be evacuated from each panchayath is estimated approximately, based on

census data 2011 and built-up area. Evacuation starting points and end points are

determined. In third phase optimum route for evacuation are found using Network Analysis.

Fourth phase constitutes the development of user interface and publishing of thematic layers

generated. Methodology of each of these phases is explained in detail below.

Fig. 4.3 Methodology of this study

23
4.4.1 Suitability Classification of Shelters:

There are a number of parameters which affect the suitability of shelters. In the present

study the location suitability of shelters is considered. The study area is not situated in an

earthquake faulty region and there are no hazardous industries in the region. So while

checking the site suitability of shelters mainly three factors are considered:

1. Accessibility to shelters.

2. Proximity to inundated area.

3. Proximity to medical centers.

Fig. 4.4 Methodology for Shelter suitability classification

Accessibility refers to the easiness for getting to the shelter from the affected area.

Here, the distance of shelter from major roads is considered as the deciding factor for

accessibility. Evacuees will undergo physical and physiological stress during a disaster.

Therefore, a shelter should be located as near as possible to medical centers. Shelter should

24
be evenly distributed around inundated area so that people can arrive there quickly before-

in-and-after disaster. Based on these three factors shelters are categorized based on their

suitability. Fig 4.4 shows the methodology followed to prepare the priority map of shelters.

Here only the site suitability is considered to prioritize the candidate shelters. First

thematic layers are prepared for the facilities like schools and colleges, hospitals, roads, etc.

Then the facilities in the inundated zone are excluded. Multiple buffer of varying distance

was provided around major roads and inundated region. Service area of hospitals are found

using Network Analysis. Service areas are a series of polygons representing the distance

that can be reached from a facility within specified time duration. These service area

polygons classify the area based on accessibility. AHP method is used to assign weights to

each parameter. Then Score of each shelter is found by WLC method. For that, Weighted

Overlay tool in ArcGIS is used and a site suitability map is prepared. The candidate shelter

layer is then overlaid over this and shelters are ranked and classified based on the suitability

area they fall in.

For doing Network Analysis, topology corrections have to be made for the road

network and then a network data set is created.

i. Creation of Topology

Before creating a network dataset we have to build topology and correct topological errors.

Topology can be built and validated for feature dataset. Node Renderer tool available in

ArcGIS platform was used to see where regular nodes, dangles and pseudo nodes exist and

corrections are made in the feature using editing tools. Topologically correct dataset is then

used to build the network dataset.

25
ii. Building Network Dataset

The ArcGIS Network Analyst extension enables to build a network data set and perform

analysis on a network data set. A network data set is created inside a geo-database. Once all

the thematic layers are added to it the next step is to model turns. Check ‘Global Turns’

which enables you to add default turn penalties. Next step is creating connectivity models

for the network. The connectivity policy of Streets is set to ‘End Point’ so that all streets

are connected to each other at endpoints. Next step is to set the network attributes. Network

attributes are properties of the network that control navigation. Common examples are cost

attributes that function as impedance over the network and restriction attributes that

prohibit traversal in both directions or in one direction, like one-way roads. In next step the

evaluators for each attribute were set. There is option to model driving directions too. Once

the finish button is clicked a summary of all the settings is displayed. Once the network is

created, the system asks if you want to build it. The build process determines which

network elements are connected and populates the attributes of the network data set. The

network should be built before performing any network analysis on it.

4.4.2 Fixing Evacuation Points:

In order to prepare an evacuation plan, it is necessary to have a rough estimate of the

population to be evacuated. Panchayath wise population data is obtained from LSGD

(Local Self Governing Department). To find the population in inundated zone, the spatial

distribution of population is needed. Built-up area map of study area was prepared from

satellite image and from inundated built-up area affected population is estimated. Starting

point for evacuation points was then determined by considering population distribution.

Nearby shelters in the study area are grouped together to form shelter clusters, so that
26
evacuation can be done to these clusters. Flow chart of the methodology is shown in Fig

4.5.

Fig. 4.5 Methodology for fixing evacuation points

4.4.3 Optimum Route for Evacuation

Evacuation of affected population is done panchayath wise. Optimum route for evacuation

is found by performing Network Analysis in ArcGIS. Optimum route is the route having

least cost. Cost can be distance or time. Dijkstra’s algorithm is used in Network Analysis to

find the route with minimum cost. Here time of travel is considered as the cost attribute.

While selecting the shelter cluster, the site suitability of that cluster and its capacity to

accommodate the evacuated population are also considered. Evacuation routes are proposed

in such a way that it causes minimum traffic conjunction. Steps to find evacuation route

(Fig 4.6) is explained below.

27
 Fig. 4.6 Methodology for evacuation routing

4.4.4 Development of User Interface

In order to provide the information required for handling the dam break flood hazard to a

large number of users, a Web GIS enabled Emergency Information has been developed.

Open Source Software (OSS) was used to develop the Web GIS framework. User interface

was developed using XAMPP as the Web Server and GeoServer as the GIS server.

PostgreSQL is the database in which all the data are stored. Workflow for developing Web

GIS enabled information system is shown in Fig 4.7. OpenGeo Suite which is a complete

GIS package (QGIS, GeoServer, PostGIS etc.) is used in this study, rather than installing

stand-alone versions of GIS software. In user interface, user can view information in

tabular form or as maps and query the displayed information. Provisions are made for the

authorities to update the information if needed.

28
 Fig. 4.7 Procedure for developing Web GIS enabled user Interface.

4.5 SUMMARY

The procedure adopted for developing Web GIS enabled Emergency information system is

discussed in this chapter. Detailed procedure for shelter priority mapping, evacuation

routing etc. are discussed. Web GIS architecture adopted in this work is also explained in

detail. This same procedure can be adopted to develop an Emergency Information System

for entire Pampa River basin. Dam breach in Pampa River valley, floods a very large

portion of land in the downstream side. About sixty panchayaths will be inundated.

Preparing an EIS for entire affected area is beyond the scope of present study, mainly due

to time constraints.

29
 CHAPTER 5

RESULTS AND DISCUSSION

5.1 GENERAL

In this project, an Emergency Information System was developed for selected panchayaths

in Pampa river basin. If breach occurs in any of the three dams, Pampa, Kakki or Anathode,

a large area will be flooded and causalities will be large. Emergency managers need to

organize rescue operations and affected population has to be evacuated to safer locations.

