Indonesian Journal of Geography.
Vol 54.
No.
ISSN 2354-9114 .
ISSN 0024-9521 .
Indonesian Journal of Geography Vol 57.
No.
: 440-447
DOI: 10.
22146/ijg.
99419 website: htps://jurnal.
id/ijg A2025 Faculty of Geography UGM and The Indonesian Geographers Association ARTICLEARTICLE
REVIEW
RESEARCH
Flood Risk Food Mapping Using in GISIndonesia:
and Multi-Criteria Analysis at Nanga West Mapping Security Geographic Clusters andPinoh Regional Kalimantan Disparities Area *Ajun Purwanto1.
Rustam2.
Eviliyanto3.
Dony Andrasmoro4 Rusli Abdulah 1,3,4 .
Departmen of Geography Education IKIP PGRI Pontianak Graduate School of International Development (GSID).
Nagoya University.
Japan, .
Institute for Development of Departmen of Counseling Guidance Education IKIP PGRI Pontianak Economics and Finance (INDEF).
Jakarta-Indonesia Received: 2021-12-22 Received: 2024-08-26 Accepted: 2022-10-13 Revised: 2025-11-10 Accepted: 2025-11-10 Keywords:
Published: 2025-11-11 Flood Risk.
GIS.
Multi-Criteria Analysis.
Nanga Pinoh Abstract.
one of the hit various regions in Indonesia, in WestDevelopment Kalimantan.
Abstract Flood Food issecurity become that a global and represents the first ofspecifically the Sustainable The in Nanga District.
Regency, 18 villages Therefore.
Goals, is zeroPinoh
ManyMelawi Indonesia, have et
the central
aimed to map
flood There risk areas Nanga
Pinoh
and their on their has in into food
security issues, butSecondary primarily this multi-criteria
GIS
no spatial context.
This research identifies spatial clustersAihigh-high, low-low, high-low, and low-highAi that the theFood had low, and high was found that areas four The Security Vulnerability Atlas, the ItDietary Diversity Score,with lowCalorie risk class are 1,515.
ha, 514 30,194.
ha,in21,953.
ha, respectively.
Food prone.
Varietymedium.
Score, and Intake.
It explores Indonesiaha.
Exploratory These implied(ESDA, that the GIS approach are effective tools(LISA).
for flood Spatialfindings Data Analysis Global MoranAos I)and andmulti-criteria Local Indicators of Spatial Association The in anticipating and mitigating the disasters.
security come
SUSENAS
he National Socio-Economic Surve.
Statistics of Indonesia and the National Food Agency.
The research reveals the presence of regional food security in Indonesia.
Eastern Indonesia faces challenges from food insecurity issues.
The LISA result shows that there are low-low clusters in eastern Indonesia because of geographical isolation, poor economic performance, and a lack of infrastructure.
Keywords: food security.
Conversely, high clusters in western Indonesia, especially in Java Island, benefit from favorable agricultural ESDA.
MoranAos I.
LISA, conditions, a robust infrastructure, and diverse food markets.
High-low clusters highlight that there are urban Indonesia centers with better food access amidst less secure areas, low-high areas and logistical in the West Kalimantan.
This Introductin being near food-secureoccurred This beyond including the results Nanga Pinoh Police Floods occur when a riverchallenges exceeds despite its storage of a standard ESDA, emphasizing the need for tailored policies to address regional Future Tanjung Tanjung Lay Village.
Tembawang Panjang.
Pal Village, forcing the excess water to overflow the banks and fill the should explore the determinants of food security using spatial and non-spatial econometric approaches and Niaga.
Kenual.
Baru and Sidomulyo Village in Nanga Pinoh adjacent low-lying lands.
Thisshould represents the apply convergence analysis to identify the factors driving prosperous regions, providing benchmarks for Spectacle.
Melawi Regency (Supriyadi, 2.
most frequent disasters affecting enhancing food security across all districts.
*Correspondeny email:
ajunpurwanto@ikippgriptk.
A2022 by the authors.
Licensee Indonesian Journal of Geography.
Indonesia.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution(CC BY NC) licensehttps://creativecommons.
org/licenses/by-nc/4.
