Journal of Natural Resources and Environmental Management
http://dx.doi.org/10.29244/jpsl.15.5.804

RESEARCH ARTICLE

Land Use Change and Future Prediction in Banggai Islands Regency, Central
Sulawesi, Indonesia
Hidayat Arismunandar Katilia,b, Syartiniliac, Firman Irmansyahd, Widiatmakae
a

Natural Resources and Environmental Management Science Study Program, IPB University, IPB Baranangsiang Campus, Bogor, 16129,
Indonesia
b
Faculty of Agriculture, Tompotika Luwuk University, Luwuk, Banggai Regency, 94711, Indonesia
c
Department of Landscape Architecture, Faculty of Agriculture, IPB University, IPB Dramaga Campus, Bogor, 16680, Indonesia
d
Indonesian System Dynamics Center, Bogor, 16128, Indonesia
e Department of Soil Science and Land Resources, Faculty of Agriculture, IPB University, IPB Dramaga Campus, Bogor, 16680, Indonesia

Article History
Received
27 December 2024
Revised 24 April 2025
Accepted 27 May 2025
Keywords
Banggai Islands, land use
mapping, prediction,
spatial distribution

ABSTRACT
Land use and land cover (LULC) changes can influence policies in a region due to economic and social
conditions caused by population growth. The objective of this study is to analyze and map LULC
changes in 2002, 2012, and 2022 using the Random Forest approach on Google Earth Engine, and
to predict land use in 2042 using Markov-CA, thereby supporting the provision of accurate and
sustainable policy data related to LULC in Banggai Islands Regency. This method can provide
accurate information about the spatial distribution of rational LULC, balancing development
demands with sustainable environmental protection. The study's results indicate that LULC has
undergone significant changes from 2002 to 2022. There has been an increase in plantation land,
open land, and settlements originating from forest and scrubland. Predictions of LULC changes in
2042 show an increase in plantations, settlements, and open land, while other land uses are
declining. Effective land use policies require spatial planning that considers the potential and
limitations of land, as well as the space needs for residential, agricultural, and forest areas. This
approach will facilitate the application of land conservation principles in sustainable, balanced
agricultural and non-agricultural development in Banggai Islands Regency.

Introduction
Land is a vital natural resource that humans have utilized for various activities and sustenance. Over the years,
land use and land cover change (LULC) has become a significant concern for many nations. Land use refers to
the alterations made by humans to the Earth's surface, while land cover describes the physical and
environmental characteristics of that surface [1]. Land is defined by distinct properties and essential elements
that characterize its use and management. LULC is a prevalent issue in Indonesia, including the Banggai
Islands Regency. Key drivers of LULC include rapid population growth, migration, and the transformation of
rural areas into urban centers [2]. Given the region's limited land resources, the government must implement
effective regulations for land management. Over the past two decades, numerous studies have explored
various aspects of LULC; however, information remains incomplete in certain areas. Recent work by Anandhi
et al. [3] has provided both narrow and broad definitions of land resources, encompassing ecological
components such as climate, water, soil, landscape, flora, and fauna, as well as the socioeconomic systems
that interact with agriculture, forestry, and other land uses within a defined system boundary.
Identifying land use change is critical to addressing the rapid and uncontrolled change that is occurring, and
is critical to understanding land development, loss, and degradation [2]. Assessing LULC changes is essential
for the sustainable planning of each region [4], as unmanaged land use can lead to degradation and loss of
Corresponding Author: Widiatmaka
widiatmaka@apps.ipb.ac.id
Agriculture, IPB University, IPB Dramaga Campus, Bogor, Indonesia.

Department of Soil Science and Land Resources, Faculty of

© 2025 Katili et al. This is an open-access article distributed under the terms of the Creative Commons Attribution (CC BY) license, allowing
unrestricted use, distribution, and reproduction in any medium, provided proper credit is given to the original authors.