For that they need information regarding the time of arrival of flood wave, area inundated,

depth of inundation etc. Information regarding infrastructure and other facilities in the

study area are also needed for preparing evacuation plan. To initiate evacuation procedure,

suitability of temporary shelters in the locality and an estimate of affected population are

needed. Finding suitable shelters and evacuation are a part of emergency management and

it is discussed in detail in this chapter. A User Interface is developed to share all the

required information to the stakeholders.

5.2 SHELTER SELECTING CRITERIA

Shelters in the study area include facilities like schools and colleges. To check the

suitability of these facilities during a disaster, they are classified based on site suitability.

Three parameters are considered for ranking of candidate shelters and they are accessibility,

proximity to hospitals and proximity to inundated zone. We are only considering

accessibility through road in this study (water ways and railways are not considered).

30
Accessibility is measured in terms of proximity to major roads. Medical care is an

unavoidable factor while choosing a shelter. Medical assistance should be available to

people at the time of disaster and during their stay in temporary shelters. So we are

considering proximity to hospitals as a parameter in determining the suitability of shelters.

Other important criterion is the coverage of shelter to danger zone. Evacuation to nearby

safe shelters is preferred to distant places, so we are considering proximity to inundation

zone as another parameter for site suitability check. Thematic layers of hospitals,

educational institutions, road network and inundation map prepared are shown in Fig 5.1,

Fig 5.2, Fig 5.3 and Fig 5.4.

Fig. 5.1 Hospital facilities in study area

31
Fig. 5.2 Educational Institutions in study area

Fig. 5.3 Road network of study area

32
 Fig 5.4 Inundation map

First step is to assign weights to each parameter using AHP. Pair wise comparison

between each parameter is done based on their importance level and comparison matrix is

shown in Table 5.1. Importance is given after referring to the shelter requirements in

Federal Guidelines for Dam Safety, FEMA (2013). After deciding the importance of each

parameter with respect to other, the comparison matrix is prepared (Table 5.1). Then each

value in the matrix is divided by the sum of each column to get normalized relative weight.

Weights are then derived from the obtained normalized matrix (Table 5.2). The normalized

principal Eigen vector can be obtained by averaging across the rows.

33
 Table 5.1: Comparison matrix.

Parameter Inundated Proximity to Proximity to

area hospitals major roads

Inundated area 1 1/2 1/3

Proximity to
2 1 1
hospitals

Proximity to
3 1 1
major roads

Sum 6 2.5 2.33

To ensure that the weights given are consistent, consistency property of the matrix

is checked. Consistency Index (CI) and Consistency Ratio (CR) are calculated as in (5.1)

and (5.2):

CI = (λmax−n) / (n−1) (5.1)

CR = CI / RI (5.2)

Where n is the order of the matrix, λmax is the principal Eigen value of the matrix

and RI is the Random Index (Table 5.3). Consistency ratio (CR) should be less than or

equal to 0.10. The value of RI for suitability parameters with n value 3 is 0.52. Principle

Eigen value calculated is 3.042. Here the CR values obtained are 0.04 which indicates the

weights assigned are consistent. From the above paired comparison weight for proximity to

inundated area is 18 %, proximity to hospitals is 38% and proximity to major road is 44%.

34
 Table 5.2: Deriving weights from Normalized matrix.

Parameter Inundated Proximity to Proximity to Weights

area hospitals major roads

Inundated 0.17 0.2 0.143 0.18

area

Proximity to 0.33 0.4 0.43 0.38

hospitals

Proximity to 0.5 0.4 0.43 0.44

major roads

Table 5.3: RI Values for different order of the matrix [Saaty, T., 1980]

N 2 3 4 5 6 7 8

RI 0 0.52 0.59 1.12 1.24 1.32 1.41

Inundation zone is defined by including the regions which are surrounded by flood

water to the inundated area map, as they are not accessible till the flood water recedes. The

facilities which fall in the inundation zone are erased from the layer as they cannot be

accessed during the disaster.

Proximity to hospitals is found by defining service area for hospitals. Service area is

created using Network Analysis. First step in Network Analysis is the creation of Network

Dataset. Network Database is created using topology corrected road network layer. In order

35
to find the service area, hospital layer is loaded as ‘Facilities’ into the ‘Network Analyst’

toolbar. Here the impedance for analysis is travel distance, so distances for the generation

of service area are specified. Once all the parameters for analysis are specified, service area

polygons are generated. A new thematic layer is created by exporting service area polygons

(Fig. 5.5). Proximity to inundation zone and major roads are found based on the distance of

shelters from them. For measuring that distance, multilayer buffer maps of varying distance

are created around inundated zone and major roads (Fig 5.6 and 5.7).

Fig. 5.5: Service area of hospitals.

36
Fig. 5.6: Multiple buffers provided to inundation zone.

Fig. 5.7: Multiple buffers provided to major roads.

37
5.3 SUITABILITY CLASSIFICATION OF SHELTERS

The WLC was implemented to estimate site suitability of shelters. ‘Weighted Overlay’ tool

in ArcGIS performs the WLC analysis. Service area layer of hospitals and buffer maps are

converted to raster format and added to weighted overlay tool. Weights for each input raster

are specified before analysis. As the input criteria layers are of different ranges, to combine

them in a single analysis, each cell of each criterion must be reclassified into a common

preference scale. The preference values are on a relative scale with the highest value being

most favorable. Locations close to major roads, hospitals and evacuation zone are assigned

higher scale value. The Layers, the criteria used, their rating, and their weights are

summarized in Table 5.4. Weighted overlay tool then multiplies the cell values of input

raster by its weight of importance and adds the cell values together to produce the output

raster. Higher values in the output raster generally indicate that a location is more suitable.

As a result of the weighted overlay operation, site suitability map of the study area

is obtained. To get a simplified map, number of classes in the output map is restricted to

three. So we get a suitability map showing areas of high, medium and low suitability (Fig

5.8). The classification is done based on Jenk’s Natural Breaks Algorithm. It is a data

clustering method designed to arrange the available values into different classes in the most

suitable way. This optimization method seeks to reduce the variance within classes and

maximize the variance between classes.

