role of geospatial technologies in...

ROLE OF GEOSPATIAL TECHNOLOGIES IN

DISASTER MANAGEMENT

By :- Wan Mohd Naim Wan Mohd, PhD

Centre of Studies for Surveying Science and Geomatics Faculty of Architecture, Planning and Surveying

Universiti Teknologi MARA, Shah Alam

9 August 2016

AIM OF PRESENTATION

• To highlight the use of geospatial technology in Disaster Management (Before, During, After)

GEOSPATIAL TECHNOLOGY

GEOSPATIAL TECHNOLOGY

GIS

RS

GNSS

OTHERS

• refers to equipment used in visualization, measurement, and analysis of earth's features

Global Navigation Satellite System

Source : Havell, 2014

Cycle of Disaster Management

Four (4) phases • Prevention and mitigation

(before) • Preparedness (before) • Response (during) • Recovery (after)

(Source : Sudheer, 2014)

Disaster Management ? - organisation and management of resources and responsibilities for dealing with all humanitarian aspects of emergencies, in particular preparedness, response and recovery in order to lessen the impact of disasters (International Federation of Red Cross and Red Crescent Societies)

Current research projects related to Disaster Management

• Development of Landslide Hazard Zonation Mapping

• Evaluation of Various DEMs for flood inundation Modeling

• Development of Integrated Flood Management System

• GIS-based Flood Vulnerability Index

LANDSLIDE HAZARD ZONATION MAPPING

6

BACKGROUND • The increasing population and

expansion of settlements over hilly areas has greatly increased the impact of natural disasters such as landslide.

• Over the years, various techniques and models have been developed to predict landslide hazard zones.

• The development of these models are based on different landslide inducing factors such as:

Main Groups Factors

Ground Condition Geomorphology

Geology

Soil

Land use

Distance Related Roads

River

Drainage density

Faults

Geomorphometry DEM

Slope

Aspect

Elevation

Triggering Rainfall

Earth quakes

7

BACKGROUND • Slope is one of the most important

factor in assessing landslide hazard areas – need high accuracy and high resolution DEM

• LiDAR technology and Geographical Information System (GIS) are important tools in assessing landslide hazards

• Multi-criteria Decision Making (MCDM) Multi-criteria decision making approach also play important role in determining relative importance of landslide factors

Source : Sight Power 8

METHODOLOGY

DEM (LiDAR DTM) DEM (SRTM 30m)

SLOPE 1 LAND USE LITHOLOGY SOIL SLOPE 2

Phase 2 - Landslide Model Development • Expert opinion to rank factors • Modify previously developed models based on only

slope, land use, lithology and soil properties factors (Othman, W. Mohd. N. Surip, 2013)

LHZ Model 1

Phase 5 - Validation of Models

Phase 3 - Data Acquisition

Phase1 - Selection of Study Area Cheras and Kajang (5 x 5 km)

Phase 4 - Data Processing/analysis in GIS Rank Criteria

Calculate Weight and Standardize Score for the criteria used

Generate Landslide Hazard Zone Maps using different models

MODEL 1 MODEL 2 MODEL 3 MODEL 1 MODEL 2 MODEL 3

LHZ Model 3 LHZ Model 2 LHZ Model 1 LHZ Model 3 LHZ Model 2

9

STUDY AREA – PART OF CHERAS AND KAJANG Area Coverage

• Size : 5 x 5 km

• From Cheras to Kajang

• Elevation Range : 20 – 321 m above MSL

• Mukim : Kajang, Semenyih and Cheras

10

DATA COLLECTION

• Digital Terrain Model (DTM) – from LiDAR

• Digital Elevation Model from SRTM – from USGS website

• Soil Properties - derived from soil map

• Land use – Digitised from Orthoimage

• Lithology

11

DATA ACQUISATION FROM LIDAR

EQUIPMENT DETAILS:

• LiDAR System is LiteMapper 6800-400(Riegl 680i-400kHz) • This Laser Scanner is Full Waveform which

has unlimited number of return echoes. • This System comes with high resolution RGB Camera System

