SlideShare a Scribd company logo
1 of 61
Download to read offline
GIS-based Climate Change Vulnerability Assessment
of the Municipality of Del Carmen, Surigao del
Norte Province, Philippines
Diomedes A. Racelis, Elenita L. Racelis and Angela A. Limpiada
University of the Philippines Los Banos
PRESENTATION OUTLINE
• Country Background: Philippines
• Study Site: Del Carmen, Siargao Island
Surigao del Norte, Mindanao
• Objectives of the Study
• Methodology
• Results of Vulnerability Mapping
• Lessons Learned
The Philippines: Paradise at Risk
The Philippines is the most vulnerable country to climate change in SEA
EEPSEA
List of Most Disaster Risk Countries
World Disaster Risk Index
Storm tracks (1900-2014)
Located in the typhoon belt, average of 18-22 typhoons
a year almost half of which is destructive
36,000 km of
coastlines
RISK TO SEA LEVEL RISE
RUGGED TOPOGRAPHY
RAPID URBANIZATION and OVERPOPULATION
Photo Credit: Skyscrapercity.com
Top 17 Megadiverse Countries
Australia
The Congo
Madagascar
South Africa
China
India
Indonesia
Malaysia
Papua New Guinea
PHILIPPINES
Brazil
Colombia
Ecuador
Mexico
Peru
United States
Venezuela
Study Site: Del Carmen, Siargao Island , Surigao del Sur
MANILA
MINDANAO
SIARGAO
Photo Credit: http://www.mindanaotours.com
Siargao Island is well known as “The Surfing
Capital of the Philippines”.
DEL CARMEN, SIARGAO ISLAND, SURIGAO DEL NORTE
DEL CARMEN
Village Map of Del Carmen
Siargao Island contains the largest mangrove forest reserves in Mindanao, at Del Carmen.
Photo Credit: http://www.mindanaotours.com
Sugba Lagoon Del Carmen, Siargao Island
Photo Credit: http://www.mindanaotours.com
OBJECTIVES OF THE STUDY
Determine using GIS the vulnerability to the
impacts of climate change and climate risks of
ecosystems, communities, and infrastructure in
the municipalities of Del Carmen in Surigao del
Norte
METHODOLOGY
A. Site Reconnaissance
B. Coordination with Stakeholders
C. Gathering of Secondary Data
D. Community Mapping
E. Map Consolidation
F. Scanning
G. Georeferencing
H. Digitizing
I. Generation of Vulnerability Maps
A. Site Reconnaissance: Del Carmen
METHODOLOGY
B. Coordination with Stakeholders
METHODOLOGY
C. Gathering of Secondary Data
METHODOLOGY
• Thematic maps prepared by the GIS team of
CENERGY – topographic, vegetation, land use,
political, etc.
• Natural resources data – forest resources,
agriculture, fisheries
• Socio-economic data – settlement,
population, infrastructure
D. Community Mapping
METHODOLOGY
• Stakeholders were invited to annotate
existing maps as to their first-hand
knowledge on the information shown on the
thematic maps
• Map sheets, transparencies, colored pens,
colored papers, sticky notes were used to
mark details on the draft maps
E. Map Consolidation
METHODOLOGY
• Involved compiling all analog (paper-based)
spatial information prior to GIS mapping.
• In this particular activity, a total of twenty (20)
draft thematic maps for the four (4) selected
municipalities of Siargao Island were received
for conversion to GIS-based digital maps
F. Scanning
METHODOLOGY
• transferred analog (paper) maps into a digital
format the computer can read
• the maps were scanned individually in
preparation for on-screen digitizing.
• each map was placed carefully on ordinary
printer-scanner and scanned to produce a
digital image for subsequent digitizing
F. Georefencing
METHODOLOGY
• involves aligning geographic data to a known
coordinate system so it can be viewed, queried,
and analyzed with other geographic data
• to georeference an image, one first needs to
establish control points, input the known
geographic coordinates of these control points,
choose the coordinate system and other
projection parameters and then minimize
residuals
• Residuals are the difference between the actual
coordinates of the control points and the
coordinates predicted by the geographic model
created using the control points
F. Georefencing
METHODOLOGY
In this project, the scanned maps were georeferenced
individually by clicking the “add data” button in the
ArcMap including the boundary per municipality that
was previously used in earlier maps
F. Georefencing
METHODOLOGY
…georeferencing tool was activated to allow adding of the
control point on the image; … the four (4) control points
were chosen preferably along the corners of the map;
…“update” button was clicked to update the georeferenced
of the image…
F. Digitizing
METHODOLOGY
• analog (or paper) map is converted into a
digital format
• in this particular set of maps, the scanned