Think twice before printing this journal paper. Save paper, trees, and Earth!

land capability, as well as overlapping regulations in land use in each sector. According to Alam et al. [5],
evaluating land use and cover changes from various perspectives, such as land use planning and sustainable
development, is critical to meeting the population's diverse needs [6].
Recent research on LULC has gained significant attention among researchers in Sulawesi [5–7]. According to
Statistics Indonesia 2022, the area of forest cover in Central Sulawesi is over six million hectares [8]. This data
is the basis for the importance of the research in Banggai Islands Regency. The area's rapidly increasing
population is driving the need for land expansion to meet growing demands [9,10]. Furthermore, Dibaba et
al. [11] note that population growth is a primary factor influencing LULC changes. Additionally, exploring LULC
changes aims to enhance understanding through integrated research priorities [12], focusing on the causes
and effects of these phenomena in the present and future contexts [13]. Furthermore, LULC studies are
essential for updating data accurately [12] and providing policymakers with a comprehensive understanding
of effective and sustainable land resource management [13,14]. There is a growing acknowledgment that
policies significantly influence land use and management practices, extending beyond the traditional
frameworks of land use and environmental planning systems. According to Munthali et al. [15], Kullo et al.
[16], LULC information is crucial for environmental planners and observers due to its significant impact on
both local and global environments. A wide variety of incentives and disincentives, including fiscal and tax
incentives, industry subsidies, and the design of infrastructure or transport programmes, significantly
influence decision-making and interventions across our region. However, these measures often result in
negative consequences, such as loss of agricultural land, environmental degradation, and an increase in costly
services [17].
Research on LULC is crucial for urban planners [18], policymakers in natural resource management [16,18],
and agricultural and environmental experts [19,20]. Krawchenko and Tomaney [18] assert that the
framework promotes an evaluation of incentives, disincentives, and complementarities across various
policies and practices, highlighting the necessity for greater alignment to achieve land management
objectives. Remote sensing serves as an effective tool for swiftly detecting changes in land cover. In the
Banggai Islands, researching LULC change is essential for formulating timely policies, making remote sensing
a necessary component of this study. Remote sensing involves processing satellite images through opensource platforms that utilize big data, such as Google Earth Engine (GEE), which is commonly used for
mapping and monitoring land cover, vegetation, and urban expansion [21]. The GEE platform supports
various sensors, including Landsat and Sentinel, allowing for the storage and processing of large datasets to
aid decision-making [22]. Geospatial remote sensing data available on GEE has been extensively applied in
research related to regional socio-economics [23], land cover mapping [24], land change detection [25],
urban studies [26], rice mapping [27], rice crop detection [28], and oil palm plantations [29]. Furthermore,
GEE features a programming language that enables rapid and accurate analysis of land cover changes [30].
The platform is both efficient and computationally advantageous, providing various algorithms, including the
Random Forest (RF) algorithm, which facilitates access and classification [31–33]. Developed by Breiman in
2001, the RF method is a non-parametric classification technique that employs multiple decision trees for
effective decision-making [34,32]. Landsat imagery with a spatial resolution of 15–60 m and sentinel-2 with
a spatial resolution of 10–60 m can be used in LULC classification [35,36] with a classification process using
the Random Forest (RF) algorithm through the Google Earth Engine (GEE) mapping platform to monitor land
use in the Banggai Islands Regency. The medium resolution of these images allows for adequate detail in area
identification. Additionally, land use predictions in this study are conducted using Markov Chain and Cellular
Automata (Markov-CA). Both Markov-CA models have been widely applied in spatial and geographic
analyses, yielding accurate results in land use change assessments [37–39]. The Markov-CA model provides
a more robust and comprehensive framework for predicting land use changes compared to single-method
models. Its application in the Banggai Islands serves as a valuable tool for forecasting future land use
dynamics. The Cellular Automata (CA) component utilizes neighbourhood interactions to model systems that
evolve based on simple, predefined rules over time. Meanwhile, Markov chains, a fundamental aspect of
probability theory, have increasingly been integrated into spatial sciences. This study employs the MarkovCA approach to project land use changes, aiming to offer a spatial perspective that can inform regional
planning aligned with anticipated land use patterns in the Banggai Islands Regency.
The Banggai Islands are home to a variety of endemic species, including primates, reptiles, and birds.
However, the biodiversity of the Banggai Islands is threatened by human activities such as illegal logging and
unsustainable practices, including land use change from forest to agricultural land, which results in the loss
of habitat for these endemic species. Current regulations governing the management of natural resources

