Assessing the Vulnerability of Agricultural Crops to Riverine Floods

in Kalibo, Philippines using Composite Index Method

Ronalyn Jose

, Ransie Joy Apura

, Daniel Marc dela Torre

, Ariel Blanco

1,2

Patricia Kristen Dela Cruz

, Therese Anne Rollan

, Eric Luis Tañada

, Joyce Anne Laurete

Nerissa Gatdula

and Edgardo Macatulad

Phil-LiDAR 2 Project 1 Agricultural Resources Extraction (PARMap) from LiDAR Surveys, Training Center for Applied

Geodesy and Photogrammetry, University of the Philippines Diliman, Quezon City, Philippines

Department of Geodetic Engineering, College of Engineering, University of the Philippines Diliman,

Quezon City, Philippines

Phil-LiDAR 1 Data Archiving and Distribution, Training Center for Applied Geodesy and Photogrammetry,

University of the Philippines Diliman, Quezon City, Philippines

Keywords: Vulnerability Assessment, Agriculture, River Flood, Analytic Hierarchy Process, GIS, Composite Index

Method.

Abstract: Evaluating the vulnerability of a system can serve as an effective planning tool in increasing resilience to a

certain hazard. In this study, a vulnerability assessment of agricultural crops to river flooding in Kalibo, a

municipality in Aklan, Philippines, was performed. The analysis included physical, agro-ecological, and

socio-economic indicators clustered under the components of exposure, sensitivity, and adaptive capacity.

Indicators relevant for a composite index measuring degree of vulnerability to flooding were identified and

corresponding weights were determined using Analytic Hierarchy Process (AHP). Various datasets were

acquired using Light Detection and Ranging (LiDAR) remote sensing and participatory methods such as focus

group discussions (FGDs) and key informant interviews (KIIs). The barangay-level (village-level) and

gridded (500m x 500m) vulnerability maps produced using Index Method and GIS were validated through

field surveys and by comparison with historical accounts of disasters and their corresponding impacts on

agricultural productivity. It was concluded that the most exposed barangays were those near bodies of water

and having vast agricultural land cover. Though the physical and environmental attributes of an area are

substantial in determining risk, the vulnerability of the subject area was shown to be influenced by its internal

(exposure) and external (sensitivity and adaptive capacity) factors. Thus, knowing and acting on indicators

that are within human influence is essential in minimizing the effects of inevitable and uncontrollable

phenomena.

1 INTRODUCTION

Climate change involves processes that are complex

and diversified. However, it is mainly characterized

by the intensification of the water cycle, which results

in the occurrence of extreme weather phenomena like

excessive rainfall, flooding, storms, drought, etc.

(Heng et al., 2013).

Due to the Philippines’ geographical location and

physical environment, it has become highly

vulnerable to the impacts of natural disasters,

including the global incidents of the effects of climate

change (Senate Economic Planning Office [SEPO],

2013). According to the SEPO (2013), the Philippines

is one of the most hazard-prone countries in the

world. In the recent World Risk Index published by

the United Nations University - Institute of

Environment and Human Security [UNU-EHS]

(2016), the Philippines ranked third as the most

disaster risk country worldwide with a Risk Index of

26.70 percent. Risk Index is the computed disaster

risk of a country taking into account its exposure,

vulnerability, susceptability, coping capabilities and

adaptive capacities to natural disasters such as storms,

184

Jose, R., Apura, R., Torre, D., Blanco, A., Cruz, P., Rollan, T., Tañada, E., Laurete, J., Gatdula, N. and Macatulad, E.

Assessing the Vulnerability of Agricultural Crops to Riverine Floods in Kalibo, Philippines using Composite Index Method.

DOI: 10.5220/0006284601840194

In Proceedings of the 3rd International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2017), pages 184-194

ISBN: 978-989-758-252-3

flooding, earthquakes, droughts and sea level rise

(UNU-EHS, 2016).

In the context of climate change, the impact of

flooding on socio-ecological systems is of global

significance (Doch et al., 2015). Flooding is one of

the most frequent, typical, and costly natural disasters

which causes abrupt damage on these systems. It

occurs when a body of water rises or overflows

beyond its normal confines, which then causes

inundation in adjoining areas that are usually dry

lands (Doch et al., 2015).

Modifications in rainfall pattern have a very

significant effect on the water level, especially of

basins (Heng et al., 2013). Intensified increase in the

water level may cause run-off, resulting in flooding

of nearby zones. When this phenomenon, i.e.,

sustained heavy rainfall over a specific river basin

takes place, river flood materializes.

In the last decade, the Centre for Research on the

Epidemiology of Disasters [CRED] (2016) recorded

around 196 significant damaging natural disasters in

the Philippines – 72 of which were flooding. Of the

72 flooding events, 41 were classified as riverine

flooding. River flooding alone resulted in 565

casualties and an estimated USD 16 billion worth of

damages. With fluvial flooding accounting for almost

21 percent of the total number of disastrous events in

the past 10 years, researches on their occurrence and

the vulnerability of specific areas should be given

considerable attention.