60 Mega Pixel and automatic geo-correction system which is equipped with 512kHz Fiber Optic IMU.

DATA ACQUISITION: • Date: 19 December 2014, 30 December 2014 – 3 January 2015 • Requirement RSGIS & JMG for data acquisition: • Helicopter type : Eurocopter EC 120B • Helicopter Speed : 60 knot • Flying Altitude : 600 m AGL • Laser Scan Angle : 600 • PRR laser : 400 kHz (maximum range)

• Data acquisition - Hazard and Slope Risk Mapping Project at Cheras Selatan-Kajang-Bangi-Putrajaya, Selangor for RS & GIS Consultancy Sdn Bhd and Department of Mineral & Geoscience.

12

LiDAR Project Area

13

Digital Terrain Model

14

LiDAR SRTM

DEM – Shuttle Radar Topographic Mission (SRTM) – 30 x 30 m

15

Orthoimage of Study Area

16

Slope map derived from_LiDAR Slope map derived from_SRTM

17

DEVELOPMENT OF LANDSLIDE HAZARD ZONATION MODELS

• Based on earlier studies by Ainon Nisa, Wan Mohd and Noraini Surip

• Study Areas - Ampang Jaya and Hulu Langat

• Technique used – GIS-based Multicriteria Decision Making (MCDM)

18

Landslide Hazard Models Tested Model

No Technique/ Criteria Slp Lu Litho SP Geomor Asp Elev Rf Priv Prd Facc Drg

1 Ranking (Rank Sum) 0.333 0.133 0.267 0.2 0.067

2 Ranking (Rank Reciprocal) 0.438 0.109 0.219 0.146 0.088

3 Ranking (Rank Exponential) 0.454 0.073 0.291 0.164 0.018

4 Rating 0.335 0.168 0.252 0.211 0.034

5 AHP (Expert Opinion) 0.162 0.082 0.116 0.277 0.023 0.061 0.21 0.041 0.032

6

Pairwise Comparison (Expert

Opinion) 0.5 0.036 0.143 0.214 0.107

7

Pairwise Comparison (Expert

Opinion) 0.294 0.088 0.236 0.265 0.029 0.088

8 AHP (Expert Opinion) 0.361 0.113 0.091 0.199 0.141 0.051 0.044

9 AHP (Expert Opinion) 0.301 0.089 0.073 0.152 0.108 0.045 0.037 0.195

Slope (slp) Land use (Lu) Lithology (Litho) Soil Properties (SP)

Geomorphology (Geomor) Aspect (Asp) Elevation (Elev) Rainfall (Rf)

Proximity to river (Priv) Proximity to road (Prd) Flow Accumulation (Facc) Drainage Pattern (Drg)

FACTORS CONSIDERED :

19

Developed Models Model

No Technique Formula

1 Rank Sum 0.333(s_slp) + 0.133(s_lu) + 0.267(s_lit) + 0.2(s_sp) +0.067(s_geomorf)

2 Rank Reciprocal 0.438(s_slp) + 0.109(s_lu) + 0.219(s_lit) + 0.146(s_sp) + 0.088(s_geomorf)

3 Rank Exponent 0.454(s_slp) + 0.073(s_lu) + 0.291(s_lit) + 0.164(s_sp) + 0.018(s_geomorf)

4 Rating 0.335(s_slp) + 0.168(s_lu) + 0.252(s_lit) + 0.211(s_sp) + 0.034(s_geomorf)

5 AHP 0.162(s_slp) + 0.082(s_lu) + 0.116(s_lit) + 0.277(s_sp) + 0.023(s_asp) + 0.061(s_elev) + 0.207(s_rfal) +

0.041 (s_priv) + 0.032(s_prd)