The flood disaster in Melawi Regency should be mitigated worldwide (Rincyn et: al.
, 2018.
Zwenzner & Voigt, 2.
Correspondent future consequences by mapping the risk.
Indonesia.
Flooding A2025 Indonesian Journal Geography s6@s.
This article is an open accessfeatures article distributed under the terms and conditions of Remote the Creative Commons Various such as Sensing and Geographic man-made nagoya-u.
Attribution(CC BY NC) licensehttps://creativecommons.
org/licenses/by-nc/4.
Information Systems have been developed for monitoring flood (Du et al.
, 2013.
Falguni & Singh.
Tehrany et al.
, 2013.
This technology has significantly contributed to flood Youssef et al.
, 2.
monitoring and damage assessment helpful for the disaster There are flood risks in many regions resulting in great (Biswajeet & Mardiana.
Haq The used for the FSVA employ the framework Introduction (Alfieri et al.
, 2016.
Mahmoud & Gan, 2.
with et al.
,and Pradhansecurity et al.
, 2.
as the Furthermore, basis for analysis.
Food Food security become and a global issue that represents social, has have been to map flood vulnerability and extent three pillarsAiavailability, access and utilizationAi first of the Sustainable Development Goals, (Falguni & Singh.
Geographic.
Komolafe the damage.
These techniques and assess nutrition guide and the Manyetcountries set foodet as the Rincyn , 2018.
Skilodimou , 2.
Theprimary of Remote Sensing (RS) andofGeographic (Ministry AgricultureInformation Indonesia.
Systems World of their In 2012 on loss of human life,agendas.
adverse impacts (GIS) to improve the efficiency of monitoring and managing Food Programme.
Indonesia Act No 18 Yearessential of 2012 about Food, damage toissued the infrastructure, floodThe et al.
, 2.
guide forthe policy inand Indonesia.
The purpose of spread food of iseases, water contamination FSVA (Haq is a thematic map that depicts a geographical In the age of technology, to Act et is stated in Article 3: AuFood implementation is carried (Rincyn , 2.
of modern areas susceptible food insecurity.
extracted through Geographical Information System (GIS) and out Food to meet basic human in a fair, for 34%needs and 40% global benefits natural disasters the food security index values into six categories:
Remote Sensing (RS) into other datasets manner based Sovereignty, in quantityand losses, respectively (LyuonetFood , 2019.
Petitvery mild weak, for very andNasional.
Food SecurityAy.
Boix etIndependence, , 2.
, with theFood increasing significantly resistant monitoring, (Badan Pangan (Biswajeet Mardiana.
Haq (Komolafe As regards the implementation of the Act, the National In the FSVA, as mentioned above, the concept of food Pradhan rests et al.
,on2.
the causes access, of flooding RozalisAgency et al.
, (NFA)
isThe Food that include has the threeUnderstanding main pillars: availability, (Ozkan Tarhan.
Zhou task of carrying out government duties in the food sector in Food availability pertains to the physical presence Different hazard area, land structure (Jha et , 2011.
Zwenzner & Voigt,through 2.
, and to maintain in Indonesia, food inflood a specific areasAo (Curebal coordination, formulation, and determination of policies production, government reserves, and external sources like flooding and These for the Other are land-use change, suchofasfood the stabilization supplies and Food access processes refers to a are householdAos (Huong Pathirana.
Rincyn prices, food insecurity and nutrition, the diversification of obtain sufficient nutritious food.
Food security can be achieved floods, and flood management Zhang et al.
, 2018.
Zhou et al.
(Presidential Regulation self-production, (Bubeck Falguni Singh.
Mandal The (Perpre.
Number 66 of 2021 Concerning the National Food bartering, or receiving aid.
It is important to note Chakrabarty.
Shafapour Tehrany sub-districts West Kalimantan Agency, 2.
if food is available in an area, specific households might face GIS and remote or map the Thousands of uses 18 villages Melawi Regency Atlas social technologies barriers preventing The NFA theinFood Securityinand Vulnerability acquiring adequate quantities or an adequate varietyhazards of food (FSVA) to monitor food security in each district in Indonesia.
Indonesian Journal of Geography.