The staggering effects of climate change,

particularly the effect of flooding on agricultural

crops, has become a very serious concern worldwide

(Mallari, 2016). A consensus had been established

that inter-annual, monthly, and daily variations in the

distribution of climate variables such as temperature,

radiation, precipitation, water vapour pressure, wind

speed, and rainfall patterns may consequently reduce

agricultural productivity due to a number of physical,

chemical, and biological processes (Cuesta and

Rañola, 2009; Parry et al., 2007). Damage occurs

because flooding causes depletion of soil oxygen

which is crucial for normal metabolism, growth and

development of crops; moreover, it causes

intensification of nitrogen losses and disease

infections which reduce stands and yields (Butzen,

n.d.).

However, assessment of flooding is not only

based on the physical and environmental indicators,

but also on the interaction between flooding and other

agro-ecological, socio-economic factors and human

activities (Dang et al., 2011). Therefore, changes in

the current situation of society like population

density, population ageing, population literacy,

population income source, existing mitigation

measures, road density, access to crop insurance,

access to typhoon forecasting information, access to

planting calendar bulletins, and others contribute to

the vulnerability of a society to the natural hazard

(Dang et al., 2011; Doch et al., 2015; Heng et al.,

2013; Mallari, 2016).

2 STUDY AREA

The municipality of Kalibo, Aklan (Figure 1)

encompasses the mouth of Aklan River. On the north,

it is bounded by the Sibuyan Sea, and on the other

borders by other municipalities of Aklan. Kalibo has

a total land area of 5,075 hectares which is divided

into 16 barangays (PSA, 2016). The first inset map in

Figure 1 shows the location of Panay Island in the

Philippines. The island is consist of the provinces of

Aklan, Antique, Capiz and Iloilo. The location of

Kalibo, Aklan is outlined in red in the second inset

map.

Figure 1: The study area of Kalibo, the capital municipality

of Aklan (Google Earth, 2016). Red lines indicate political

boundaries acquired from the National Statistics Office

(NSO) thru the National Mapping and Resource

Information Authority (NAMRIA).

Kalibo is predominantly an agricultural domain,

as such, agricultural lands cover the biggest portion

of the municipality. Around 1,111 hectares (21.89%)

of soil in Kalibo is suitable for diversified forest

Assessing the Vulnerability of Agricultural Crops to Riverine Floods in Kalibo, Philippines using Composite Index Method

185

crops. An estimated area of 1,555 hectares (30.64%)

was declared highly suitable for tree crops, while

another 1,150 hectares (22.66%) is highly suitable for

and currently planted with rice (Municipality of

Kalibo, 2016). Thus, the major thrust of the local

government of Kalibo is to make it the center of

agriculture-based economic industry and eco-tourism

(Municipality of Kalibo, 2016).

The agricultural areas are highly vulnerable to

flooding according to the records of the Municipal

Agriculture Office of Kalibo (2008; 2012). Based on

the damage reports (MAO, 2012), Typhoon Quinta

destroyed more than 350 hectares of agricultural

crops costing more than PHP 1.2 million in damages.

In July 2008, Typhoon Frank, one of the strongest

typhoons to hit Kalibo, resulted in crop losses

amounting to PHP 23.4 million (MAO, 2008).

3 METHODOLOGY

Agricultural vulnerability assessment to flooding was

performed using composite index method. Having

dimensions reflected by various indicators, an index

is described as a composite measure of any social

phenomena (Mallari, 2016). Calculating composite

indices from indicators is a common way of

quantifying and communicating vulnerability to

hazards, which is visualized using maps (Wiréhn et

al., 2015).

Vulnerability is the propensity or predisposition

to be adversely affected. Vulnerability encompasses a

variety of concepts and elements including sensitivity

or susceptibility to harm and lack of capacity to cope

and adapt (IPCC, 2014).

According to Doch et al. (2015), vulnerability

links the social and biophysical dimensions of

environmental change.

Vulnerability is most often conceptualized as

being constituted by the degree of exposure of the

system to the hazard, sensitivity of the system to

change, and its adaptive capacity to the changes in the

environment (Heng et al., 2013; Bogardi et al., 2005)

A vulnerability assessment of the agricultural sector

in Mabalacat City relative to Typhoon Santi was

performed considering such components and using

GIS to generate vulnerability index maps, identify the

barriers to adaptation, and to provide planning

recommendations (Mallari, 2016). The same

composite index method was used by Heng et al.

(2013) and Doch et al. (2015) to gauge the

vulnerability of agricultural production to flooding in

the Sangkae River watershed in Battambang

province, Cambodia.

Figure 2: The methodological framework for the

agricultural vulnerability assessment.