38
 Table 5.4: Factor rating and weights assigned to each parameter.

Sl.
Factor Buffer Distance (m) Factor rating Weights
No.

<500 8
Proximity to
500-1000 6
1 health care 38
1000-2000 4
facility
2000-3000 2

<250 8

Proximity to 250-500 6
2 44
major roads 500-1000 4

1000-1500 2

<500 8

Proximity to 500-1000 6
3 18
danger zone 1000-2000 4

2000-2500 2

In order to classify shelters, the ‘candidate shelter’ layer is overlaid over the

suitability map. Candidate shelters are the shelters obtained after erasing shelters in

inundated area. The shelters are classified based on the score for the location it falls within.

All the candidate shelters present are classified into three classes: low suitability, medium

suitability and high suitability (Fig 5.9). It is observed that about 64 % of the available

shelters belong to highly suitable shelters. Among them 11 % are classified to be in

medium suitable location. Only 25 % of the facilities are classified as less suitable.

39
 Fig. 5.8: Suitability map of study area

Fig. 5.9: Suitability classification of shelters.

40
5.4 DETERMINATION OF EVACUATION POINTS

Evacuation start points are the locations from which evacuation starts. Evacuation starting

points depend on the population distribution of study area. In order to fix evacuation starting

points, the built-up area in the study area has to be identified. Built-up area map is prepared

by applying built-up index (equation 5.3) in Landsat 8 image. The Normalized Difference

Built-up Index (NDBI) map (Fig. 5.10) is calculated using ‘Model maker’ tool in Erdas

Imagine. The pixels in white, bright grey and light grey tones show built-up areas, whereas

the dark grey and black tones represent other land covers. The variability in the bright tones

indicates the differences in built-up density, where the areas with high built-up density

appear the brightest and vice versa (Bhatti and Tripathi (2014)).

The built-up area map of study area that is prepared from NDBI map using

‘knowledge engineer tool’. Actual built-up area in the study area is identified by field

survey. Then the pixel values of these identified areas in the NDBI image are noted. It is

observed that the threshold value of built-up in the NDBI image ranges from 0.45 to 0.7. A

new hypothesis is created in the knowledge engineer to specify the output class and its color.

Rules are defined for the created hypothesis. The ‘Rule properties’ provides a way to enter

variables and their values with their relationship with the rule. Threshold value for

calculating built-up are provided as variable values and then the hypothesis is executed.

Built-up area map obtained is shown in Fig. 5.11.

NDBI= (SWIR-NIR) / (SWIR+NIR) (5.3)

SWIR = Shortwave Infrared.

41
NIR = Near Infrared.

Fig. 5.10 NDBI map of study area

Fig. 5.11 Built-up area map

42
 Population to be evacuated from each panchayath was roughly estimated by finding

population density of inundated built-up area. A 100 meter buffer is provided around the

inundated area and that area is declared as the evacuation zone. Population data is obtained

from Census data 2011. Only the census data at Panchayath level was available. Hence it is

assumed that the population in a panchayath is uniformly distributed in the built-up area.

Inundated built-up area is found by clipping the built-up area map to the extent of

evacuation zone. Population density of each panchayath was found by dividing the

population with built-up area of panchayath. Then affected population was found by

multiplying population density with inundated built-up area of each panchayath. Table 5.5

shows the approximate population to be evacuated from each panchayath.

Table 5.5 Population to be evacuated from each panchayath

Sl. No. Panchayath Population Evacuated

1 Aranmula 13339

2 Chengannur 15393

3 Eraviperoor 2678

4 Kuttoor 17918

5 Thiruvanvandoor 15663

6 Thiruvalla 3957

Evacuation starting points were provided after considering accessibility via roads

and nearness to prominent land marks. These starting points should be well known and

easily accessible by people of that locality. Shelters are the evacuation destinations. But in

43
order to prepare an evacuation plan they should be properly organized. Nearby shelters

were grouped together to form shelter clusters. Nearby Churches and temples were also

included in these clusters so that the evacuated people can be efficiently accommodated.

Places of worship in the study area are shown in Fig 5.12. Shelter clusters are classified

based on the capacity of facility included in it. Capacity is assumed to be more for colleges

than for schools and churches. So colleges are given a higher rating than schools. Churches

are given a lower rating when compared to other two facilities. Then number of schools,

colleges and churches present in each cluster are multiplied by their respective rating.

Based on the value obtained, shelter clusters are classified roughly as having high, medium

and low capacity (Fig 5.13). Fig. 5.14 shows the evacuation start and shelter clusters

located in the inundated built-up area map of study area. Facilities included in each cluster

are shown in Table 5.6 and Table 5.7 shows the capacity wise classification of shelter

clusters.

Fig. 5.12: Places of worship in study area.

44
 Fig. 5.13: Classification map of shelter clusters

Fig. 5.14: Evacuation start points and shelter clusters.

45
 Table 5.6 Facilities included in shelter clusters
Sl.
Cluster Place Schools/colleges Churches/temples
no.
1 CLUS_CHE_A Angadikkal MMAR School, Govt. HSS Angadical South, Govt. Nil
,Chengannur ITI for Women, Christian College.
2 CLUS_CHE_B Chengannur Karavali Group of College, Chinmaya Vidyalaya, Devi Temple Vandimala, Chengannur
Sree Ayyappa College. Temple
3 CLUS_CHE_C Puthencavu, St. Ann's Girls High School, SCRV TTI angadical. Nil
Chengannur
4 CLUS_CHE_D Chengannur College of Engineering, Govt Girls High School, Malankara suriyani catholica church, St
Government Girls Higher Secondary School. Andrews CSI church, Marymatha
church moolapadav
5 CLUS_AR_A Thannikunnu, Govt. SNDP UPS Vallana, MTLP School St. Marys Malankara Catholic
Aranmula Idayaranmula west Church.St. Stephens Mount Orthodox
Church
6 CLUS_AR_B Kidangannoor, Govt. LPS Kidangannur, SVGV Higher Secondary Pallimukkam Devi Temple, Anikkad
Aranmula School, SV GV High School Mahadeva Temple
7 CLUS_AR_C Manappally St. Marys MTLP School, Vallana Higher Secondary IPC Kidangannur Town Church, Peniel
,Aranmula School, ST. Gabrial English Medium School Centre,