images were individually digitized on-screen
by encoding all the relevant information
shown on each image
F. Digitizing
METHODOLOGY
…the specific map features were digitized by clicking in sequence
the ArcMap catalog and add new shape files buttons; …spatial
information such as plants, crops, buildings, and schools were
digitized as “points” while roads, streams, partial boundary were
digitized as “polylines” with the same coordinates as “points”;
Municipal and barangay boundary was digitized as “polygons”…
F. Digitizing
METHODOLOGY
…the digitized map was then assigned a coordinate system
which in this particular project is “WGS 1984 UTM Zone 51 N”…
F. Digitizing
METHODOLOGY
…The finished map is then added as a new shapefile; the GIS
mapping process is completed by editing the map by clicking
“editor” button; this is done by drawing each item per point,
polyline and polygon…
G. Preparation of Vulnerability Maps
1. Storm surge hazard map analysis
The digital elevation model (DEM) of the municipality
was used to derive the storm surge map. Five-meter
elevation interval was generated through interpolation.
METHODOLOGY
G. Preparation of Vulnerability Maps
1. Storm surge hazard map analysis
METHODOLOGY
G. Preparation of Vulnerability Maps
1. Storm surge hazard map analysis
METHODOLOGY
G. Preparation of Vulnerability Maps
2. Drought hazard map analysis
METHODOLOGY
• Drought risk map was generated based on land use map/data
by identifying the agricultural lands (rice, coconut, orchard,
etc.) to be affected and label them as high (red), populated
areas medium (yellow), others low (green).
• Crops and agricultural areas were selected and reclassified as
high while forests and water bodies were reclassified as low
LAND USE DROUGHT
'Arable land, crops mainly cereals and sugar' High
'Crop land mixed with coconut plantation' High
'Cultivated Area mixed with
brushland/grassland'
High
'Mangrove vegetation' nd/grassland' Low
'Unclassified' Low
‘Built-up’ Medium
G. Preparation of Vulnerability Maps
2. Drought hazard map analysis
METHODOLOGY
G. Preparation of Vulnerability Maps
2. Flood and Landslide map analysis
METHODOLOGY
Based on
existing flood
and landslide
map of the
areas and
rainfall data
from PAGASA
MONTH
OBSERVED CHANGE PROJECTED
(1971-2000) (2006-2035)
(2036-
2065)
(2006-2035)
(2036-
2065)
JAN 603.4 22.1 1.9 736.8 615.1
FEB 428.6 12.3 -0.5 481.2 426.3
MAR 369.8 18.7 -23.8 438.9 281.7
APR 203.1 -16.4 -30.2 169.8 141.8
MAY 132.6 -5.5 -9.8 125.3 119.6
JUN 148.6 31.7 -0.4 195.8 148.0
JUL 170.7 10.0 -15.1 187.8 144.9
AUG 136.9 17.5 6.5 160.8 145.8
SEP 165.7 21.7 2.9 201.7 170.5
OCT 267.9 -12.1 26.0 235.5 337.6
NOV 510.8 18.6 18.3 606.0 604.4
DEC 510.8 10.2 41.2 562.7 721.1
Average 304.1 10.7 1.4 341.8 321.4
Average
/30 10.1359906 0.35748333 0.0472675 11.3947482 10.713652
(Averag
e/30)/3
65 0.02776984 0.00097941 0.0001295 0.03121849 0.02935247
G. Preparation of Vulnerability Maps
2. Flood and Landslide map analysis
METHODOLOGY
The average/30 means that the result of the average per year is
divided into 30 which is the number of years that it is projected
(1971-2000, 2000-2020, 2020-2050)
Topo-
corrected
Observed
Hazard
Map
Observed Mean Frequency of Daily Rainfall 100 mm
and above (2020, 2050)
Low (1)
Moderately
Low (2)
Moderate
(3)
Moderately
High (4)
High (5)
Very High
(6)
Low (2) 2 4 6 8 10 12
Moderate
(4) 4 8 12 16 20 24
High (6) 6 12 18 24 30 36
G. Preparation of Vulnerability Maps
2. Flood and Landslide map analysis
METHODOLOGY
…the result of this is divided again into 365 which is the number
of days per year…
0-6 Low
7-12 Moderately Low
13-18 Moderate
19-24 Moderately High
25-30 High
>30 Very High
Average number of daily rainfall (100 mm above)
0-1 Low
1-2 Moderately Low
2-3 Moderate
3-4 Moderately High
4-5 High
>5 Very High
G. Preparation of Vulnerability Maps
2. Flood and Landslide map analysis
METHODOLOGY
G. Preparation of Vulnerability Maps
2. Flood and Landslide map analysis
METHODOLOGY
 Areas affected by storm surge (Population)
 Areas affected by rain-induced landslide (Population)
 Areas affected by drought (Agriculture sector)
Vulnerability Maps for Del Carmen
OTHER GIS-RELATED ACTIVITIES
1. Preparation of hazard map in jpeg format:
–Flood map overlayed with topo map. Map
showing barangay boundaries and other
principal/key features that should be included
in the map.
–Landslide map overlayed with topo map.
Map showing barangay boundaries and other
principal/key features that should be included
in the map.
OTHER GIS-RELATED ACTIVITIES