In this research, the vulnerability assessment of

agricultural crops to flooding in Kalibo was based on

indicators clustered into respective components,

namely exposure, sensitivity, and adaptive capacity.

Literature reviews, internal meetings, discussions

with experts in the academe, local and national

government agencies, focus group discussions

(FGDs) and key informant interviews (KIIs) were

performed for the assessment, review, validation, and

finalization of indicators.

3.1 Data Collection and Preparation

Data collection commenced once the list of indicators

was finalized. Collection of primary socio-economic

datasets was done through farmer household (HH)

surveys. Survey materials consisted of questions

investigating the status and characteristics of the farm

lands and the socio-economic and agro-ecologic

attributes of the farmer households.

Since there was no available official list with the

number of farmer households per barangay, as many

farmer households as possible were interviewed

during a one-week fieldwork in Kalibo. A total of 243

farmer households were interviewed, coming from 14

barangays. No HH survey was conducted in

barangays Poblacion and Andagaw because there

were no agricultural crops in the area, therefore

assuming that there were no farmer households. This

was verified using the detailed LiDAR-derived

agricultural land cover maps and with the barangay

officials. The lack of agricultural crops in the area

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

186

was attributed to the urbanization of these two

barangays.

The flood hazard maps of Kalibo were obtained

from the Disaster Risk and Exposure Assessment for

Mitigation (DREAM) Program of the University of

the Philippines (UP) and the Department of Science

and Technology (DOST). Flood models were

generated using LiDAR data, Synthetic Aperture

Radar (SAR) DEM and other datasets such as river

water level and discharge, soil shapefile, land cover,

meteorological data from Philippine Atmospheric

Geophysical and Astronomical Services

Administration (PAGASA) and DOST-Advanced

Science and Technology Institute (DOST-ASTI), and

the software FLO-2D GDS Pro (DREAM, 2016).

Agricultural features, i.e., land use/land cover

(LULC) including crop distribution, were extracted

from LiDAR datasets and orthomosaics. Derivative

layers (e.g., DTM, DSM, nDSM, CHM, intensity,

number of returns, etc.) that were used for

classification were prepared using LAStools. LULC

was extracted from these derived layers,

orthomosaics, object-based image analysis (OBIA)

and SEparability and THresholds (SEaTH). Accuracy

assessment was also performed to validate the results;

training and validation points were collected from

fieldworks, and visual inspection and interpretation of

orthomosaics and LiDAR data was conducted. Figure

3 shows the generated LULC map of Kalibo, Aklan.

Figure 3: LULC Map of Kalibo, Aklan derived from

LiDAR datasets and orthomosaics.

Since the LULC map does not cover the entire

municipality, exposure computation of the

agricultural classes was based on the processed

LiDAR data area, rather than the entire boundary

area. The implications of performing calculations

based on the processed LiDAR data area would be

minimal since most of the area without LiDAR data

belongs to residential, commercial, and industrial

zones, as verified with the general land use map of the

municipality of Kalibo (Municipality of Kalibo,

2015). Further, areas with the least LiDAR coverage,

i.e., Poblacion and Andagaw, were previously tagged

as non-agricultural barangays.

In addition to the primary datasets, secondary

datasets were requested and acquired from

government-funded research projects, non-

government organizations (NGOs), and the local

planning and agriculture office.

3.2 Normalization

To make all the indicator values comparable and

congruent, they were standardized to fit within the

range zero (0) to one (1) using either linear

normalization or Z-score, depending on the type of

data. Linear normalization was done using the

formula:

= (X

- X

min

)/(X

max

- X

min

)

(1)

Where:

= normalized value of indicator i to barangay j

= original value of indicator i to barangay j

max

= highest value

min

= lowest value

For ordinal data, i.e., hierarchical arrangement but

without meaningful interval, standardization using Z-

score was applied using the online calculator

Measuring U (www.measuringu.com/pcalcz.php)

3.3 AHP

To compute for the composite index, which is defined

as the weighted average of all the normalized

indicator values, weights were determined and

assigned using Analytic Hierarchy Process (AHP.

AHP, which is also called Saaty Method, is a

complex decision making tool introduced by Thomas

Saaty (1980). It is a theory of measurement primarily

performed through pairwise comparisons and relies

on the judgements of experts to derive priority scales

(Saaty, 2008; Mendoza et al., 2014).

AHP was used to determine the weights of

indicators in every component of the composite

index. The same weighting method was utilized in

assigning the exposure weights of the land cover

classes.

Assessing the Vulnerability of Agricultural Crops to Riverine Floods in Kalibo, Philippines using Composite Index Method

187

The selection of experts is crucial as the

credibility of the results substantially depends on it.

The criteria for expert selection were familiarity with

the agricultural characteristics of the barangays

and/or municipality and sufficiency of knowledge on

flooding and its impact on the agricultural sector.

Thirteen to fifteen experts were consulted per

indicator type to rate the indicators for exposure,

sensitivity, adaptive capacity, and the LULC classes.