46
 Table 5.6 Facilities included in shelter clusters contd..
Sl.
Cluster Place Schools/colleges Churches/temples
no.
8 CLUS_ER_A Vallamkulam, Govt. UPS Vallamkulam, National High School, St,Marys Orthodox Church, Nanoor Devi
Eraviperoor Govt DVLPS, Govt. LPS Vallamkulam. Temple, St.mary's church Vallamkulam
9 CLUS_ER_B Othera, NSS TTI Othera, DVNSS High School, Govt. Ebenezer Marthoma Church, St. Mary's
Eraviperoor LPS east Othera Orthodox Church
10 CLUS_ER_C Kozhimala, AMM High School, Nazareth College of AG Church Kozhimala, MYM knanaya
Eraviperoor Pharmacy, St.Marrys U P School church, Little Flower Malankara Catholic
Church
11 CLUS_ER_D Eraviperoor Immanuel Marthoma Central School, St. Marys St. Annes Church, St. Thomas Orthodox
UPS,Govt. UPS Iraviperoor, Eraviperoor St john's Church, Immanuel Mar Thoma Church,
school, L P School, Muringaseril Eraviperoor
12 CLUS_ER_E Nellikkal, NSS KUPS east othera, CMS LPS Othera St. Johns CSI Church
Eraviperoor
13 CLUS_ER_F Nellimala, EALPS thevarkad, OEM Public School Mar Gregorios Orthodox Church,
Eraviperoor Vallamkulam St. Luke's Mar Thoma
Church, Mar Gregorious Orthodox
Church

47
 Table 5.6 Facilities included in shelter clusters contd..

Sl.
Cluster Place Schools/colleges Churches/temples
no.
14 CLUS_THR_A Thiruvalla MGM Higher Secondary School, Govt. Model St.John's Metropolitan Cathedral, St.
GHS, National college, S.C. Seminary High John's Cathedral, St Mary's Bethany
School, Nampoothirys College Orthodox Church.
15 CLUS_THR_B Thirunoolapuram, St. Thomas HSS Eruvelipra, Balikamadom St.Marys Malankara Catholic Church,
Thiruvalla HSS, Thirumoola Vilasam UPS, SNV High Subrahmanya Swami Temple
School
16 CLUS_THR_C Kuttapuzha, Christ central cbse school, Muthoor NSS High St. George Orthodox Church, St
Thiruvalla School Antony's Church, Muthoor Sree
Bhadrakali Temple
17 CLUS_THR_D Pulimootil padi, CSI VHSS for Deaf, Amrita School, AIM Kadalimangalam temple, St Mary's
Thiruvalla School Orthodox Syrian Church, St Thomas C
S I Church

48
 Table 5.7 Capacity classification of shelter clusters

Sl. No. Name Place Panchayath District Capacity

1 CLUS_CHE_A Angadikal Chengannur Alappuzha medium capacity

2 CLUS_CHE_B Chengannur Chengannur Alappuzha medium capacity

3 CLUS_AR_A Thannikunnu Aranmula Pathanamthitta medium capacity

4 CLUS_ER_A Vallamkulam Eraviperoor Pathanamthitta medium capacity

5 CLUS_ER_B Othera Eraviperoor Pathanamthitta low capacity

6 CLUS_ER_C Kozhimala Eraviperoor Pathanamthitta low capacity

7 CLUS_THR_A Thiruvalla Thiruvalla Pathanamthitta high capacity

8 CLUS_AR_B Kidangannoor Aranmula Pathanamthitta low capacity

9 CLUS_AR_C Manappaly Aranmula Pathanamthitta medium capacity

10 CLUS_ER_D Eraviperoor Eraviperoor Pathanamthitta high capacity

11 CLUS_THR_B Thirunoolapuram Thiruvalla Pathanamthitta medium capacity

12 CLUS_CHE_C Puthencavu Chengannur Alappuzha low capacity

13 CLUS_CHE_D Chengannur Chengannur Alappuzha medium capacity

14 CLUS_ER_E Nellikkal Eraviperoor Pathanamthitta medium capacity

15 CLUS_THR_C Kuttapuzha Thiruvalla Pathanamthitta low capacity

16 CLUS_THR_D Pulimootil padi Thiruvalla Pathanamthitta low capacity

17 CLUS_ER_F Nellimala Eraviperoor Pathanamthitta low capacity

5.5 EVACUATION ROUTING

Preparing an evacuation plan is of great importance in emergency management. Evacuation

is usually done to nearby facilities which act as temporary shelters during an emergency. In

this study, facilities like schools, hospitals, churches, temples etc. which are in the same
49
locality are grouped together to form shelter clusters which will act as evacuation

destinations. For a particular evacuation point, its destination has to be selected by

considering many factors like travel time, shelter capacity, time of arrival of flood wave,

traffic in the route etc. Here evacuation shelters are decided by not only considering

nearness to evacuation start point but also these factors. Time to reach inundation for

panchayaths in study area based on different breach conditions are given in Table 5.8.

Optimum route for evacuation is found by Network Analysis. Speed limit for

traversing a link is set to 10 km/hr. Travel time is calculated from the length and speed

limit of the link. Routing is done with travel time as the link impedance. ‘New Route’

option in the Network Analyst tool bar is used to find shortest route between two points. A

‘start point’ and a ‘shelter cluster’ are selected as origin and destination stops respectively.

Then the route between origin and destination stops with minimum impedance is

calculated. Dijkstra’s algorithm is the algorithm used for finding shortest route between two

points. Different combinations of start point and shelter cluster are tried to find the most

efficient evacuation plan. The distance to be traveled, travel time and step by step direction

are available for each route. Fig 5.15 shows the evacuation routing in ArcGIS. Table 5.9

shows the route details of study area. Panchayath wise evacuation route for Aranmula,

Chengannur, Thiruvanvandoor, Kuttoor, Thiruvalla and Eraviperoor are shown in Fig 5.16

to Fig 5.20 in the same order.