2. Preparation of overlay/thematic maps and
attribute tables:
• Spatial Extent for Rainfall-induced Landslides
• Exposed Population for Rainfall-induced
Landslides
• Spatial Extent for Flooding
• Exposed Population for Flooding
Agricultural Areas Affected by Drought in Del Carmen
Table 23. Agricultural areas at risk to drought in Del Carmen.
Coconut Corn Rice Root Crops Uncultivated Sub-total Coconut Corn Rice Root Crops Uncultivated Sub-total
Antipolo 41.9 121.2 350.5 513.6 35.8 163.6 199.3 713.0
Bagakay(Alburo) 54.1 54.1 85.8 93.2 53.5 650.9 883.4 937.5
Bitoon 114.8 114.8 32.5 1.1 27.5 170.0 231.2 0.1 346.1
Cabugao 18.3 1.8 159.5 179.6 46.9 199.7 246.6 426.3
Cancohoy 7.6 7.6 4.8 49.0 8.4 535.0 597.2 604.8
Caub 13.0 347.5 360.4 - 0.1 360.5
Del Carmen 30.0 139.9 169.9 13.9 33.6 47.5 217.3
Domoyog 16.7 12.3 614.1 643.2 52.9 52.9 0.1 696.2
Esperanza 333.5 333.5 5.5 46.2 296.6 348.3 681.8
Jamoyaon 470.0 470.0 73.3 24.6 1,036.7 1,134.6 0.2 1,604.7
Katipunan 65.0 65.0 5.9 121.4 13.2 352.9 493.6 558.5
Lobogon 0.5 178.7 106.7 273.0 558.8 - 558.8
Mabuhay 130.0 130.0 4.4 200.3 204.6 334.7
Mahayahay - 6.2 2.7 5.8 150.1 164.9 164.9
n.a. ( 30199) 64.5 64.5 - 64.5
Quezon 44.6 44.6 7.2 198.8 205.9 250.5
San Fernando 646.4 11.3 1,089.9 1,747.6 1,186.3 1,186.3 0.6 2,934.4
San Jose 91.7 91.7 6.8 18.2 158.2 183.2 275.0
Sayak 126.5 143.7 185.5 455.7 3.3 41.8 45.1 500.9
Tuboran 34.9 34.9 1.1 115.9 117.0 152.0
Total 706.6 220.6 266.0 275.9 4,570.4 6,039.4 224.9 259.0 124.8 189.7 5,543.3 6,341.6 1.2 12,382.3
Total
No Risk
Barangay
Agricultural Areas at Risk to Drought (ha)
High
Low
Areas Affected by Rain-induced Landslide
Table 24. Populated areas at risk to rain-induced landslide in Del Carmen.
Populated Unpopulated Populated Unpopulated Populated Unpopulated Populated Unpopulated
Antipolo 7.4 24.0 49.3 290.6 341.7 713.0
Bagakay(Alburo) 3.3 172.4 9.4 126.8 482.4 143.1 937.5
Bitoon 23.9 72.1 1.6 36.9 211.7 346.1
Cabugao 25.7 36.9 99.9 263.8 426.3
Cancohoy 66.0 79.6 93.4 205.7 160.1 604.8
Caub 1.7 46.0 124.6 14.3 173.9 360.5
Del Carmen 132.9 84.4 217.3
Domoyog 67.2 92.0 536.8 696.2
Esperanza 47.9 0.1 107.1 171.6 355.0 681.8
Jamoyaon 73.3 799.2 20.7 124.2 249.7 337.5 1604.7
Katipunan 112.5 41.5 64.7 195.5 144.5 558.5
Lobogon 85.9 36.0 401.4 35.5 558.8
Mabuhay 1.9 11.1 42.9 278.7 334.7
Mahayahay 37.9 125.1 1.9 164.9
n.a. ( 30199) 64.5 64.5
Quezon 10.0 24.3 207.9 8.3 250.5
San Fernando 7.0 15.5 21.5 334.4 2556.0 2934.4
San Jose 0.0 6.3 192.1 76.5 275.0
Sayak 14.0 34.0 407.6 45.2 500.9
Tuboran 8.2 7.0 86.4 1.7 48.7 152.0
Total 78.6 1,224.2 31.3 473.9 136.7 1,644.2 2,925.2 5,867.0 12,382.3
Total
Barangay Moderately Low Moderately High Very High No Risk
Populated Areas at Risk to Rain-Induced Landslide (ha)
Areas Affected by Storm Surge in Del Carmen
Table 25. Populated areas at risk to storm surge in Del Carmen.
High
Populated Unpopulated Populated Unpopulated None Populated Unpopulated
Antipolo 46.5 83.5 7.8 298.0 277.1 713.0
Bagakay(Alburo) 8.5 28.2 4.7 485.7 410.4 937.5
Bitoon 27.8 19.1 5.0 60.8 233.5 346.1
Cabugao 16.5 31.5 5.6 99.9 272.9 426.3
Cancohoy 0.0 205.7 399.1 604.8
Caub 20.4 17.8 52.7 60.4 209.4 360.5
Del Carmen 30.0 4.1 49.3 0.0 0.4 53.6 79.9 217.3
Domoyog 38.9 121.6 5.3 530.4 696.2
Esperanza 23.6 70.3 1.4 148.0 438.4 681.8
Jamoyaon 41.7 95.3 5.5 323.1 1,139.2 1,604.7
Katipunan 0.6 0.4 195.5 362.0 558.5
Lobogon 401.4 157.4 558.8
Mabuhay 13.7 51.4 0.9 44.8 224.0 334.7
Mahayahay 37.9 127.0 164.9
n.a. ( 30199) 13.3 9.7 18.0 23.5 64.5
Quezon 24.3 226.3 250.5
San Fernando 315.0 1,024.1 147.1 21.5 1,426.7 2,934.4
San Jose 12.0 3.8 28.3 3.9 1.0 151.9 74.0 275.0
Sayak 1.3 0.8 407.6 91.1 500.9
Tuboran 8.7 143.3 152.0
Low Medium None
Barangay
Populated Areas at Risk to Storm Surge (ha)
Total
LESSONS LEARNED
• Local Government Units (LGUs) play an
important role in mainstreaming Climate Change
Adaptation (CCA) strategies
• Attitude, commitment and competence of Local
Chief Executives (LCEs) determine the level of
success in Climate Change Adaptation strategies
• Local awareness and appreciation of the value of
CCA determine the extent of support afforded
by communities
• Local and indigenous knowledge complement
science-based solutions to CC impacts
ACKNOWLEDGEMENT
The authors wish to thank the following:
• Philippine Climate Change Commission and
the Climate Change Office
• Global Green Growth Initiatives
• Local Government of Del Carmen, Surigao del
Norte
• University of the Philippines Los Banos
• Philippine Department of Science and
Technology
THANK YOU!!!