According to Saaty (1980, 2008), for a set of

ratings to be considered acceptable, its computed

consistency ratio (CR) should be less than or equal to

0.1, otherwise, ratings should either be repeated or

disregarded. However, Kluhto (2013) and Alonso and

Lamata (2006) argued that the tolerance value can be

raised to 20 percent which corresponds to an

acceptable CR of less than 0.2. This research used a

CR threshold of 0.2.

3.4 Vulnerability Indicators

The indicators were finalized based on the judgement

of experts from different fields and on the availability

of data. Experts included barangay agriculture

technicians, barangay and municipal agricultural

officers, representatives from farmer organizations,

researchers conducting VA studies, and

representatives from agricultural schools, among

others. A total of 30 indicators were determined: 5 for

exposure, 13 for sensitivity, and 12 for adaptive

capacity.

3.4.1 Exposure

Exposure, an external factor, is the nature or degree

to which a system is exposed to significant climatic

variations taking into account the frequency,

duration, and/or extent in which the system is in

contact with a hazard (Locatelli et al., 2008; Heng et

al., 2013; Doch et al., 2015).

Inputs in the flood hazard and flow depth models

(see Figure 4) provided by DREAM in September

2016 include data from three (3) PAGASA RIDF

(rain intensity, duration, and frequency) stations

located in Romblon, Roxas City and Iloilo City.

Moreover, precipitation data was derived from 14

DOST-ASTI ARG (automated rain gauge) stations

distributed throughout the province of Aklan – two of

which are located within Kalibo. The produced vector

layers have 10-meter resolution showing floods for

storm events with a five-year return period (i.e., 20

percent chance of occurrence in any given year).

Figure 4: Flood hazard map (left) and flow depth map

(right) from DREAM Phil-LiDAR 1.

The flood hazard data was divided into three

levels based on the flood height and the product of

velocity and height (VH); these levels were as

follows:

 Low (flood height of 0.1m to 0.5m and VH of

0.1m

/s to 0.5m

/s)

 Medium (flood height of 0.5m to 1.5m or VH

of 0.5m

/s to 1.5m

/s)

 High (flood height of more than 1.5m or VH

greater than 1.5m

/s)

Two indicators were derived from the flood

hazard data; one accounted for the area covered by

the hazard while the other provided the level of the

hazard.

On the other hand, flow depth data was divided

into five (5) classes with the following ranges: less

than 0.500m, 0.501m to 1.000m, 1.001m to 2.000m,

2.001m to 5.000m and more than 5.000m.

LULC was also included in the exposure

component. Classes were divided into non-

agricultural features, grain crops, tree crops, and oil

crops. Each class was assigned an exposure weight

based on the ratings of experts. Exposure weight per

LULC in the area should be identified as crop reaction

and/or response to flooding varies.

Out of the 15 experts consulted to rate the degree

of exposure of LULC classes to fluvial flooding, 13

expert ratings were able to meet the tolerance value

of 20 percent. The computed weights using AHP are

shown in Table 1.

Table 1: Weights of LULC classes.

LULC Weight

Non-agricultural features 0.0000

Grain crops 0.5695

Oil crops 0.1617

Tree crops 0.2688

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

188

Resulting weights indicated grain crops as the

most exposed crop at almost two times more exposed

than tree crops and almost four times more than oil

crops. Some of the factors that affected the ratings

were the degree of resilience of the crops to flooding,

the sturdiness of the crops, and the degree of potential

damage to the crops based on cropping period.

The finalized exposure indicators were based on

the ratings of the 13 experts, 11 of which were

considered consistent. The list of exposure indicators

and their corresponding weights are shown in Table

Table 2: Computed weights of exposure indicators.

Indicator Weight

Flood hazard (area) 0.2429

Flood hazard (level) 0.2869

Flow depth 0.1582

LULC 0.3120

3.4.2 Sensitivity

Sensitivity, an internal factor, is the degree to which

a system is affected, either adversely or beneficially,

by a hazard (Locatelli et al., 2008). It is also defined

as the extent to which a system can absorb impacts

without having a significant change in state (Heng et

al., 2013).

For the study, 13 sensitivity indicators as listed in

Table 3 were determined and rated by the experts;

most of these indicators were collected through KIIs

with farmer households. Five experts (out of the 14

local experts) met the required 0.2 CR value.

Table 3: Sensitivity indicators identified by the experts and

corresponding weights determined using AHP.

Indicator Description Weight

Population

density

Ratio of the number of

people per 1km

0.1040

Dependence

ratio

Ratio of the number of

unemployed HH

members to total number

of HH members

0.0646

Human

sensitivity

Ratio of number of HH

members that are elderly

(65y/o<), children

(>5y/o), PWD (person

with disability),

pregnant, with chronic

illness to total number of

HH members

0.0786

Hunger

incidence

Households were asked

if they experienced

hunger in the past three

months

0.0562

Poverty rate

Number of households

meeting the poverty

threshold of the province

over the total number of

households

0.0490

Level of

education

HH member with highest

educational attainment

0.0381

Tenurial

status

Ownership of the

agricultural land (i.e.

owned, leased, tenant,

etc.)