Vol 57.
No.
(Ministry of Agriculture Indonesia.
World Food Programme.
Lastly, food utilization represents how households use the food they can access and how individuals absorb nutrients.
This pillar encompasses various factors such as food storage and preparation methods, water safety, hygiene practices, feeding habits .
specially for those with special dietary need.
, fair food distribution within households, and the overall health status of family members.
Interestingly, because of the significant role played by women in enhancing their familyAos nutritional intake, particularly for young children, the motherAos education level is often used to indicate a householdAos food utilization (Ministry of Agriculture Indonesia.
World Food Programme, 2.
Besides the FSVA measurement, there are other measurements that could be used to monitor food security:
calorie intake, the food variety score (FVS), and the dietary diversity score (DDS).
The calorie intake measurement refers to the number of calories in the daily diet.
At the global level, the average calorie needs vary by country according to its age and gender distribution, but the consensus is that a globally average diet should provide 2,100 kcal per capita per day (FAO, et al.
, 2.
The FVS is measured as the number of items eaten by any household member on the previous day from a list of 104 food items (Hatlyy, et al.
, 2.
The more diverse the food consumed by a household or individual, the more secure they are.
In addition, the DDS is used, which measures how many of several food groups were consumed by a household or individual on the previous day (Hatlyy et al.
, 2.
The essential difference between the FVS and the DDS measurements is the categorization of the food basket.
The FVS uses items, while the DDS uses food groups.
Food security is still unequal in Indonesia.
In general, the western region of Indonesia, especially the island of Java, is more food-secure than the eastern region.
Figure 1 shows a darker color in the western region of Indonesia, which means that this region is more food-secure.
For example, using the FSVA figures, all districts/cities on the island of Papua, except the city of Merauke, have an FSVA index of less than This illustrates that most regencies/cities on the island of Papua have a very vulnerable .
or vulnerable .
econd priorit.
FSVA status (Badan Pangan Nasional, 2.
Some of the existing research on food security in Indonesia uses cross-sectional data or data panels (Aguilera & Jatmiko, 2023.
Amrullah, et al.
, 2023.
Herlina et al.
, 2020.
Kadir, et al.
, 2023.
Nugroho et al.
, 2022.
Vidyarini, et al.
, 2021.
Widada, et al.
, 2017.
Yuliana, 2.
However, it does not describe the spatial aspect.
The inclusion of spatial elements in research with cross-sectional data or data panels is beneficial for researchers and policymakers because it provides detailed nuances for each region.
Furthermore, there are several reasons why it is essential to understand the location patterns for the different food security First, identifying spatial clusters for food insecurity allows for a more efficient allocation of resources and targeted policy interventions.
Second, different measurements capture different aspects of food security, and analyzing their spatial patterns provides a more comprehensive view of the Third, a comparison of spatial patterns over time can help to evaluate the effectiveness of food security interventions and policies.
Fourth, understanding the spatial patterns of food security can provide information for related sectors such as agriculture, health, and economic development.
This study explores 514 districts in Indonesia for 2019 using ESDA.
The data used come from Statistics of Indonesia and the NFA.
Methods The study employs ESDA, specifically Global MoranAos I, and Local Indicators of Spatial Association (LISA).
ESDA
uses a number of visual and numerical methods to analyze spatial data, and these methods can be categorized into two primary types.
The first type use classical non-spatial descriptive statistics that are dynamically connected to GIS maps and spatial objects.
The second type focus on identifying spatial interactions, relationships, and patterns by utilizing a spatial weights matrix, hypothesis testing, and various metrics (Grekousis, 2.
Furthermore.
ESDA describes and summarizes spatial data distributions, identifies spatial outliers and clusters, and examines spatial autocorrelations to explore spatial relationships through Global MoranAos I and LISA analysis.
Calorie Intake
FSVA
Dietary Diversity Score (DDS) Food Variety Score Source: Statistics of Indonesia and National Food Agency, 2019, processed Figure 1.
Food Security Map 2019
MAPPING FOOD SECURITY IN INDONESIA
Rusli Abdulah.
Specifically.