6 Pairwise Comparison 0.5(s_slp) + 0.036(s_lu) +0.143(s_lit) + 0.214(s_sp) + 0.107(s_asp)

7 Pairwise Comparison 0.294(s_slp) + 0.088(s_lu) + 0.029(s_geomorf) + 0.265(s_sp) + 0.236(s_lit) + 0.088(s_flowacc)

8 AHP 0.361(s_slp) + 0.141(s_asp) + 0.091(s_lit) + 0.113(s_lu) + 0.199(s_sp) + 0.051(s_priv)+0.044(s_prd)

9 AHP 0.301(s_slp) + 0.108(s_asp) + 0.073(s_lit) + 0.089(s_lu) +0.152(s_sp) +

0.045(s_priv) + 0.037(s_prd) + 0.195(s_drg)

20

Landslide Hazard Zonation Maps Generated from Model 1, 2 and 3 21

Landslide Hazard Maps Generated from Model 4, 5 and 6 22

Landslide Hazard Maps Generated from Model 7, 8 and 9 23

Comparison between landslide hazard class and landslide historical data – Area Hulu Kelang

24

Models Used – For this study Criteria Considered • Slope • Lithology • Land use • Soil Properties

LHZ (Model 1) = (0.400 x s_slp) + (0.100 x s_lu) + (0.300 x s_litho) + (0.200 x s_sp) ------------(1)

LHZ (Model 2) = (0.347 x s_slp) + (0.219 x s_lu) + (0.218 x s_litho) + (0.174 x s_sp) -----------------------(2)

LHZ (Model 3) = (0.481 x s_slp) + (0.240 x s_lu) + (0.159 x s_litho) + (0.120 x s_sp) -----------------------------------(3)

25

LHZ based on LiDAR data

RESULT – LHZ MAP BASED ON 3 DIFFERENT MODELS

LHZ based on SRTM data 26

Criteria Considered • Slope • Lithology • Land use • Soil Properties LHZ (Model 1) = (0.400 x s_slp) + (0.100 x s_lu) + (0.300 x s_litho) + (0.200 x s_sp) --------(1) LHZ (Model 2) = (0.347 x s_slp) + (0.219 x s_lu) + (0.218 x s_litho) + (0.174 x s_sp) ---(2) LHZ (Model 3) = (0.481 x s_slp) + (0.240 x s_lu) + (0.159 x s_litho)+ (0.120 x s_sp) ---(3)

SITE 1

27

SITE 4 SITE 3

SITE 2

SITE 5

28

SITE 6

• Aim : to evaluate the accuracy of NextMap IFSAR DTM for flood inundation mapping

• Study Area : Padang Terap, Kedah

ACCURACY ASSESSMENT OF DEMS FOR FLOOD INUNDATION MAPPING

STUDY AREA AND DATASETS

NextMap Airborne IFSAR

DTM Orthorectifies Radar Images (ORI) DSM

Source : JPSurvey

methodology

RESULTS

GPS ASTER SRTM GPS ASTER SRTM

(m) DTM (m) DSM (m) (m) (m) (m) DTM (m) DSM (m) (m) (m)

1 2.982 2.469 2.897 11 6 22.328 22.193 24.84 26 31

2 4.223 5.297 5.438 10 4 22.419 22.391 23.593 26 21

3 3.14 3.047 3.495 7 5 20.366 20.638 19.951 21 26

4 1.691 2.092 6.206 8 6 19.717 19.61 24.366 18 21

5 1.821 2.298 3.928 7 7 18.656 19.556 20.24 24 27

6 3.298 3.837 3.667 7 2 28.041 25.134 27.972 29 31

7 3.517 3.819 2.78 8 8 19.3 20.025 19.735 17 26

8 3.551 3.822 3.425 6 5 19.982 19.349 20.069 23 26

9 1.657 2.528 1.03 8 3 22.897 22.599 27.066 28 27

10 2.81 3.689 3.183 14 10 20.19 20.219 22.118 23 26

11 2.482 3.039 3.65 7 7 21.145 21.194 20.893 18 24

12 2.263 2.345 1.738 6 6 19.14 17.361 19.872 18 19

13 2.742 4.378 3.438 9 6 41.192 41.088 41.631 40 42

14 3.012 3.061 3.475 8 10 21.809 21.74 21.94 17 25

15 1.886 3.167 2.97 8 6 21.927 20.09 21.481 17 33

KUALA NERANG

POINT NO

ALOR SETAR

NEXTMap IFSAR NEXTMap IFSAR

Descriptive statistics of the differences between various DEMs and reference DEM

ALOR SETAR (Flat area) KUALA NERANG (Terrain area)