0.1008

Membership

Membership in a

farming organization

0.0691

Percent agri

income

Percent of income from

agriculture over total

income

0.1128

Percent debt

Percent of debt over total

income

0.0983

Access to

PHF

Access to post-harvest

facilities

0.0954

Access to

roads and

bridges

Access to various

transportation media

0.0761

Road density

Ratio of road area to

land area

0.0570

Legend

Gathered from HH surveys

From CBMS (Community-Based Mapping System)

Derived from LULC

3.4.3 Adaptive Capacity

Adaptive capacity, an external factor like exposure, is

the ability of a system to adjust to a hazard (Locatelli

et al., 2008) or evolve (Doch et al., 2015) in order to

accommodate environmental hazards and neutralize

potential damages, or to take advantage of

opportunities of planning to expand its range of

variability for coping. The data for the 12 adaptive

capacity indicators (Table 4) were gathered from HH

surveys and were consistently rated by 13 experts.

Table 4: Adaptive Capacity indicators identified by the

experts and corresponding weights determined using AHP.

Indicator Description Weight

Percent non-

agri income

Percent of income from

non-agriculture activities

over total income

0.1004

Percent

savings

Percent assets/savings

over total income

0.0932

Access to loan

HH access to loans or

credits

0.0969

Access to

financial aid

HH access to financial

aid from government

and/or non-government

organizations

0.0878

Access to

rehabilitation

HH access to

rehabilitation and aid

0.1210

Assessing the Vulnerability of Agricultural Crops to Riverine Floods in Kalibo, Philippines using Composite Index Method

189

Table 4: Adaptive Capacity indicators identified by the

experts and corresponding weights determined using AHP

(cont.).

Indicator Description Weight

Access to

training

courses

HH access to training

courses focused

particularly on climate

variability and farming

0.0685

Farming

experience

Length of farming

experience of the HH

0.0646

Crop selection

Whether crop selection is

determined by

season/potential hazards

0.0746

Level of

mechanization

Level of agricultural

mechanization (e.g. use

of machineries)

0.0846

Insurance

Insurance of agricultural

land and/or crops

0.1045

Disaster and

hazard policy

Disaster preparedness 0.0557

Means of

transportation

Means of transportation

of the HH (motorized

and non-motorized)

0.0482

3.5 Vulnerability Index

The overall vulnerability index of the agricultural

crops was computed using the formula:

Vulnerability V = [(Exposure E +

Sensitivity S) - (Adaptive Capacity A)]

(2)

It was acknowledged by Doch et al. (2015) that

vulnerability is scale dependent since the scale of

analysis affects the result and the pattern being

identified. Because the datasets have different scales,

the level of analysis was set to barangay level and

with a 500m x 500m grid. Political boundaries

acquired from NSO were intersected with a generated

500m x 500m fishnet to produce clipped tiles based

on the agricultural barangay boundaries. Calculations

were made per tile. Moreover, area weighted average

per barangay was also computed and a comparison of

the results was then performed.

3.6 Validation of Vulnerability Maps

To validate the results of the vulnerability index,

historical accounts and damage reports were

requested from the local planning and agriculture

office of the municipality. Farmer HHs were

interviewed regarding the frequency and magnitude

of flooding in the area and the damage caused to the

agricultural sector. First-hand experiences were also

recorded to aid in the analysis and validation of the

vulnerability maps.

4 RESULTS AND DISCUSSION

A model was created in ArcMap to automate the

processing of datasets from data pre-processing to the

production of exposure, sensitivity, adaptive capacity

and vulnerability shapefiles. Scripts were also

provided for the normalization of exposure data.

Using the constructed model, values of exposure,

sensitivity, adaptive capacity per agricultural

barangay were calculated as illustrated in Figure 5.

The stacked bars depict the vulnerability index

equation. The first set of bars represents the potential

impact, i.e., the combined effects of exposure and

sensitivity, which may be caused by the hazard. The

vulnerability ratings of the barangays were obtained

by subtracting the positive implications of adaptive

capacity from the potential impact.

As observed in the results, Tinigaw was tagged to

be the most vulnerable barangay with a rating of

0.7286 followed by Mobo with 0.7081. Linabuan

Norte and Tigayon were not far behind with ratings

of 0.6878 and 0.6393, respectively. Another barangay

above the 0.6 vulnerability rating was Caano with

0.6149.

When the flood hazard, flow depth, and LULC

maps were inspected, four out of the five most

vulnerable barangays were found to be located along

the river. Since overflow of water from the river is

one of the primary drivers of flood hazard in the

subject area, barangays adjacent to the river were

expected to be more vulnerable, as supported by the

produced maps. Furthermore, based on actual

interviews with the farmers, the effects of riverine

flooding in their barangays were certainly prominent

especially when the water level in the river became

higher than usual, further amplified by heavy rains

brought by typhoons.