More Related Content

Similar to Racelisd agriculture and mapping for food

Application of Geo-informatics in Environmental Management
Application of Geo-informatics in Environmental ManagementApplication of Geo-informatics in Environmental Management
Application of Geo-informatics in Environmental Management
MahaMadhu2
 
Geoland-EUROGEO conference 2022.pptx
Geoland-EUROGEO conference 2022.pptxGeoland-EUROGEO conference 2022.pptx
Geoland-EUROGEO conference 2022.pptx
Luc Zwartjes
 

Similar to Racelisd agriculture and mapping for food (20)

Process of hazard mapping
Process of hazard mappingProcess of hazard mapping
Process of hazard mapping
 
The 4th NENA workshop - Sudan
The 4th NENA workshop - SudanThe 4th NENA workshop - Sudan
The 4th NENA workshop - Sudan
 
Application of Geo-informatics in Environmental Management
Application of Geo-informatics in Environmental ManagementApplication of Geo-informatics in Environmental Management
Application of Geo-informatics in Environmental Management
 
Geoland-EUROGEO conference 2022.pptx
Geoland-EUROGEO conference 2022.pptxGeoland-EUROGEO conference 2022.pptx
Geoland-EUROGEO conference 2022.pptx
 
Presentation of Academic Writting .pdf
Presentation of Academic Writting .pdfPresentation of Academic Writting .pdf
Presentation of Academic Writting .pdf
 
Geographic information system (GIS) and its application in precision farming
Geographic information system (GIS) and its application in precision farmingGeographic information system (GIS) and its application in precision farming
Geographic information system (GIS) and its application in precision farming
 
Geographical information system
Geographical information system Geographical information system
Geographical information system
 
IRJET- Investigation of Changes in Water, Residential, Bare Land and Green Ar...
IRJET- Investigation of Changes in Water, Residential, Bare Land and Green Ar...IRJET- Investigation of Changes in Water, Residential, Bare Land and Green Ar...
IRJET- Investigation of Changes in Water, Residential, Bare Land and Green Ar...
 
Geohazard mapping and assessment program
Geohazard mapping and assessment programGeohazard mapping and assessment program
Geohazard mapping and assessment program
 
XIIa - Sudan Soil Information System (SUSIS)
XIIa - Sudan Soil Information System (SUSIS)XIIa - Sudan Soil Information System (SUSIS)
XIIa - Sudan Soil Information System (SUSIS)
 
GIS and Remote Sensing
GIS and Remote SensingGIS and Remote Sensing
GIS and Remote Sensing
 
DEM-based Methods for Flood Risk Mapping at Large Scale
DEM-based Methods for Flood Risk Mapping at Large ScaleDEM-based Methods for Flood Risk Mapping at Large Scale
DEM-based Methods for Flood Risk Mapping at Large Scale
 
Selection of Suitable Drainage Pattern using Remote Sensing and GIS Technique...
Selection of Suitable Drainage Pattern using Remote Sensing and GIS Technique...Selection of Suitable Drainage Pattern using Remote Sensing and GIS Technique...
Selection of Suitable Drainage Pattern using Remote Sensing and GIS Technique...
 
GIS Tech in INETER China-Taiwan 2007
GIS Tech in INETER China-Taiwan 2007GIS Tech in INETER China-Taiwan 2007
GIS Tech in INETER China-Taiwan 2007
 
REMOTE SENSING & GIS APPLICATIONS IN WATERSHED MANAGEMENT
REMOTE SENSING & GIS APPLICATIONS IN WATERSHED MANAGEMENT REMOTE SENSING & GIS APPLICATIONS IN WATERSHED MANAGEMENT
REMOTE SENSING & GIS APPLICATIONS IN WATERSHED MANAGEMENT
 
Remote Sensing and GIS and GPS
Remote Sensing and GIS and GPSRemote Sensing and GIS and GPS
Remote Sensing and GIS and GPS
 
wasteland mapping
wasteland mappingwasteland mapping
wasteland mapping
 
Applications of gis
Applications of gisApplications of gis
Applications of gis
 
Applications of gis
Applications of gisApplications of gis
Applications of gis
 
Land use and land cover of Ethiopia
Land use and land cover of Ethiopia Land use and land cover of Ethiopia
Land use and land cover of Ethiopia
 

More from ssusera9462d1

More from ssusera9462d1 (11)

QGIS tutorial for GIS and mapping technique
QGIS tutorial for GIS and mapping techniqueQGIS tutorial for GIS and mapping technique
QGIS tutorial for GIS and mapping technique
 
KKN Sampah_Universitas dan pengelolaan nya
KKN Sampah_Universitas dan pengelolaan nyaKKN Sampah_Universitas dan pengelolaan nya
KKN Sampah_Universitas dan pengelolaan nya
 
Conceptual Framing Landscape Approach in the field
Conceptual Framing Landscape Approach in the fieldConceptual Framing Landscape Approach in the field
Conceptual Framing Landscape Approach in the field
 
Local Government Showcase Drones for the planning
Local Government Showcase Drones  for the planningLocal Government Showcase Drones  for the planning
Local Government Showcase Drones for the planning
 
Blue Carbon February 2020_When Environmentalists
Blue Carbon February 2020_When EnvironmentalistsBlue Carbon February 2020_When Environmentalists
Blue Carbon February 2020_When Environmentalists
 