50
 Table 5.8 Time to reach inundation for selected panchayaths in study area

[Jairaj and Nair (2017)]

Sl. Panchayath Time to reach inundation (Hrs)

No. Fair weather condition Probable maximum flood

Pampa Kakki & Pampa, Pampa Kakki & Pampa,

Anathode Kakki & Anathode Kakki &
Anathode Anathode
1 Aranmula 9.27 6.77 6.79 23.78 6.74 6.49

2 Chengannur 16.06 10.04 9.83 31.32 9.97 9.26

3 Thiruvanvandoor 16.06 10.04 9.83 31.32 9.97 9.26

4 Kuttoor 13.83 9.28 9.19 27.81 9.17 8.64

Fig. 5.15: Finding shortest route in ArcGIS.

51
 Table 5.9 Evacuation route details of study area

Travel
Travel Time
Sl. No Panchayath Start Place Destination Distance Directions
(min)
(m)

1 Aranmula Karimukham Thannikunnu 2171.75 13.03 Karimukham to Thannikunnu

Arattupuzha to chakittapadi to
2 Aranmula Arattupuzha Thannikunnu 3565.48 21.39
Thannikunnu
Edayaranmula to ikkara
3 Aranmula Edayaranmula Kidagannoor 2195.94 13.17 toiyyan kovikal padi to
Kidagannoor
4 Aranmula Chakittapadi Thannikunnu 2060.88 12.38 Chakittapadi to Thannikunnu
Karinthaloor to Iyyankovikal
5 Aranmula Karinthaloor Kidangannoor 1824.34 11
padi to Kidangannoor
Mezhuveli to Kidangannur to
6 Aranmula Mezhuveli Manapally 2568.75 15.41
Manapally
Althara Mundencavu to Chengannur via
7 Chengannur Mundencavu 1980.66 12
junction MC road to Althara junction
Narendrabooshan Narendrabooshan Road to
8 Chengannur Chengannur 508.28 3
Road Althara junction, Chengannur

52
 Table 5.9 Evacuation route details of study area contd..

Travel
Travel Time
Sl. No Panchayath Start Place Destination Distance Directions
(min)
(m)

9 Chengannur Nellikkal Edanadu 3123.34 8.52 Nellikkal to Edanadu

10 Chengannur Othera Mangalam 3009.51 19.52 Othera to Kallishery to Mangalam

Kallisery to Chengannur via MC

11 Chengannur Kallisery Chengannur 3106.79 22.16
road
Puthencavu to Netagi Road,
12 Chengannur Puthencavu Chengannur 1492.71 4
Chengannur
Pandavanpara temple to Chengannur
Pandavanpara Christian
13 Chengannur 3441.28 20.65 to Hatchery jn., near Christian
temple College
College

14 Chengannur Edanadu Nellikkal 1820.62 8.52 Edanadu to Nellikkal

15 Eraviperoor West Othera Nellimala 4645.55 18 West Othera to Nellimala

Nannadu to Nannoor to
16 Kuttoor Nannadu Vallamkulam 5164.13 31
Vallamkulam

53
 Table 5.9 Evacuation route details of study area contd..

Travel
Travel Time
Sl. No Panchayath Start Place Destination Distance Directions
(min)
(m)
Vennapala to Thukalassery to
17 Kuttoor Vennapala Thiruvalla 5828.34 35
Thiruvalla
Vennapala to Thukalassery to
18 Kuttoor Vennapala Thiruvalla 4885.27 30
Thiruvalla

West Othera to Pazhayakavu jn. to

19 Kuttoor West Othera Nellimala 7089.49 29
Nellimala

Vallamkulam to Eraviperoor via

20 Kuttoor Vallamkulam Eraviperoor 5538.36 27
thiruvalla kumbazha hwy.
Kuttoor to Thukalassery to
21 Kuttoor Kuttoor Ramanchira. 5640.73 33.85
Thiruvalla to Ramanchira.
Kadalimangalam Balikamadom Near Kadalimangalam temple to
22 Thiruvalla 859.87 5.16
temple high school Balikamadom high school

54
 Table 5.9 Evacuation route details of study area contd..

Travel Travel
Sl.
Panchayath Start Place Destination Distance Time Directions
No
(m) (min)
Nedumpuram to venapala to
23 Thiruvalla Nedumpuram mathilbhagom 1888.41 11.33
mathilbhagom
Muzhuker to pravinkoodu to
24 Thiruvanvandoor Muzhuker Eraviperror 8154.31 32 kuttoor to Kozhimala to
Eraviperror

25 Thiruvanvandoor Chengannur Kallissery 2103.35 13.44 Chengannur to Kallissery

Kuttiyathodu to thiruvanvandoor to
26 Thiruvanvandoor Kuttiyathodu othera 7617.75 37
pravinkoodu to othera

Balikamadom Nannadu to Kuttoor to near

27 Thiruvanvandoor Nannadu 3500.30 21
Girls school balikamadom Girls school

Thiruvanvandoor to pravinkoodu
28 Thiruvanvandoor Thiruvanvandoor Eraviperoor 9085.12 38
to kozhimala to Eraviperoor

55
 Fig. 5.16: Evacuation routes for Aranmula Panchayath.

Fig. 5.17: Evacuation routes for Chengannur Panchayath.

56
Fig. 5.18 Evacuation routes for Thiruvanvandoor Panchayath

Fig. 5.19 Evacuation routes for Kuttoor Panchayath

57
 Fig. 5.20 Evacuation routes for Thiruvalla and Eraviperoor Panchayath

5.6 USER INTERFACE

The main purpose of developing a user interface is to provide the information required to

handle the dam break flood hazard to the stake holders. Disaster specific information like

inundation map, time of inundation etc., along with the infrastructural facilities in the study

area is included in the user interface. Thematic layers of shelter suitability classification,

built-up area distribution, evacuation routes etc. are also provided. Stake holders can access

this information any time they need. They could make necessary plans and precautions for

dealing with dam break flood hazard by referring to the data available in the user interface.

58
5.6.1 Development of User Interface

For developing a GUI, a database to store the data, a Map server to publish the data and a

Web server to upload the data were needed. A server side or thin client architecture was

used to develop Web GIS application because of the added advantage like ease in updating

data, negligible resource requirement at client side etc. OpenGeo Suite is the software used

in this work. It consists of GeoServer, PostgreSQL, GeoExplorer, OpenLayers etc.