Caano is the only barangay not located along the

river. It is mostly covered with agricultural crops,

particularly grain crops which were scored as the

most vulnerable to flooding. This pulled the exposure

rating of the said barangay since LULC has the

highest weight for the exposure component.

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

190

When the exposure ratings alone were inspected,

a great difference between the values of the top two

most vulnerable barangays was observed. Mobo, in

fact, was more exposed by more than 0.06. However,

the populace of Tinigaw was considered more

sensitive.

The next barangays in the ranking, Bachaw Sur,

Pook, and Briones, showed the same trend having

relatively similar values of exposure, sensitivity, and

adaptive capacity. Also, these barangays have the

same geographical features, i.e., are near water bodies

and are covered by agricultural crops.

Mabilo and Buswang New, the succeeding

barangays in the ranking, are coastal barangays.

Based on the previously discerned trends, it could be

assumed that these barangays should have the same

vulnerability rates with the other barangays near

water bodies, in close proximity to either the river or

the sea. However, the results diverged from these

deductions, with the reason perceived to be the land

cover of the area. Since agricultural crops are not very

prominent in the area, the land cover was not subject

to much exposure.

In the case of Estancia, its computed hazard

(exposure and sensitivity) was almost 0.8 – near the

hazard value of Pook which ranked 7

. However, its

vulnerability score was not as high due to the counter

effect of its adaptive capacity.

Nalook, on the other hand, is situated far from

bodies of water. Though covered mostly by

agricultural crops, Nalook was not perceived to be

vulnerable because of its relatively greater distance

away from the river and the sea.

The least vulnerable barangays, Buswang Old and

Buswang New, have the exact opposite situation from

that of Nalook. Both barangays are near the river and

are coastal barangays. Additionally, very few

agricultural crops were planted in the area. This was

also verified during the HH survey as less than ten

farmer households were known to the barangay

offices.

These results were validated with the agricultural

crops damage report prepared by the Kalibo

agriculture office (2008). Based on the report during

Typhoon Frank, Mobo was the most devastated

barangay with damages amounting to almost PHP 7

million. In addition to that, more than PHP 3 million

worth of agricultural products and infrastructure were

destroyed in Linabuan Norte, and Tigayon. The

farmers’ first-hand experiences corroborated the

results of the map with almost 75 percent of the

respondents confirming the destruction brought by

Frank.

The computed area weighted average of cells per

barangay was compared with its corresponding

vulnerability value computed using the NSO

boundary. Identical rankings were observed in both

datasets, although the results using grids were

consistently lower as illustrated by the bar graph in

Figure 6. Further, discrepancies in the data increase

as the vulnerability values for both scales increase.

Figure 5: Computed values of every agricultural barangay. Blue bars indicate the exposure (E) rating of the barangays, yellow

bars represent sensitivity (S), green bars portray the offset caused by adaptive capacity (A), and red bars show the computed

vulnerability (V) ratings of the barangays.

Assessing the Vulnerability of Agricultural Crops to Riverine Floods in Kalibo, Philippines using Composite Index Method

191

Figures 6: Vulnerability index values computed using two

different scales.

The observed trends may be attributed to the

generalization of the barangay data. Since an

assumption was made that the primary HH data

gathered applied to the entire barangay, generalizing

the inputs amplified the effect of the indicators.

Setting a smaller grid size may further decrease the

vulnerability values. Thus, it is recommended for

future studies to conduct geotagging during the

farmer HH interviews for a more accurate dataset

coverage. Moreover, participatory mapping will be

beneficial especially in analysing the results. Further

researches on the effect of scale in the computation of

vulnerability indices is recommended.

Vulnerability maps representing the results of the

composite index are shown in Figures 7 and 8.

Figure 7: Barangay-level vulnerability of agricultural crops

to floods in Kalibo.

Figure 8: Vulnerability of agricultural crops to floods in

Kalibo using 500m x 500m grid.

Figure 9: Plot of exposure vs. sensitivity values (top) and

exposure vs. adaptive capacity values (bottom).

Shown in Figure 9 are scatter plots of the

computed component values per barangay. The plots

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

192

show the distribution of exposure values versus

sensitivity and adaptive capacity values. In the first

plot, it can be observed that the most vulnerable

barangays were also the most exposed. These kinds

of values often lead to the common misconception of

exposure equating vulnerability and sensitivity.

However, when the data was analysed such as in the

case of Bachaw Norte, the data showed that although

it was not the least exposed, it was the least

vulnerable. This was due to its relatively lower

sensitivity. A number of barangays were tagged as

having moderately high vulnerability because of the

linear relationship between exposure and sensitivity

= 0.2041). To counteract the effect of exposure,

sensitivity should be low, i.e. a plot of the two should

form a steep inverse slope.