Drone UAV Mapping in Agriculture and result
Drone UAV Mapping in Agriculture and resultDrone UAV Mapping in Agriculture and result
Drone UAV Mapping in Agriculture and result
 
Conceptual_framing_of_the_landscape_approach.pdf
Conceptual_framing_of_the_landscape_approach.pdfConceptual_framing_of_the_landscape_approach.pdf
Conceptual_framing_of_the_landscape_approach.pdf
 
Pemetaan Foto Udara Menggunakan Drone.pdf
Pemetaan Foto Udara Menggunakan Drone.pdfPemetaan Foto Udara Menggunakan Drone.pdf
Pemetaan Foto Udara Menggunakan Drone.pdf
 
2-Camera-Properties.pdf
2-Camera-Properties.pdf2-Camera-Properties.pdf
2-Camera-Properties.pdf
 
scuba.pdf
scuba.pdfscuba.pdf
scuba.pdf
 
SWOT Kabaena - INFO Grafik
SWOT Kabaena - INFO GrafikSWOT Kabaena - INFO Grafik
SWOT Kabaena - INFO Grafik
 

Recently uploaded

Recently uploaded (20)

COMMUNICATING NEGATIVE NEWS - APPROACHES .pptx
COMMUNICATING NEGATIVE NEWS - APPROACHES .pptxCOMMUNICATING NEGATIVE NEWS - APPROACHES .pptx
COMMUNICATING NEGATIVE NEWS - APPROACHES .pptx
 
General Principles of Intellectual Property: Concepts of Intellectual Proper...
General Principles of Intellectual Property: Concepts of Intellectual  Proper...General Principles of Intellectual Property: Concepts of Intellectual  Proper...
General Principles of Intellectual Property: Concepts of Intellectual Proper...
 
Food safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdfFood safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdf
 
How to Add New Custom Addons Path in Odoo 17
How to Add New Custom Addons Path in Odoo 17How to Add New Custom Addons Path in Odoo 17
How to Add New Custom Addons Path in Odoo 17
 
Basic Civil Engineering first year Notes- Chapter 4 Building.pptx
Basic Civil Engineering first year Notes- Chapter 4 Building.pptxBasic Civil Engineering first year Notes- Chapter 4 Building.pptx
Basic Civil Engineering first year Notes- Chapter 4 Building.pptx
 
SOC 101 Demonstration of Learning Presentation
SOC 101 Demonstration of Learning PresentationSOC 101 Demonstration of Learning Presentation
SOC 101 Demonstration of Learning Presentation
 
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdfUGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
 
Unit 3 Emotional Intelligence and Spiritual Intelligence.pdf
Unit 3 Emotional Intelligence and Spiritual Intelligence.pdfUnit 3 Emotional Intelligence and Spiritual Intelligence.pdf
Unit 3 Emotional Intelligence and Spiritual Intelligence.pdf
 
On_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptx
On_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptxOn_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptx
On_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptx
 
Plant propagation: Sexual and Asexual propapagation.pptx
Plant propagation: Sexual and Asexual propapagation.pptxPlant propagation: Sexual and Asexual propapagation.pptx
Plant propagation: Sexual and Asexual propapagation.pptx
 
Fostering Friendships - Enhancing Social Bonds in the Classroom
Fostering Friendships - Enhancing Social Bonds  in the ClassroomFostering Friendships - Enhancing Social Bonds  in the Classroom
Fostering Friendships - Enhancing Social Bonds in the Classroom
 
On National Teacher Day, meet the 2024-25 Kenan Fellows
On National Teacher Day, meet the 2024-25 Kenan FellowsOn National Teacher Day, meet the 2024-25 Kenan Fellows
On National Teacher Day, meet the 2024-25 Kenan Fellows
 
Graduate Outcomes Presentation Slides - English
Graduate Outcomes Presentation Slides - EnglishGraduate Outcomes Presentation Slides - English
Graduate Outcomes Presentation Slides - English
 
Mehran University Newsletter Vol-X, Issue-I, 2024
Mehran University Newsletter Vol-X, Issue-I, 2024Mehran University Newsletter Vol-X, Issue-I, 2024
Mehran University Newsletter Vol-X, Issue-I, 2024
 
How to Manage Global Discount in Odoo 17 POS
How to Manage Global Discount in Odoo 17 POSHow to Manage Global Discount in Odoo 17 POS
How to Manage Global Discount in Odoo 17 POS
 
Accessible Digital Futures project (20/03/2024)
Accessible Digital Futures project (20/03/2024)Accessible Digital Futures project (20/03/2024)
Accessible Digital Futures project (20/03/2024)
 
HMCS Max Bernays Pre-Deployment Brief (May 2024).pptx
HMCS Max Bernays Pre-Deployment Brief (May 2024).pptxHMCS Max Bernays Pre-Deployment Brief (May 2024).pptx
HMCS Max Bernays Pre-Deployment Brief (May 2024).pptx
 
This PowerPoint helps students to consider the concept of infinity.
This PowerPoint helps students to consider the concept of infinity.This PowerPoint helps students to consider the concept of infinity.
This PowerPoint helps students to consider the concept of infinity.
 