Global MoranAos I measures spatial autocorrelation, determining whether the pattern of a variable is clustered, dispersed, or random across a geographic area.
Furthermore.
LISA identifies local clusters and spatial outliers, offering a more detailed insight into spatial patterns at a regional level (Grekousis, 2.
The ESDA method has two main steps, namely global investigation and local exploration.
Calculating MoranAos I represents the global investigation.
This investigation measures the similarities and dissimilarities between observations across space.
The formula for Global MoranAos I (Moran, 1.
is written below:
LISA outputs are often shown using cluster and significance A cluster map reveals four categories: high-high clusters .
igh values surrounded by high value.
, low-low clusters .
ow values surrounded by low value.
, and spatial outliers .
ighlow and low-high cluster.
High-low clusters have high values surrounded by low values, and low-high clusters have low values surrounded by high values.
This helps to identify local patterns and spatial relationships (Anselin, 1.
The research included 514 districts in Indonesia in 2019 and applied the weight spatial matrix to calculate the Global MoranAos I and apply LISA.
In research that uses spatial analysis, the researcher should first identify the neighbors in which the contiguity matrix is commonly used.
This contiguity is based on districts across Indonesia, and the method works in continental areas but not in regions where there are islands separated by sea.
The challenge is to deal with the islands that make up Indonesia.
Previous literature has utilized the Thiessen polygon method to deal with the geographical characteristics of Indonesia.
The Thiessen polygon method creates a tessellation .
way to divide an area into regular subarea.
that encloses all locations that are closer to the central point than to any other point (Anselin, 2.
The boundary of a Thiessen polygon is artificial, based on the central point of each district.
Research by Miranti and Mendez .
and Santos-Marquez et al.
utilized Thiessen polygons to create proximity at the district level.
Other research conducted by Mendez and Siregar .
used Thiessen polygons to analyze the space-time dynamics of the administration of two levels of unemployment by simultaneously accounting for their serial persistence, spatial dependence, and common factors.
Figure 2 displays Thiessen polygons and their borders at the district level in Indonesia.
After the Thiessen polygons have been constructed, the researcher defines the neighborhood of each district.
The research uses a queen contiguity matrix to represent the neighbors for each district.
The food security measurements are based on data from Statistics of Indonesia and the National Food Agency data.
The first measurement uses the food security and vulnerability atlas (FSVA) with its index measurement.
The FSVA is a thematic map that gives a geographical visualization of areas susceptible to food insecurity.
The design of the FSVA is based on three aspects of food security, namely food availability, affordability/ .
where the represent the elements of the spatial weight matrix W between two districts and , is the mean of , and , = 1,A, n.
Global MoranAos I is a statistic used to measure the global spatial autocorrelation of a dataset, not simply the local spatial Spatial autocorrelation refers to the degree to which similar values are clustered together in a spatial distribution.
MoranAos I helps to determine whether the observed spatial pattern is clustered, dispersed, or random.
The value of MoranAos I ranges from -1 to 1, with a value near -1 indicating dispersion and a value close to 1 indicating the presence of clustering (Grekousis, 2.
Knowing the local measurements aids policymakers in making region-specific decisions and adds nuance to spatial LISA identifies local clusters and spatial outliers, unlike Global MoranAos I, which provides a single summary LISA detects local patterns, identifies Auhot spots,Ay and assesses the influence of individual locations on the global statistics, revealing spatial relationships not visible in the global analysis.
The formula for LISA (Anselin, 1.
where the represent the elements of the spatial weight matrix W between two districts and , and standardized numbers for observations and .
Figure 2.
Centroid.
Thiessen Polygon and Spatial Connectivity (Queen Matri.
at District Level in Indonesia 2019 Indonesian Journal of Geography.
Vol 57.
No.
access, and utilization (Badan Pangan Nasional, 2.
The FSVA categorizes the scores for the food security index into six categories: very vulnerable, vulnerable, slightly vulnerable, slightly resistant, resistant, and very resistant.
The second measurement is calorie intake, which measures how many calories are consumed daily.
In Indonesia, the suggested calorie intake is based on the calories consumed in one day.