In the second plot (Figure 9), Estancia, Nalook,

and Mabilo have lower vulnerability ratings despite

their high exposure values because of their adaptive

capacity values. Lower vulnerability ratings are

attained when the plot of exposure versus adaptive

capacity forms a steep slope.

In general, the relationships of exposure to

sensitivity and adaptive capacity as shown in Figure

9 provide indication that Kalibo and its barangays

over the years have somehow adapted to some degree

in dealing with flood hazards. Barangays with greater

exposure have higher sensitivity and some barangays,

which have higher exposure, have relatively lower

sensitivity. It can also be said that barangays with

relatively higher exposure have developed ways such

that they have higher adaptive capacity.

5 CONCLUSIONS

Vulnerability of the agricultural sector to fluvial

flooding is influenced by a number of factors.

External factors alone do not dictate the state of the

municipality. The socio-economic and agro-ecologic

status of the affected community greatly contributes

to the overall condition of the subject area. Thus,

internal factors i.e. factors that are within the control

of the community must be identified, assessed, and

addressed. Moreover, identifying the factors affecting

the vulnerability of Kalibo and assessing them are

crucial in improving the condition of the municipality

through planning, effective decision making and

imposing policies that can ameliorate the coping

capabilities and adaptive capacity of the community.

As illustrated by the case of Estancia,

vulnerability was alleviated because of the adaptive

measures of either the households or the barangay.

For a more efficient response to force majeure, a

collective effort from both the farmers and the local

government unit is required. Likewise, vulnerability

maps, plans, and policies are useless when they are

not utilized and followed. Therefore, action from and

coordination of all involved individuals and groups is

essential. Although vulnerability assessment is not a

panacea, this study can serve as a practical guide in

alleviating the damage caused by riverine flooding in

the agricultural sector of Kalibo, Aklan.

The results will be provided to local government

units and stakeholders to aid in the decision making

process for generating policies and/or programs that

may improve the adaptability of the agricultural

system during the occurrence of flooding.

ACKNOWLEDGEMENTS

This research was made possible through the funding

of Department of Science and Technology (DOST)

and monitoring by the Philippine Council for

Industry, Energy and Emerging Technology Research

and Development (PCIEERD). The algorithms and

workflows were developed using the spatial and

satellite image data from DREAM/Phil-LiDAR 1

Program. Technical facilities and support were

provided by the UP College of Engineering,

Department of Geodetic Engineering and Training

Center for Applied Geodesy and Photogrammetry.

The project was also supported by the Department

of Agriculture through the Information and

Communications Technology Service (ICTS).

The project acknowledges everyone who has

contributed to this research especially the consulted

national and local experts, local government offices,

and farmer household survey respondents.

REFERENCES

Alonso, J. and Lamata, T. (2006). Consistency in the

Analytic Hierarchy Process: A New Approach.

International Journal of Uncertainty, Fuzziness and

Knowledge-Based Systems, 16(4), 445-459.

Bogardi, J.J., Villagran De Leon, J.C., Birkmann, J.,

Renaud, F., Sakulski, D., Chen, X., Affeltranger, B.,

Mensa, A., Kaplan, M. (2005). Vulnerability in the

context of climate change [Workshop]. Retrieved from

https://www.academia.edu/15601966/Vulnerability_in

_the_context_of_climate_change.

Bündnis Entwicklung Hilft & United Nations University –

Institute for Environment and Human Security (2016).

World Risk Index. Retrieved from

http://weltrisikobericht.de/english.

Assessing the Vulnerability of Agricultural Crops to Riverine Floods in Kalibo, Philippines using Composite Index Method

193

Butzen, S. (n.d.). Flooding Impact on Crops. Retrieved

from:

https://www.pioneer.com/home/site/us/agronomy/crop

-management/adverse-weather-disease/flood-impact/

Centre for Research on the Epidemiology of Disasters -

CRED (2016). In The International Disaster Database.

Retrieved November 15, 2016, from

http://www.emdat.be/country_profile/index.html.

Cuesta, M. and Rañola, R., Jr. (2009). Adaptive Capacity of

Rice Farmers to Rainfall Variability and Extremes in

the Province of Camarines Sur, Philippines.

Philippagric Scientist, 92(4). Retrieved from:

https://journals.uplb.edu.ph/index.php/PAS/article/vie

w/350.

Dang Dinh Kha, Tran Ngoc Anh and Nguyen Thanh Son

(2011). Flood vulnerability assessment of downstream

area in Thach Han river basin, Quang Tri province.

Retrieved from:

hmo.hus.vnu.edu.vn/file.php?file_id=459.

Disaster Risk Exposure and Assessment for Mitigation

[DREAM] (2016). Aklan: DREAM Flood Forecasting

and Flood Hazard Mapping. Quezon City: UP Training

Center for Applied Geodesy and Photogrammetry

(TGACP).