In this study, PostgreSQL (Postgres) which is an object relational database

management system (ORDBMS) along with its spatial extension PostGIS was used as the

database server. Thematic layers prepared in ArcGIS software were first imported into

Postgres database. The conversion of shapefiles to Postgres database is achieved using the

sph2pgsql utility included as part of the PostGIS extension. Fig 5.21 is a screenshot

showing shapefiles imported to Postgres. Shapefiles imported to the database includes

study area, inundation area, infrastructural facilities, road network, shelter clusters, start

points, evacuation routes etc. After the layers are imported, the PostGIS database was

connected to GeoServer. Before importing from database, a new workspace is created in

GeoServer (Fig 5.22). A workspace is a separate, isolated space relating to a certain project.

A workspace is created by providing a name and a name space URI (Uniform Resource

Indicator). Then a store is created which will retrieve data from database. Imported layers

are then published by specifying coordinate system, workspace, style, and publishing

parameters. Styling of each layer is done in QGIS software and individual styles are

exported as SLD (styled layer descriptor) files. This SLD file is then imported into

GeoServer as new style format. ‘Layer preview’ option is used to check whether the layers

59
are published correctly or not. Maps are displayed using OpenLayers which is a free

mapping library associated with GeoServer (Fig 5.23).

Fig. 5.21 Importing shapefiles to Postgres

Fig. 5.22: Creating workspace in GeoServer.

60
 Fig. 5.23 Previewing published layers in OpenLayers

GeoExplorer is a map composer application in OpenGeo Suite. It is a Web-based

map configuration and styling utility. The shapefiles published in GeoServer is accessible

from its interface. Layers can be overlaid over one another to create the desired appearance.

Icons for panning, editing, querying, legend etc. can be attached with the map window.

Then these maps can be embedded in Emergency Information interface. It is a more easy

way to publish the maps in user interface than using mapping libraries like OpenLayers or

Leaflet as it does not demand extensive programming knowledge.

XAMPP is the Web server used in the study. ‘PhpPgAdmin’ is the database

interface used in XAMPP to connect with PostgreSQL. It is perfect for hosting services.

Database connection is established by providing the host name and port number in which

the database runs. Interface was designed using different technologies like web designing

language HTML and scripting languages like CSS, PHP and OpenLayers API. The code for

61
development of user interface is given in Appendix. Body of interface is developed using

HTML language and styling is done using CSS scripts. JavaScript is used to introduce

dynamicity to the web content. PHP codes are used for database connection and querying.

5.6.2 Working of User Interface

In the home page of user interface, study area layer is overlaid over Google Maps. A

window showing attribute data will be displayed if any panchayath is selected. Screenshot

of the home page is shown in Fig 5.24. An ‘About’ page is included in the interface that

explains the working of the interface and includes all the necessary information regarding

this work. It describes what all functions can be performed by the user interface. Figures

and demoes are included to help the users to better understand the working of the interface.

By selecting the ‘Facilities’ menu user can view the facilities in the study area. Safe

facilities and inundated facilities can be viewed separately. Fig 5.25 shows a screenshot of

user interface showing safe facilities in the study area. Facilities like hospitals, churches

and temples, educational institutions, Government offices, road network etc. are overlaid

over a base map of study area. Toolbar with functions like zooming, panning, measuring

distance and area etc. are provided. Map scale, legend and location coordinates are included

in interface along with identity tool which make a popup window showing attribute

information of selected feature. Users can also query these layers based on the attribute data

associated with it.

62
Fig. 5.24 User Interface for Emergency Information System.

Fig. 5.25 Interface showing safe facilities in study area

63
 Thematic layers of shelter suitability, shelter cluster, evacuation start points,

evacuation routes etc. are available from the ‘Evacuation plan’ menu. Fig 5.26 shows the

screenshot of the GUI showing evacuation routes for Aranmula panchayath. Evacuation

routes are provided panchayath wise. A study area map overlaid with evacuation start

points and shelter clusters are provided. When a user selects a start point, optimum

evacuation route for that point is displayed. Travel time, travel distance, directions etc. are

included as attribute information of the evacuation routes.

Fig. 5.26 Interface showing evacuation route details of Aranmula Panchayath.

Sometimes due to some unforeseen incidents the provided evacuation destinations

or routes may not be adopted. In such situations emergency managers can try alternative

routes and select the best possible one. This is made possible by including Google Maps in

64
the interface. ‘Evacuation Route’ icon is provided in the user interface to find alternate

routes. When that icon is selected users are directed to Google map interface. Evacuation

start point and shelter cluster layer are overlaid over this map. User can select any two

points as origin and destination by simply clicking on that feature. Then shortest route

connecting these points will be displayed (Fig. 5.27). Any number of destinations can be

added to find the best route connecting them.

Fig. 5.27 Evacuation routing using Google Maps embedded in interface.

By accessing the ‘Administrative’ Interface the maps can be updated. In order to

access this page user has to login with an ID and password. Map modification and addition

of new data are done by the concerned authority and the changes made will be available to

all stake holders through the Graphical user interface.

65
5.7 SUMMARY

In this project an effort is made to develop a model of Emergency Information for dam

break flood hazard in Pampa River basin. Due to time constraints only six panchayaths in

Pampa river basin is considered for present study. Information system developed for a

hazard management should provide relevant details of the disaster, affected area,

infrastructural facilities present, population affected etc. In the case of a dam break flood

disaster, information on the location of temporary shelters, their suitability and an

evacuation plan to rescue the affected people is also necessary. Evacuation plan is prepared

based on the inundation map obtained from dam break flood simulation done by Jairaj and

Nair (2017). Other details like time of arrival of flood wave, depth of inundation, spatial

distribution of facilities in the study area etc. are also collected from their work.

Before preparing an evacuation plan, the suitability of facilities available in the

study area as temporary flood shelters should be checked. Site suitability of temporary

shelters is calculated by considering three parameters: proximity to major roads, proximity

to inundation zone and proximity to hospitals. Weight of importance for these parameters

are found using AHP. Shelter suitability classification is done using WLC method and

shelters are classified into three suitability classes. 64 % of the shelters are classified as

highly suitable, 11% as moderately suitable and 25 % of the shelters are found to be of low

suitability.

A rough estimate of population affected is needed to prepare an evacuation plan.