In 2013, the Government of Indonesia issued the Regulation of Ministry of Health Republic of Indonesia No 75 Year 2013 about Nutrition Standard .
which states a minimum requirement for daily food of 2,150 kcal/person/day and for daily protein of 57 grams/person/day.
The threshold was changed in 2019, when the caloric intake threshold was cut to 2,100 kcal/person/day (Regulation of Ministry of Health Republic of Indonesia No 28 Year 2019 about Nutrition Standard, 2.
Households or persons whose calorie intake is below the threshold are deemed to have experienced food The third measurement is the food variety score (FVS).
The FVS is defined as the number of food items, on a list of 104 food items, eaten by any household member on the previous day (Hatlyy, et al.
, 2.
In the original research, all the food items were given an equal weight.
The households were divided into tertiles, to give a high, a medium and a low FVS, where a high was for more than 18 items, a medium score for between 14 and 18 and a low score for between 4 and 13 Because this study only had limited data, it counted the percentage of food items consumed by the household directly.
The maximum percentage was 100 per cent, which would mean that the household consumed all the food items on the SUSENAS list .
food item.
The last food security measurement that was used in this study is the dietary diversity score (DDS), which means the number of food groups represented in what was consumed by the household on the previous day (Hatlyy et al.
, 2.
The DDS was generated from the same list as the FVS.
Different definitions of DDS have been suggested in recent years, with variations in the number of food groups to include and their composition and in the dietary assessment methods used.
The number and composition of the food groups are often different since they reflect the aim of a specific study.
therefore, there is no consensus regarding the ideal number of food groups in a DDS.
There are challenges to the application of the DDS, because of the differences in food in different cultures.
DDS developed for one culture will not necessarily be the same as one used in another, but the theory and approach for developing the score can be used across cultures (Hodgson, et , 1.
This study used 13 food groups because of the data available in the SUSENAS data, which aligns with the study by Hatlyy, et al.
The food groups are grains, tubers, fish, shrimp or shellfish, meat, egg or milk, vegetables, nuts, fruits, oil and coconuts, beverages, seasonings, other foods, and prepared food.
In this research, the DDS was expressed as a percentage.
The higher the percentage, the greater the food The assumption is that a household will first secure stable food and then move into tertiary food.
In the research of Hatlyy, et al.
, the DDS was divided into tertiles, which gave three categories: high for a score of 3 to 8, medium for a score of 6 or 7 and low for a score of 2 to 4.
Results and Discussion This results and discussion section will be divided into three main parts: a statistical description of the data, the Global MoranAos I analysis, and the LISA analysis.
Descriptive statistics .
istribution of data, rang.
are essential for understanding the data: they help to quantify the extent of variation and allow for comparisons between different datasets.
The output for the first analysis is presented in Table 2, which provides descriptive statistics on the food security Table 1.
Variables and Data Sources Variable Food Security and Vulnerability Atlas (FSVA) Source National Food Agency Calorie Intake SUSENAS 2019a.
Central Bureau of Statistics Food Variety Score (FVS) SUSENAS 2019a.
Central Bureau of Statistics Dietary Diversity Score (DDS) SUSENAS 2019a.
Central Bureau of Statistics Source: Statistics of Indonesia and National Food Agency Table 2.
Descriptive Statistics for Food Security Measurements Description Food Security Vulnerability Atlas
Observations Mean
Dietary Diversity Score (DDS) Calorie Intake
Std
Min Max
Source: Statistics of Indonesia and National Food Agency, 2019
Food Variety Score
(FVS)
MAPPING FOOD SECURITY IN INDONESIA
Rusli Abdulah.
The data shows that the FSVA ranges from 56 to 90.
05, with a standard deviation of 14.
38 and a mean The standard deviation of the calorie intake is the highest among the food security measurements constructed from household surveys.
The value is 13, whereas the standard deviations for the DDS and FVS are 5.
35 and 3.
69, respectively.
In other words, the calorie intake scores are more spread out, indicating greater variability.
The second analysis calculates the Global MoranAos I Figure 3 presents the Global MoranAos I values for the four variables related to food security in Indonesia.
Overall, the Global MoranAos I value for all the variables and the significant p-values indicate spatial clustering in food security and nutritional metrics across Indonesia.