RMSE (m)

Min (m)

Max (m)

RMSE (m)

Min (m)

Max (m)

IFSAR DTM - GPS 0.497 0.049 0.879 0.841 0.029 1.837

IFSAR DSM - GPS 1.529 0.085 4.515 2.092 0.069 4.649

ASTER - GPS 5.848 2.449 11.19 3.278 0.634 5.344

SRTM - GPS 4.268 1.298 7.190 5.300 0.14 8.672

Elevation points from GPS and manually observed from different DEMs

Field Work

Flood inundation map – different discharge values

Flood Inundation Map based on different Cross-section interval

200 meter cross-section interval

Legend

4000 m3

ValueHigh : 7.48727

Low : 7.62939e-006

300 meter cross-section interval 400 meter cross-section interval 500 meter cross-section interval

Accuracy of Flood Inundation Maps

Location Elevation

Water Depth from

Simulated Water

Simulated Water

Water Depth

Photograph (m) Level Depth (m) Differences

Point 1 19.108 1.2 19.702 0.592 -0.608

Point 2 16.805 0.3 16.992 0.186 -0.114

Point 3 14.813 1.65 17.074 2.26 0.61

Point 4 16.216 0.6 16.881 0.66 0.06

Point 5 16.771 2.1 17.665 0.89 -1.21

Point 6 16.297 2 19.315 2.83 0.83

Point 7 15.253 1.45 18.503 3.25 1.8

Point 8 19.000 1.2 20.461 1.456 0.256

Point 9 18.159 2.4 19.598 1.438 -0.962

Point 10 16.679 1.45 18.455 1.77 0.32

Point 11 17.374 1.8 19.675 2.3 0.5

Point 12 19.304 0.9 19.552 0.25 -0.65

Point 13 20.67 0.45 21.95 1.26 0.81

Point 14 15.847 0.6 17.115 1.269 0.669

SYSTEM CAPABILITIES

FLOOD DAMAGE INVENTORY USING

MOBILE DATA COLLECTION

SAMPLE OF FLOOD DAMAGE INVENTORY

IN MOBILE APPS GEOJOT+ APP

Online mode data collection using collector

Features update in ArcGIS Online

Offline mode data collection using collector

DISPLAY DATA

DATA DISPLAY IN GIS SOFTWARE

MANAGEMENT OF EVACUATION CENTRE USING GIS

CLOSEST FACILITIES AND SHORTEST PATH

FLOOD VULNERABILITY INDEX ASSESSMENT USING

GIS-BASED MULTI CRITERIA DECISION MAKING

Factors Considered:

• Social Vulnerability

• Economic Vulnerability

• Infrastructure Vulnerability

• Physical Vulnerability

Location of study area (Google Earth, 2014)

FVI = (SocVul) + (EconVul) + (InfraVul)

+ ( PhyVul)

where,

SocVul – Social Vulnerability

EconVul – Economic Vulnerability

InfraVul – Infrastructure Vulnerability

PhyVul – Physical Vulnerability

FVI = (SocVul) + (EconVul) + (InfraVul) +

( PhyVul)

where,

SocVul – Social Vulnerability

EconVul – Economic Vulnerability

InfraVul – Infrastructure Vulnerability

PhyVul – Physical Vulnerability

FVA maps using Rank Sum method

(a) Social (b) Infrastructure (c) Economic and (d) Physical Vulnerabilities

(a) (b)

(c) (d)

FVA maps using AHP method

(a) Social (b) Infrastructure (c) Economic and (d) Physical Vulnerabilities

(c) (d)

(a) (b)

FVI map - using Rank Sum method

FVI using AHP method

Top 8 most vulnerable mukim based on AHP Method

Top 8 most vulnerable mukim based on Rank Sum Method

SUMMARY

• Geospatial technologies (remote sensing, GIS and GPS) – widely used in all aspects of disaster management.