Population to be evacuated from each panchayath is calculated based on inundated built-up

66
area. Built-up area map of study area is prepared from Landsat-8 image using built-up

index. Starting points for evacuation is selected by considering the spatial distribution of

built-up area and roads. Instead of using facilities as separate shelter destinations, a number

of shelters in the study area are grouped together to form shelter clusters. Facilities like

schools, colleges, churches, temples etc. are included in these clusters. Suitable shelter

cluster for each starting point is decided after considering population to be evacuated,

capacity of shelter, travel time etc. Evacuation routing is done by performing Network

Analysis. Network Analyst tool calculates the shortest route between origin and destination

points using Dijkstra’s algorithm.

User Interface is developed using XAMPP as the web server and GeoServer as the

GIS server. PostgreSQL is the database used to store the data and it is connected to

XAMPP using PhpPgAdmin. Postgres is used along with its spatial extension PostGIS.

Interface was designed using web designing language HTML and scripting languages like

CSS, PHP and OpenLayers API. All the facilities in the study area, inundation map, shelter

suitability map, built-up area map, evacuation routes etc. are provided in the interface.

Users can also do spatial query on the layers based on the attribute information associated

with it. Provisions are made in the interface to modify existing layers and to add new data.

One of the many advantages of using Web GIS is that users can access the

information efficiently in a common platform. Multiple users are able to visualize the same

information simultaneously. Any modifications made in the interface will be readily visible

to the users. This saves a lot of time and effort. Information system developed in this

project provides details of the infrastructure facilities in the study area, suitability
67
classification of shelter and a detailed evacuation plan with routes. By including

Emergency Action Plan for Dam break flood hazard this prototype can be developed into a

complete Emergency Information System. Moreover, in the future, this system can be

expanded to include entire Pampa River Basin.

68
 CHAPTER 6

CONCLUSION

The main focus of this study is the development of a Web GIS based Emergency

Information System. Such a system is developed for Dam break flood hazard for selected

panchayaths in Pampa river basin. Important problems addressed in this project are shelter

suitability classification, preparation of evacuation plan and development of user interface.

The major conclusions from the study are as follows:

i. Shelters in the study area are classified based on their site suitability by considering

three parameters: Proximity to major roads, Proximity to hospitals and Proximity to

inundation zone.

ii. Facilities like schools, colleges, churches and temples are grouped together to form

shelter clusters which acts as evacuation destinations and evacuation start points are

selected by considering spatial distribution of built-up area and road network.

Optimum route for evacuation is found for each start point by performing Network

Analysis. Evacuation plan is prepared for entire study area.

iii. User interface is developed with XAMPP as the Web Server, GeoServer as the GIS

server and PostgreSQL as the spatial database.

iv. User interface displays the entire infrastructural feature in the study area, inundation

map, shelter suitability map, evacuation plan etc. Evacuation plan is prepared

panchayath wise. Map querying, panning, zooming, etc. are facilitated in the

69
 interface. Routing between any two points of interest is made possible by

embedding Google Maps into the interface.

v. Provisions are made to modify existing maps in the interface and to add new maps

to the database.

The developed Information system is used to provide information required for dam

break hazard management to stake holders in a fast and efficient way. Modifications made

on the interface will be available to the users without any delay. This prototype can be

modified into a complete Information System by incorporating Emergency Action Plan for

Pampa Dam break flood hazard. Details of other panchayaths can be included in this

interface to use this Information system for entire river basin.

70
 REFERENCES

1. Saaty, T., “The Analytic Heirarchy Process : Planning, Priority Setting, Resource

Allocation”, MacGraw Hill, 1980.

2. Alesheikh, A., Helali, H. and Behroz, H. (2002). “Web GIS: technologies and its

applications”, In Symposium on geospatial theory, processing and applications.

Retrieved from http://www.isprs.org/proceedings/xxxiv/part4/pdfpapers/422.pdf.

3. Hodgson, M. E., and Kar, B., (2008) “A GIS based model to determine site

suitability of Emergency Evacuation Shelters”, Transactions in GIS, 12(2): 227–

248.

4. Koehl, M. and Joseph, J., (2008) “Interface for dissemination of GIS application

on internet”, The International Archives of the Photogrammetry, Remote Sensing

and Spatial Information Sciences. Vol. XXXVII. Part B4. Beijing.

5. Fu, P., and Sun, J. Web GIS Principles and Applications, ESRI Press, 2011.

6. Lariyah, M. S., Vikneswaran, M., Hidayah, B., Muda, Z. C., Thiruchelvam, S.,

Abd Islam, A. K., and Rohani, H. (2013) “Numerical Modelling dam break

analysis for water supply project” 4th International Conference on Energy and

Environment, Universiti Tenaga Nasional.

7. Agrawal, S. and Gupta, R. D.,(2014) “Development and comparison of open

source based Web GIS frameworks on WAMP and Apache Tomcat web servers”,

The International Archives of the Photogrammetry, Remote Sensing and Spatial

Information Sciences, Volume XL-4, 2014.

71
8. Bhatti, S. S. and Tripathi, N. K., (2014) “Built-up area extraction using Landsat 8

OLI imagery”, GIScience & Remote Sensing, Vol. 51, No. 4, 445–467.

9. Karnatak, H., Kamal, P., Roy, A., Joshi, D., and Oberai, K., (2014) “Geospatial

data sharing online spatial analysis and processing of Indian biodiversity data in

internet GIS domain- A case study for raster based online goeprocessing”, The

International Archives of the Photogrammetry, Remote Sensing and Spatial

Information Sciences, Volume XL-8, 2014.

10. Singh, S. P. and Singh, P., (2014) “Mapping spatial data on the web using free and

open-source tools:A prototype implementation”, Journal of Geographic

Information System, 2014, 6, 30-39.

11. Bhanumurthy, V., Jai Shankar, G., Rao, K. R., and Nagamani, P. V., (2015)

“Defining a framework for integration of geospatial technologies for emergency

management”, Geocarto International.

12. Gohain, G. B., Mipun, B. S. and Singh, K. K.,(2015) “ Web based GIS for

infrastructure information using Google Earth imagery, open source software and

standards”, International Journal of Remote Sensing & Geoscience, Volume 4,

Issue 4.

13. Hwang, H., Ra, K. & Kim, C. (2015), “A Web-based System for Shelter and

Evacuation Path Selection Using Spatial Models in Disaster Situations”,

International Journal of Smart Home, Vol. 9, No. 10, pp. 87-96.