The highest degree of spatial autocorrelation is the food variety score, with a value for MoranAos I of 0.
92 and a p-value of 0.
001, suggesting a high degree of spatial clustering in the food variety score.
the other hand, the calorie intake variable has the lowest value for the Global MoranAos I, at 0.
A Global MoranAos I scatter plot only informs us that there is spatial autocorrelation, but does not specify the location of that Local indicators for spatial autocorrelation can identify and analyze local spatial patterns within a dataset, whereas global measures like the Global MoranAos I cannot achieve this through the decomposition of global indicators into the contributions of individual observations.
The decomposition allows the study to detect local clusters, hot spots, cold spots, and spatial outliers (Anselin, 1.
This localized analysis is essential for understanding the spatial heterogeneity and pinpointing areas that exhibit significant spatial autocorrelation.
It is valuable as it allows policymakers to deliver the policy more specifically in smaller areas.
As a result, the policy outcome will get good results with minimum Based on the LISA cluster map in Figure 4, regions with a low-low cluster appear mainly in eastern Indonesia for all the food security measurements.
The low-low clusters .
lue area.
have low food variety scores and are surrounded by other regions with low scores.
This indicates areas where food insecurity arises from limited food variety that is due to geographical isolation, a lack of infrastructure, or economic challenges affecting food access and diversity.
Research by Amrullah et al.
highlights that households in rural or backward regions are more likely to experience food insecurity, which is the result of geographical In addition.
Akbar et al.
revealed that living in an isolated location, especially in an underdeveloped nonurban region, can contribute to food insecurity.
The problems with geographical access may hinder access to markets, resources, and infrastructure, impacting food availability and affordability (Thow et al.
, 2.
Geographical isolation leads to a lack of infrastructure because of the difficulty of establishing access in a complex geographical region.
Eastern Indonesia is an underdeveloped region with challenging geographical conditions, such as in Papua, which means that investment in infrastructure such as roads and ports is rare.
Research by Wahyuni et al.
found that, in 2014, only 26.
39 per cent of the area of Papua and West Papua had paved roads and only 39.
4 per cent had all-season road access.
In the next six years, these percentages Figure 3.
MoranAos I Scatter Plot on Food Security at District Level in Indonesia 2019 Indonesian Journal of Geography.
Vol 57.
No.
did not increase significantly, rising to 29.
59 per cent of rural areas with paved roads and 39.
32 per cent with all-season road In contrast, the high-high clusters .
ed area.
are primarily located in the western part of Indonesia.
These regions have high food security and are surrounded by other regions with similarly high scores.
In addition, this clustering suggests areas where food security is consistently high, which may be due to favorable agricultural conditions, diverse food markets, effective distribution systems and a high wealth level.
For instance.
Java Island has a high cluster presence for all food security measurements.
Java Island has good infrastructure, such as roads, for effective distribution systems (Wahyuni et al.
, 2.
Java Island also has a long history, under the Dutch colonial government, of a focus on establishing infrastructure, such as building roads, railways, telegraphs, bridges, and irrigation This policy encouraged Java to produce more profitable commodities for the European market (Rinardi.
Java Island also has good soil for planting, compared to outer Java Island, because of its volcanic mountains (Anda & Dahlgren, 2020.
Purwanto et al.
, 2.
In addition.
Western Indonesia, especially Java Island, has diverse food markets because of its wealth.
The western part of Indonesia has higher socio-economic characteristics than eastern Indonesia.
As for the other areas, high-low .
areas are outliers that represent regions with high food security scores surrounded by areas with food insecurity .
ow scores for food These outliers might be urban centers or regions with better access to food that stand out from their surroundings in terms of food access.
High-low areas could be used as the starting point for policymakers to enhance their policy on food access equity.
In other words, high-low areas could help the government to spotlight areas that they should prioritize.
For instance.
Merauke City in Papua Province is a highlow area in three of the food security measures .
he food variety score does not show this resul.
The government could quickly identify why the Mappi and Boven Digoel Districts .
n the north side of Merauk.
include low-low clusters by comparing the determinants of food security in the two The government could then deliver specific policies to enhance particular determinants to raise the food security level in Mappi and Boven Digoel.