http://dx.doi.org/10.29244/jpsl.15.5.804

JPSL, 15(5) | 805

and biodiversity in the Banggai Islands may be ineffective, partly due to limited human and financial resources
that hinder conservation efforts. In addition, the involvement of local communities in managing these
resources is still suboptimal, potentially leading to conflicts over resource use. The case study conducted in
the Banggai Islands can provide valuable lessons on how to overcome complex land management challenges
so that natural resources can be utilized appropriately and effectively, monitoring changes in land use
provides valuable recommendations for policymakers in formulating appropriate and effective strategies.
Furthermore, the findings of this study can be used as a reference for land management and for future
research on more effective integration of various land uses to protect biodiversity in the Banggai Islands.
Over the past three decades, the Banggai Islands Regency has undergone significant changes in LULC,
primarily driven by population growth, agricultural expansion, and various development initiatives. However,
systematic monitoring of these changes remains insufficient, and assessments are infrequently conducted.
This study aims to evaluate and simulate the spatial and temporal dynamics of LULC changes in the Banggai
Islands Regency. Specifically, it seeks to map and quantify LULC changes from 2002 to 2022 using Landsat 5,
7, and Sentinel-2 satellite imagery. Additionally, the research projects on land use conditions in 2042 provide
insights into land use conditions and the relevance of policies aimed at promoting appropriate and
sustainable land use in the region.

Materials and Methods
Research Location
The research was conducted in the Banggai Islands district of Central Sulawesi province, located
geographically between 1°06‘30″–1°35’58″ N and 122°37‘6.3″–123°40’1.9” E, in the northeastern part of
Sulawesi Island, covering an area of 2,376.71 km². To the north, it borders Banggai Regency, while the Maluku
Sea lies to the east (Banggai Islands Central Statistics Agency 2022). The study area comprises 12 districts, as
illustrated in Figure 1.

Figure 1. Research location.

This journal is © Katili et al. 2025

JPSL, 15(5) | 806

Data Collection and Processing
The processing and analysis of remote sensing imagery begins with data pre-processing. According to Andries
et al. [40], Wei et al. [41], land use and cover can be visually classified from images as needed. Agricultural
patterns in the Banggai Islands Regency apply mixed agricultural patterns, such as agroforestry systems, so
agricultural land use and plantations are classified into plantations. In this study, LULC types will be digitally
classified, such as forests, plantations, Built-up, open land, mangroves, rice fields, shrubs, and water bodies,
using the Random Forest (RF) algorithm (ntree=500) on the Google Earth Engine (GEE) [42]. The GEE platform
offers corrected Landsat image data, streamlining the pre-processing phase. The RF algorithm, an ensemble
model based on decision trees, was selected for its effectiveness in classifying remote sensing data [42,43].
Landsat and Sentinel image processing is done directly in GEE, with the output image set at a spatial
resolution of 30 metres [44]. It employs a bootstrap method on 70% (1,600 sampling total) of the training
dataset to create a new dataset, followed by a majority decision process that determines the mode from
each decision tree's output. The dataset was obtained using LULC class polygons and distributed across all
geographical characteristics of the study location, covering the entire administrative area.
This case study utilizes Landsat-5 Thematic Mapper (TM) and Landsat-7 Enhanced Thematic Mapper Plus
(ETM+) imagery from 2002 and 2012, along with Sentinel-2 MultiSpectral Instrument (MSI) imagery from
2022. The approach refers to Simonetti et al. [45], combining various bands to assess land use in the study
area. The bands from all satellites include Green Band, Blue, Red, Near Infrared (NIR), and Short-wave
infrared (SWIR). In addition, there are derivative indices used, namely NDVI, NDWI, NDMI, EVI, SAVI, ARVI,
IBI, SLAVI, NDBI, GNDVI, MNDWI, MMRI, DVI, RVI, PSSRA, and LWSI, to improve processing performance.
Validation involves assessing 30% of all samples for accuracy using the RF algorithm result data, with
classification results compared against field checks or available reference data [46,47]. Field checks were
conducted to gather information on land use, and the validation results were used to evaluate the accuracy
of the 2022 image interpretation using the kappa value.
Accuracy testing is conducted to evaluate classification errors and determine the percentage of accuracy.
This is achieved by creating a contingency matrix, also known as an error or confusion matrix, calculated using
the error matrix function in the GEE platform (Table 1). From this matrix, various accuracy metrics can be
derived, including overall accuracy, user accuracy, producer accuracy, and kappa accuracy. According to
Sivanpillai et al. [48], Hidayah et al. [49], the minimum accuracy value of more than 80% is required for land
use interpretation to be classified as excellent. Overall accuracy is calculated using the formula where X
represents the number of correctly interpreted land cover points aligned with ground truth results, and N is
the total number of ground truth points [50]. Kappa value equal to 1 indicates the classification results get
perfect agreement, while a kappa value equal to 0 indicates the classification results cannot be approved,
Furthermore, Foody [51] defines kappa accuracy with equations (1 to 4). The kappa coefficient (κ) assesses
how classification results compare to chance, calculated from the error matrix by combining diagonal and
off-diagonal elements [48]. The range of kappa values and their classifications is presented in Table 2.
Table 1. Error matrix.
Classification result of land use
Pi+
Pi+
…
…
Pi+
Total