14. Nappi, M. M. L., and Souza, J. C., (2015) “Disaster management: hierarchical

structuring criteria for selection and location of temporary shelters”, Natural

Hazards, Vol 75, Issue 3, pp 2421-2436.

72
15. Bhandari, A., Panwar, A. and Saklani, P., (2016) “Creation of a Web based GIS

application for spatial wasteland information system for Uttarakhand state using

effective open source software”, International Journal of Computer Science and

Information Technologies, Vol. 7 (6).

16. Jairaj, P. G., and Nair, B. T.,(2017) “Emergency Action Plan for Kakki, Anathode

and Pampa dams in Sabarigiri Hydroelectric project.” Center for Continuing

Education, College of Engineering Trivandrum.

73
 APPENDIX-A

A.1 CODES FOR DEVELOPING USER INTERFACE

<html>

<head>

<meta charset="utf-8"/>

<link rel="stylesheet" href="style2.css">

<script>

// Get the modal

var modal = document.getElementById('id01');

// When the user clicks anywhere outside of the modal, close it

window.onclick = function(event) {

if (event.target == modal) {

modal.style.display = "none";

}}

</script>

</head>

<body>

<div class="wrap">

<h1> EMERGENCY RESPONSE CENTER, PAMBA RIVER BASIN </h1></br>

<marquee direction ="left"><font color="#483D8B">

EVACUATION PLAN FOR DAM BREK FLOOD IN PUMBA RIVER BASIN.....A

PROJECT WORK </font></marquee>

74
<div id="container">

<ul>

<li><a href

="https://www.google.com/maps/d/embed?mid=1I2bmTm3jbYL59zSRfeUWNEInaZw"tar

get="main">Home</a></li>

<li><a href ="acordtion.html" target="_blank" >About</a></li>

<li>Facilities<ul>

<li><a href ="http://localhost:8080/geoexplorer/viewer/#maps/8" target="main">Safe

Facilities</a></li>

<li><a href ="http://localhost:8080/geoexplorer/viewer/#maps/13"

target="main">Inundated Facilities</a></li>

<li><a href ="http://localhost:8080/geoexplorer/viewer/#maps/14"

target="main">Candidate Shelters</a></li>

<li>Query Map<ul>

<li><a href ="http://localhost:8080/geoexplorer/composer/#maps/8" target="main">Safe

Facilities</a></li>

<li><a href ="http://localhost:8080/geoexplorer/composer/#maps/13"

target="main">Inundated Facilities</a></li>

</ul></li></ul></li>

<li>Evacuation Plan

<ul>

<li><a href ="http://localhost:8080/geoexplorer/viewer/#maps/10" target="main">

Evacuation Points </a></li>

75
<li><a href

="https://www.google.com/maps/d/embed?mid=12Aj4vs7uBpzTsXXycsNSsj_aos4"

target="main">Shelter Clusters</a></li>

<li>Evacuation Routes

<ul> <li><a href ="http://localhost:8080/geoexplorer/viewer/#maps/11" target="main">

Aranmula </a></li>

<li><a href ="http://localhost:8080/geoexplorer/viewer/#maps/16"

target="main">Chengannur</li>

<li><a href ="http://localhost:8080/geoexplorer/viewer/#maps/17"

target="main">Kuttoor</li>

<li><a href ="http://localhost:8080/geoexplorer/viewer/#maps/18"

target="main">Thiruvanvandoor</li>

<li><a href ="http://localhost:8080/geoexplorer/viewer/#maps/19"

target="main">Thiruvalla</li>

</ul></li></ul> </li>

<li><a href ="https://www.google.com/maps/d/embed?mid=1F3CdfciANE-

5IrcQEtn5ngvdNH4" target="main">Evacuation Routes</a></li>

<li><button onclick="document.getElementById('id01').style.display='block'"

style="width:auto;">Login</button></li>

</ul></div>

<div class="content" >

<iframe

src="https://www.google.com/maps/d/embed?mid=1I2bmTm3jbYL59zSRfeUWNEInaZw"

width="1050" height="550" name="main"></iframe></div>

76
<div id="id01" class="modal">

<form class="modal-content animate" action="check.php" method="post"

target="_blank">

<div class="imgcontainer">

<span onclick="document.getElementById('id01').style.display='none'" class="close"

title="Close Modal">&times;</span>

<img src="mushroom.jpg" alt="Avatar" class="avatar">

</div>

<div class="container">

<label><b>Username</b></label>

<input type="text" placeholder="Enter Username" name="uname" required>

<label><b>Password</b></label>

<input type="password" placeholder="Enter Password" name="psw" required>

<button type="submit">Login</button>

<input type="checkbox" checked="checked"><font color="blue"> Remember me </div>

<div class="container" style="background-color:#f1f1f1">

<button type="button" onclick="document.getElementById('id01').style.display='none'"

class="cancelbtn">Cancel</button>

<span class="psw"><font color="blue">Forgot <a href="#"><font

color="blue">password?</a></span>

</div> </form></div>

<aside>

<br/><br/><br/><br/>

77
<h2><a href="hallo.html" target="_blank"><font color="red">Notifications

</font></a></h2>

<marquee direction ="up" SCROLLDELAY=300><ul>

<li><a href="time.pdf" target="_blank"><font color="blue">

Time to Reach Inundation </font></a></li></br>

<li><a href="clusters.pdf" target="_blank"><font color="green">

Cluster Capacity and Details</font></a></li> </ul>

</marquee>

</br></br>

<h3><font color="red">Details of Facilities:</h3></font></br>

<form action="formtry.php" target="_blank" method="post" autocomplete="on"

autofocus="on">

<b>Facility:</b><input type="text" name="facility">

<b>Panchayath/Municipality:<b/><input type="text" name="panchayath">

<input type="submit">

<input type="reset">

</form></aside><br/></div>

</body>

</html>

A.2 PHP CODE FOR DATABASE CONNECTION

<?php

$host = "host= localhost";

$port="port=5432";

78
$dbname ="dbname=fulldetails";

$credentials= "user=postgres password=pampadam";

$db= pg_connect("$host $port $dbname $credentials");

if(!$db)

{echo "Error unable to open database\n";}

else

{echo "connection established </br>";}

?>

79