Like high-low regions, the outliers in low-high .
ight blu.
areas could help policymakers to decide what to do.
Low-high outliers could be areas facing specific challenges, such as economic constraints or logistical issues, preventing them from benefiting from the food security achievements of their surrounding regions.
Local analysis of the clusters using the four food security measures provides greater nuances than the ESDA standard, which focuses on a single food security measurement.
Researchers can also identify areas that fall into low-low clusters from the perspective of the four food security As a result, it can be determined which areas should be prioritized for development by applying different approaches simultaneously.
Furthermore, the districts mentioned in Table 3 are in a low-low cluster for all four food security measurements.
This indicates that these regions have consistently low scores across all the indicators, suggesting significant challenges and systemic issues in food security.
These may include limited agricultural productivity, inadequate supply chains, and socio-economic barriers that prevent access to a diverse and sufficient diet.
By focusing on these areas, policymakers can work towards improving food security and reducing the disparities highlighted in these regions by the LISA map.
For example.
West Southeast Maluku is included in the low-low cluster for all food security measurements.
The government must improve peopleAos purchasing power so that they can meet the recommended calorie intake.
In line with that, improving logistics, based on sea transportation into the region, is necessary to provide different foods.
West Southeast Maluku .
he Tanimbar Islands Distric.
is located in the middle of the Arufu Sea.
Different policies need to be applied to low-low regions in mainland areas, such as Paniai District.
This district is in the middle of West Papua Province, which has no sea .
t is a landlocked distric.
The policies that need to be implemented in this area differ from those in the West Southeast Maluku Calorie Intake Dietary Diversity Score Food Vulnerability Atlas Food Variety Score Figure 4.
LISA Map of Food Security in Indonesia 2019
MAPPING FOOD SECURITY IN INDONESIA
Rusli Abdulah.
District ID Table 3.
Regions with Similarity (Low-Low Cluster.
in Food Securities Measurements Food Dietary Calorie Food Security and District Intake Vulnerability Atlas Variety Score Diversity Score West Southeast Maluku Southeast Maluku Aru Islands Seram Bagian Timur Tual City Fakfak Kaimana Nabire Paniai Asmat Sarmi Keerom Waropen Mamberamo Raya Central Mamberamo Yalimo Puncak Dogiyai Deiyai Source: AuthorAos analysis of data from Statistics of Indonesia and National Food Agency.
Repairing roads or improving access to pioneer logging can be a solution to encourage the distribution of varied foodstuffs.
Furthermore, high-low clusters are outliers in which areas with low scores surround regions with high food security These areas, often urban centers with better food access, highlight disparities and can guide policymakers in addressing food access equity.
In addition, low-high clusters indicate regions that, despite being surrounded by food-secure areas, face challenges such as economic or logistical issues.
These clusters provide valuable insights for targeted policy interventions to improve food security in less advantaged Local analysis using the four food security measures provides greater nuance than standard ESDA, identifying areas in low-low clusters across all four indicators.
Regions like West Southeast Maluku and Paniai District face systemic food security challenges, including low agricultural productivity and socio-economic barriers.
West Southeast Maluku requires better sea-based logistics and improved purchasing power to meet the inhabitantsAo calorie needs, while Paniai District, being landlocked, needs a better road infrastructure to facilitate food Tailored policies for each area can help address these disparities and improve overall food security.
Further research should include investigating the determinants of each food security measurement using spatial and non-spatial econometric approaches.
Secondly, a Conclusion This research identifies high-high, low-low, low-high and high-low clusters for four different food security measurements:
food security and vulnerability atlas (FSVA), dietary diversity score (DDS), food variety score (FVS), and calorie intake.
There are low-low regions for all food measurements in eastern Indonesia.
The potential determinants are geographical isolation, economic performance, and infrastructure .
onnectivity and irrigatio.
On the other hand, high-high clusters, which are predominantly located in the western part of Indonesia and particularly on Java Island, indicate regions with consistently high food security.
This is likely due to favorable agricultural conditions, diverse food markets, effective distribution systems, and higher wealth levels.