FSB Advising Checklist - Orientation 2024
FSB Advising Checklist - Orientation 2024FSB Advising Checklist - Orientation 2024
FSB Advising Checklist - Orientation 2024
 
Single or Multiple melodic lines structure
Single or Multiple melodic lines structureSingle or Multiple melodic lines structure
Single or Multiple melodic lines structure
 

Racelisd agriculture and mapping for food

  • 1. GIS-based Climate Change Vulnerability Assessment of the Municipality of Del Carmen, Surigao del Norte Province, Philippines Diomedes A. Racelis, Elenita L. Racelis and Angela A. Limpiada University of the Philippines Los Banos
  • 2. PRESENTATION OUTLINE • Country Background: Philippines • Study Site: Del Carmen, Siargao Island Surigao del Norte, Mindanao • Objectives of the Study • Methodology • Results of Vulnerability Mapping • Lessons Learned
  • 4. The Philippines is the most vulnerable country to climate change in SEA EEPSEA
  • 5. List of Most Disaster Risk Countries
  • 7.
  • 8. Storm tracks (1900-2014) Located in the typhoon belt, average of 18-22 typhoons a year almost half of which is destructive
  • 10. RISK TO SEA LEVEL RISE
  • 11.
  • 12.
  • 13.
  • 15. RAPID URBANIZATION and OVERPOPULATION Photo Credit: Skyscrapercity.com
  • 16. Top 17 Megadiverse Countries Australia The Congo Madagascar South Africa China India Indonesia Malaysia Papua New Guinea PHILIPPINES Brazil Colombia Ecuador Mexico Peru United States Venezuela
  • 17. Study Site: Del Carmen, Siargao Island , Surigao del Sur MANILA MINDANAO SIARGAO
  • 18. Photo Credit: http://www.mindanaotours.com Siargao Island is well known as “The Surfing Capital of the Philippines”.
  • 19. DEL CARMEN, SIARGAO ISLAND, SURIGAO DEL NORTE DEL CARMEN
  • 20.
  • 21. Village Map of Del Carmen
  • 22. Siargao Island contains the largest mangrove forest reserves in Mindanao, at Del Carmen.
  • 24. Sugba Lagoon Del Carmen, Siargao Island Photo Credit: http://www.mindanaotours.com
  • 25. OBJECTIVES OF THE STUDY Determine using GIS the vulnerability to the impacts of climate change and climate risks of ecosystems, communities, and infrastructure in the municipalities of Del Carmen in Surigao del Norte
  • 26. METHODOLOGY A. Site Reconnaissance B. Coordination with Stakeholders C. Gathering of Secondary Data D. Community Mapping E. Map Consolidation F. Scanning G. Georeferencing H. Digitizing I. Generation of Vulnerability Maps
  • 27. A. Site Reconnaissance: Del Carmen METHODOLOGY
  • 28. B. Coordination with Stakeholders METHODOLOGY
  • 29. C. Gathering of Secondary Data METHODOLOGY • Thematic maps prepared by the GIS team of CENERGY – topographic, vegetation, land use, political, etc. • Natural resources data – forest resources, agriculture, fisheries • Socio-economic data – settlement, population, infrastructure
  • 30. D. Community Mapping METHODOLOGY • Stakeholders were invited to annotate existing maps as to their first-hand knowledge on the information shown on the thematic maps • Map sheets, transparencies, colored pens, colored papers, sticky notes were used to mark details on the draft maps
  • 31. E. Map Consolidation METHODOLOGY • Involved compiling all analog (paper-based) spatial information prior to GIS mapping. • In this particular activity, a total of twenty (20) draft thematic maps for the four (4) selected municipalities of Siargao Island were received for conversion to GIS-based digital maps
  • 32. F. Scanning METHODOLOGY • transferred analog (paper) maps into a digital format the computer can read • the maps were scanned individually in preparation for on-screen digitizing. • each map was placed carefully on ordinary printer-scanner and scanned to produce a digital image for subsequent digitizing
  • 33. F. Georefencing METHODOLOGY • involves aligning geographic data to a known coordinate system so it can be viewed, queried, and analyzed with other geographic data • to georeference an image, one first needs to establish control points, input the known geographic coordinates of these control points, choose the coordinate system and other projection parameters and then minimize residuals • Residuals are the difference between the actual coordinates of the control points and the coordinates predicted by the geographic model created using the control points
  • 34. F. Georefencing METHODOLOGY In this project, the scanned maps were georeferenced individually by clicking the “add data” button in the ArcMap including the boundary per municipality that was previously used in earlier maps
  • 35. F. Georefencing METHODOLOGY …georeferencing tool was activated to allow adding of the control point on the image; … the four (4) control points were chosen preferably along the corners of the map; …“update” button was clicked to update the georeferenced of the image…