Disaster Risk Exposure and Assessment for Mitigation

[DREAM] (2016). Aklan: DREAM LiDAR Acquisition

and Processing Report. Quezon City: UP Training

Center for Applied Geodesy and Photogrammetry

(TGACP).

Doch, S., Diepart, J.-C. and Heng, C. (2015). A multi-scale

flood vulnerability assessment of agricultural

production in the context of environmental change: The

case of Sangkae River watershed, Battambang

province. In Diepart, J.-C., (ed.) Learning for

resilience: Insights from Cambodia’s rural

communities, Phnom Penh: The Learning Institute, pp.

19-49.

Google Earth (2016). Retrieved November 15, 2016.

Heng, C., Doch, S. and Diepart, J.-C. (2013). Towards

Measuring the Vulnerability of Agricultural Production

to Flood: Insight from Sangkae River Cathcment,

Battambang Province, Cambodia. (IJERD)

International Journal of Environmental and Rural

Development, 4(2). Retrieved from:

http://iserd.net/ijerd42/42015.pdf.

International Strategy for Disasrter Reduction (2004).

Living with risk: A global review of disaster reduction

initiative. Geneva: UN Publications. Retrieved from:

http://www.unisdr.org/files/657_lwr1.pdf.

IPCC (2014). Summary for policymakers. In: Climate

Change 2014: Impacts, Adaptation, and Vulnerability.

Part A: Global and Sectoral Aspects. Contribution of

Working Group II to the Fifth Assessment Report of the

Intergovernmental Panel on Climate Change [Field,

C.B., V.R. Barros, D.J. Dokken, K.J. Mach, M.D.

Mastrandrea, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O.

Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N.

Levy, S. MacCracken, P.R. Mastrandrea, and L.L.

White (eds.)]. Cambridge University Press, Cambridge,

United Kingdom and New York, NY, USA, pp. 1-32.

Kluhto, S. (2013). Mathematical Decision Making An

Overview of the Analytic Hierarchy Process. Retrieved

from:

https://www.whitman.edu/Documents/Academics/Mat

hematics/Kluhto.pdf.

Locatelli, B., Herawati, H., Brockhaus, M., Idinoba, M. and

Kanninen, M. (2008). Methods and Tools for Assessing

the Vulnerability of Forests and People to Climate

Change. Center for International Forestry Research.

Mallari, A. E. (2016). Climate Change Vulnerability

Assessment in the Agriculture Sector: Typhoon Santi

Experience. Procedia – Social and Behavioral

Sciences, 216, 440-451. Retrieved from:

http://www.sciendirect.com/science/article/pii/S18770

42815062382.

Mendoza, M.E., Bui Dung The, Heng Naret, Ballaran, V.,

Jr. and Arias, J.K. (2014). Assessing Vulnerability to

Climate Change Impacts in Cambodia, the Philippines

and Vietnam: An Analysis at the Commune and

Household Level. Journal of Environmental Science

and Management, 17(2): 78-91.

Municipality of Kalibo (2016). In Municipal Profile.

Retrieved from: http://www.kaliboaklan.gov.ph/wp-

content/uploads/2016/05/MUNICIPAL-PROFILE.pdf.

Municipal Agriculture Office (MAO) of Kalibo (2008).

Damage Reports of Typhoon Frank.

Municipal Agriculture Office (MAO) of Kalibo (2012).

Damage Reports of Typhoon Quinta.

Municipal Agriculture Office (MAO) of Kalibo (2015).

Kalibo Ecological Profile.

Parry, M, Canziani, P., Palutikof, J., van der Linden, P. and

Hanson, C. (2007). Climate Change 2007: Impacts

Adaptation and Vulnerability. Contribution of Working

Group II to the Fourth Assessment Report of the

Intergovernmental Panel on Climate Change.

Cambridge: Cambridge University Press. Retrieved

from: https://www.ipcc.ch/pdf/assessment-

report/ar4/wg2/ar4_wg4_full_report.pdf.

Philippine Statistics Authority (2016). In Philippine

Standard Geographic Code. Retrieved November 18,

2016 from

http://www.nap.psa.gov.ph/activestats/psgs/province.a

sp?provCode=060400000.

Saaty, T. (1980). The Analytic Hierarchy Process. New

York: McGraw-Hill.

Saaty, T. (2008). Decision making with the analytic

hierarchy process. Int. J. Services Sviences, 83-98.

Senate Economic Planning Office (2013). Natural

Disasters At A Glance. Retrieved from

https://www.senate.gov.ph/publications/AAG%20201

3-04%20-%20Natural%20Disasters_final.pdf.

Wiréhn, L., Danielsson, Â. and Neset, T.-S. (2015).

Assessment of composite index methods for

agricultural vulnerability to climate change. Journal of

Environmental Management, 156, 70-80.

doi:10.1016/j.jenvman.2015.03.020.

GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management

194