Java Island benefits from robust infrastructure, such as well-developed roads and irrigation systems, facilitating efficient food distribution.
Additionally, volcanic mountains contribute to its fertile soil, enhancing agricultural productivity compared to other regions.
Indonesian Journal of Geography.
Vol 57.
No.
convergence approach could be applied to address how the region could quickly achieve food security.
A club convergence analysis would help policymakers and researchers to identify the factors that have enabled some regions to achieve food Such areas could be used as benchmarks to help other districts to achieve food security.
Kadir.
Prasetyo.
, & Rudiana.
Food Insecurity Experience Scale Measurement of Agricultural Households in Indonesia: Analysis of the Agricultural Integrated Survey Results.
International Conference on Agriculture Statistics.
May.
https://w.
net/publication/370873070 Mendez.
, & Siregar.
Regional unemployment dynamics in Indonesia: Serial persistence, spatial dependence, and common factors.
Letters in Spatial and Resource Sciences, 16.
, 1Ae20.
https://doi.
org/10.
1007/s12076-023-00364-6 Ministry of Agriculture Indonesia.
World Food Programme.
Food Security and Vulnerability Atlas of Indonesia (FSVA) 2015.
Miranti.
, & Mendez.
Social and economic convergence across districts in Indonesia: A spatial econometric approach.
Bulletin of Indonesian Economic Studies, 59.
, 421Ae445.
https:// org/10.
1080/00074918.
Moran.
Notes on continuous stochastic phenomena.
Biometrika, 37.
Ae.
, 17Ae23.
https://doi.
org/10.
BIOMET/37.
Nugroho.
Hanani.
Toiba.
, & Sujarwo.
Does social capital improve food security? Evidence from the Indonesian family life survey.
Asian Journal of Agriculture and Rural Development, 12.
, 138Ae147.
https://doi.
org/10.
55493/5005.
Presidential Regulation (Perpre.
Number 66 of 2021 Concerning the National Food Agency .
Purwanto.
Abdul Gani.
, & Sukarman.
Karakteristik Mineral Tanah Berbahan Vulkanik dan Potensi Kesuburannya di Pulau Jawa.
Jurnal Sumberdaya Lahan, 12.
, 83.
https://doi.
org/10.
21082/jsdl.
Regulation of Ministry of Health Republic of Indonesia No 28 Year 2019 about Nutrition Standard .
Regulation of Ministry of Health Republic of Indonesia No 75 Year 2013 about Nutrition Standard .
Rinardi.
Java and Outer Island: Economic inequality and inter-island shipping policy in Indonesia until the E3S Web of Conferences, 202.
https://doi.
org/10.
e3sconf/202020207070 Santos-Marquez.
Gunawan.
, & Mendez.
Regional income disparities, distributional convergence, and spatial effects: Evidence from Indonesian regions 2010Ae2017.
GeoJournal, 87.
, 2373Ae2391.
https://doi.
org/10.
1007/s10708021-10377-7 Thow.
Sharma.
, & Rachmi.
An analysis of IndonesiaAos shrinking food security policy space under the WTO.
Food Security, 11.
, 1275Ae1287.
https://doi.
org/10.
s12571-019-00967-2 Vidyarini.
Martianto.
, & Syarief.
Evaluation of food and nutrition security level at provincial level based on outcome indicators in Indonesia.
Jurnal Gizi Dan Pangan, 16.
, 1Ae10.
https://doi.
org/10.
25182/jgp.
Wahyuni.
Ikhsan.
Damayanti.
, & Khoirunurrofik.
Inter-district road infrastructure and spatial inequality in rural Indonesia.
Economies, 10.
https://doi.
org/10.
Widada.
Masyhuri.
, & Mulyo.
Determinant factors of food security in Indonesia.
Agro Ekonomi, 28.
, 205.
https://doi.
org/10.
22146/jae.
Yuliana.
Village funds utilization strategy to improve food Efficient: Indonesian Journal of Development Economics, 1.
, 143Ae156.
https://doi.
org/10.
15294/efficient.
Acknowledgement The writer thanks the Indonesia Education Endowment Fund (LPDP).
Ministry of Finance.
Republic of Indonesia, for the scholarship.
References