  • 36. F. Digitizing METHODOLOGY • analog (or paper) map is converted into a digital format • in this particular set of maps, the scanned images were individually digitized on-screen by encoding all the relevant information shown on each image
  • 37. F. Digitizing METHODOLOGY …the specific map features were digitized by clicking in sequence the ArcMap catalog and add new shape files buttons; …spatial information such as plants, crops, buildings, and schools were digitized as “points” while roads, streams, partial boundary were digitized as “polylines” with the same coordinates as “points”; Municipal and barangay boundary was digitized as “polygons”…
  • 38. F. Digitizing METHODOLOGY …the digitized map was then assigned a coordinate system which in this particular project is “WGS 1984 UTM Zone 51 N”…
  • 39. F. Digitizing METHODOLOGY …The finished map is then added as a new shapefile; the GIS mapping process is completed by editing the map by clicking “editor” button; this is done by drawing each item per point, polyline and polygon…
  • 40. G. Preparation of Vulnerability Maps 1. Storm surge hazard map analysis The digital elevation model (DEM) of the municipality was used to derive the storm surge map. Five-meter elevation interval was generated through interpolation. METHODOLOGY
  • 41. G. Preparation of Vulnerability Maps 1. Storm surge hazard map analysis METHODOLOGY
  • 42. G. Preparation of Vulnerability Maps 1. Storm surge hazard map analysis METHODOLOGY
  • 43. G. Preparation of Vulnerability Maps 2. Drought hazard map analysis METHODOLOGY • Drought risk map was generated based on land use map/data by identifying the agricultural lands (rice, coconut, orchard, etc.) to be affected and label them as high (red), populated areas medium (yellow), others low (green). • Crops and agricultural areas were selected and reclassified as high while forests and water bodies were reclassified as low LAND USE DROUGHT 'Arable land, crops mainly cereals and sugar' High 'Crop land mixed with coconut plantation' High 'Cultivated Area mixed with brushland/grassland' High 'Mangrove vegetation' nd/grassland' Low 'Unclassified' Low ‘Built-up’ Medium
  • 44. G. Preparation of Vulnerability Maps 2. Drought hazard map analysis METHODOLOGY
  • 45. G. Preparation of Vulnerability Maps 2. Flood and Landslide map analysis METHODOLOGY Based on existing flood and landslide map of the areas and rainfall data from PAGASA MONTH OBSERVED CHANGE PROJECTED (1971-2000) (2006-2035) (2036- 2065) (2006-2035) (2036- 2065) JAN 603.4 22.1 1.9 736.8 615.1 FEB 428.6 12.3 -0.5 481.2 426.3 MAR 369.8 18.7 -23.8 438.9 281.7 APR 203.1 -16.4 -30.2 169.8 141.8 MAY 132.6 -5.5 -9.8 125.3 119.6 JUN 148.6 31.7 -0.4 195.8 148.0 JUL 170.7 10.0 -15.1 187.8 144.9 AUG 136.9 17.5 6.5 160.8 145.8 SEP 165.7 21.7 2.9 201.7 170.5 OCT 267.9 -12.1 26.0 235.5 337.6 NOV 510.8 18.6 18.3 606.0 604.4 DEC 510.8 10.2 41.2 562.7 721.1 Average 304.1 10.7 1.4 341.8 321.4 Average /30 10.1359906 0.35748333 0.0472675 11.3947482 10.713652 (Averag e/30)/3 65 0.02776984 0.00097941 0.0001295 0.03121849 0.02935247
  • 46. G. Preparation of Vulnerability Maps 2. Flood and Landslide map analysis METHODOLOGY The average/30 means that the result of the average per year is divided into 30 which is the number of years that it is projected (1971-2000, 2000-2020, 2020-2050) Topo- corrected Observed Hazard Map Observed Mean Frequency of Daily Rainfall 100 mm and above (2020, 2050) Low (1) Moderately Low (2) Moderate (3) Moderately High (4) High (5) Very High (6) Low (2) 2 4 6 8 10 12 Moderate (4) 4 8 12 16 20 24 High (6) 6 12 18 24 30 36
  • 47. G. Preparation of Vulnerability Maps 2. Flood and Landslide map analysis METHODOLOGY …the result of this is divided again into 365 which is the number of days per year… 0-6 Low 7-12 Moderately Low 13-18 Moderate 19-24 Moderately High 25-30 High >30 Very High Average number of daily rainfall (100 mm above) 0-1 Low 1-2 Moderately Low 2-3 Moderate 3-4 Moderately High 4-5 High >5 Very High
  • 48. G. Preparation of Vulnerability Maps 2. Flood and Landslide map analysis METHODOLOGY
  • 49. G. Preparation of Vulnerability Maps 2. Flood and Landslide map analysis METHODOLOGY
  • 50.  Areas affected by storm surge (Population)  Areas affected by rain-induced landslide (Population)  Areas affected by drought (Agriculture sector) Vulnerability Maps for Del Carmen
  • 51. OTHER GIS-RELATED ACTIVITIES 1. Preparation of hazard map in jpeg format: –Flood map overlayed with topo map. Map showing barangay boundaries and other principal/key features that should be included in the map. –Landslide map overlayed with topo map. Map showing barangay boundaries and other principal/key features that should be included in the map.
  • 52. OTHER GIS-RELATED ACTIVITIES 2. Preparation of overlay/thematic maps and attribute tables: • Spatial Extent for Rainfall-induced Landslides • Exposed Population for Rainfall-induced Landslides • Spatial Extent for Flooding • Exposed Population for Flooding
  • 53. Agricultural Areas Affected by Drought in Del Carmen
  • 54. Table 23. Agricultural areas at risk to drought in Del Carmen. Coconut Corn Rice Root Crops Uncultivated Sub-total Coconut Corn Rice Root Crops Uncultivated Sub-total Antipolo 41.9 121.2 350.5 513.6 35.8 163.6 199.3 713.0 Bagakay(Alburo) 54.1 54.1 85.8 93.2 53.5 650.9 883.4 937.5 Bitoon 114.8 114.8 32.5 1.1 27.5 170.0 231.2 0.1 346.1 Cabugao 18.3 1.8 159.5 179.6 46.9 199.7 246.6 426.3 Cancohoy 7.6 7.6 4.8 49.0 8.4 535.0 597.2 604.8 Caub 13.0 347.5 360.4 - 0.1 360.5 Del Carmen 30.0 139.9 169.9 13.9 33.6 47.5 217.3 Domoyog 16.7 12.3 614.1 643.2 52.9 52.9 0.1 696.2 Esperanza 333.5 333.5 5.5 46.2 296.6 348.3 681.8 Jamoyaon 470.0 470.0 73.3 24.6 1,036.7 1,134.6 0.2 1,604.7 Katipunan 65.0 65.0 5.9 121.4 13.2 352.9 493.6 558.5 Lobogon 0.5 178.7 106.7 273.0 558.8 - 558.8 Mabuhay 130.0 130.0 4.4 200.3 204.6 334.7 Mahayahay - 6.2 2.7 5.8 150.1 164.9 164.9 n.a. ( 30199) 64.5 64.5 - 64.5 Quezon 44.6 44.6 7.2 198.8 205.9 250.5 San Fernando 646.4 11.3 1,089.9 1,747.6 1,186.3 1,186.3 0.6 2,934.4 San Jose 91.7 91.7 6.8 18.2 158.2 183.2 275.0 Sayak 126.5 143.7 185.5 455.7 3.3 41.8 45.1 500.9 Tuboran 34.9 34.9 1.1 115.9 117.0 152.0 Total 706.6 220.6 266.0 275.9 4,570.4 6,039.4 224.9 259.0 124.8 189.7 5,543.3 6,341.6 1.2 12,382.3 Total No Risk Barangay Agricultural Areas at Risk to Drought (ha) High Low
  • 55. Areas Affected by Rain-induced Landslide
  • 56. Table 24. Populated areas at risk to rain-induced landslide in Del Carmen. Populated Unpopulated Populated Unpopulated Populated Unpopulated Populated Unpopulated Antipolo 7.4 24.0 49.3 290.6 341.7 713.0 Bagakay(Alburo) 3.3 172.4 9.4 126.8 482.4 143.1 937.5 Bitoon 23.9 72.1 1.6 36.9 211.7 346.1 Cabugao 25.7 36.9 99.9 263.8 426.3 Cancohoy 66.0 79.6 93.4 205.7 160.1 604.8 Caub 1.7 46.0 124.6 14.3 173.9 360.5 Del Carmen 132.9 84.4 217.3 Domoyog 67.2 92.0 536.8 696.2 Esperanza 47.9 0.1 107.1 171.6 355.0 681.8 Jamoyaon 73.3 799.2 20.7 124.2 249.7 337.5 1604.7 Katipunan 112.5 41.5 64.7 195.5 144.5 558.5 Lobogon 85.9 36.0 401.4 35.5 558.8 Mabuhay 1.9 11.1 42.9 278.7 334.7 Mahayahay 37.9 125.1 1.9 164.9 n.a. ( 30199) 64.5 64.5 Quezon 10.0 24.3 207.9 8.3 250.5 San Fernando 7.0 15.5 21.5 334.4 2556.0 2934.4 San Jose 0.0 6.3 192.1 76.5 275.0 Sayak 14.0 34.0 407.6 45.2 500.9 Tuboran 8.2 7.0 86.4 1.7 48.7 152.0 Total 78.6 1,224.2 31.3 473.9 136.7 1,644.2 2,925.2 5,867.0 12,382.3 Total Barangay Moderately Low Moderately High Very High No Risk Populated Areas at Risk to Rain-Induced Landslide (ha)
  • 57. Areas Affected by Storm Surge in Del Carmen
  • 58. Table 25. Populated areas at risk to storm surge in Del Carmen. High Populated Unpopulated Populated Unpopulated None Populated Unpopulated Antipolo 46.5 83.5 7.8 298.0 277.1 713.0 Bagakay(Alburo) 8.5 28.2 4.7 485.7 410.4 937.5 Bitoon 27.8 19.1 5.0 60.8 233.5 346.1 Cabugao 16.5 31.5 5.6 99.9 272.9 426.3 Cancohoy 0.0 205.7 399.1 604.8 Caub 20.4 17.8 52.7 60.4 209.4 360.5 Del Carmen 30.0 4.1 49.3 0.0 0.4 53.6 79.9 217.3 Domoyog 38.9 121.6 5.3 530.4 696.2 Esperanza 23.6 70.3 1.4 148.0 438.4 681.8 Jamoyaon 41.7 95.3 5.5 323.1 1,139.2 1,604.7 Katipunan 0.6 0.4 195.5 362.0 558.5 Lobogon 401.4 157.4 558.8 Mabuhay 13.7 51.4 0.9 44.8 224.0 334.7 Mahayahay 37.9 127.0 164.9 n.a. ( 30199) 13.3 9.7 18.0 23.5 64.5 Quezon 24.3 226.3 250.5 San Fernando 315.0 1,024.1 147.1 21.5 1,426.7 2,934.4 San Jose 12.0 3.8 28.3 3.9 1.0 151.9 74.0 275.0 Sayak 1.3 0.8 407.6 91.1 500.9 Tuboran 8.7 143.3 152.0 Low Medium None Barangay Populated Areas at Risk to Storm Surge (ha) Total
  • 59. LESSONS LEARNED • Local Government Units (LGUs) play an important role in mainstreaming Climate Change Adaptation (CCA) strategies • Attitude, commitment and competence of Local Chief Executives (LCEs) determine the level of success in Climate Change Adaptation strategies • Local awareness and appreciation of the value of CCA determine the extent of support afforded by communities • Local and indigenous knowledge complement science-based solutions to CC impacts
  • 60. ACKNOWLEDGEMENT The authors wish to thank the following: • Philippine Climate Change Commission and the Climate Change Office • Global Green Growth Initiatives • Local Government of Del Carmen, Surigao del Norte • University of the Philippines Los Banos • Philippine Department of Science and Technology