Pi+
Xii

Reference land use
Pi+ …
…
Pi+
Xii
Xii
Xii

Xi+

Xi+

Xi+

Xi+

Xii
Xi+

Total
X+i
X+i
X+i
X+i
X+i
N

Where, P+i: The i-th land use type of the classification result; Pi+: The i-th land use type from field check; X+i:
Number of classification result points on the i-th land use type; Xi+: Number of points from field check on the
i-th land use type; Xii: Number of land use types that correspond to the classification result and the field
check results; i: Row or column; r: Number of land use types; N: Number of land use points validated; k: Kappa
value. The overall accuracy between the remote sensing classification data and the reference data can be
calculated using the following Equations 1–4.

http://dx.doi.org/10.29244/jpsl.15.5.804

JPSL, 15(5) | 807

User accuracy (UA) =

𝑋𝑖𝑗
𝑋𝑖+

𝑥 100%

Producer accuracy (PA) =
Overall accuracy (OA) =
Kappa accuracy (K) =

1
𝑁

𝑋𝑖𝑗

(1)

𝑥 100%

(2)

∑𝑟𝑖=1 𝑋𝑖𝑖 𝑥 100%

(3)

𝑋𝑖+

𝑟
𝑁 ∑𝑟
𝑖=1 𝑥𝑖𝑖 − ∑𝑛=1(𝑋𝑖+ + 𝑋+𝑖 )

(4)

𝑁2 −∑𝑟
𝑛=1(𝑋𝑖+ +𝑋+𝑖 )

Where Xii is the diagonal value of the confusion (error) matrix, and Xij is used for the land use class. The matrix
size r = 2, Xi+ is the number of pixels in the row i obtained from remote sensing analysis, X+j is the number of
columns of j obtained from referenced land use data, and N is the number of pixels in the data.
Table 2. Categorization based on a range of kappa values.
Kappa range
<0.00
0.00–0.20
0.21–0.40
0.40–0.60
0.61–0.80
>0.81

Interpretation
Very poor
Poor
Fair
Moderate
Good
Very good

Source: [52,48].

LULC change prediction can be analyzed through multitemporal spatial changes [53,54]. This approach is
recommended for its ability to model land use based on temporal trends. Cellular Automata (CA) is a crucial
geospatial element that highlights the dynamics of change events and can simulate the characteristics of
complex systems both spatially and temporally. In this study, spatial data from the initial year (T0), specifically
2022, is utilized alongside 20 years of change data from 2002 to 2022, employing Markov Chain and Cellular
Automata methods. Research by Bondansari et al. [55], Sejati et al. [56], Bindajam et al. [57] indicates that
the Markov Chain and Cellular Automata (Markov-CA) method yields optimal predictions based on kappa
values, particularly when the predicted year distance matches that of previous periods [58–60]. This method
helps identify potential land use changes over specific time frames [61,62]. The integration of these two maps
serves as input data for training the Markov-CA model, facilitating an evaluation of the 2022 land use
simulation. The Markov-CA model can generate predictions for various LULC and development scenarios, as
well as illustrating interactions between regions.
The modelling process begins with land use data from 2012 and a transition area matrix covering 2002–2012
to predict land use for 2022, which is then validated against actual land use data from that year. The MarkovCA modelling and prediction for LULC is facilitated by the MOLUSCE (Modules for Land Use Change
Simulations) plugin within the software QGIS 2.16 series. The prediction process involves several steps: first,
input raster data from land use maps for 2002, 2012, and 2022, along with factors influencing LULC change,
is prepared. Next, a correlation assessment is performed using the MOLUSCE plugin to evaluate relationships
between rasters, where a value of 0 indicates no relationship, while values of 1 and -1 indicate strong
relationships. This information is then automatically input into the transition probability matrix (TPM), which
reflects the proportion of pixels expected to change from one land use type to another. The multilayer
perceptron method is applied to optimize iterations and minimize the loss function, enhancing accuracy.
Subsequently, Cellular Automata processes the land cover images to generate a probability transition matrix,
transition area matrix, and iteration results for the desired land use prediction. Accuracy validation can also
be conducted through ground checks, which are recorded in an error matrix. The error matrix serves as a tool
for measuring the accuracy of the LULC change map. According to Jensen and Lulla [63], Jensen [64], Artikanur
et al. [65], established guidelines, a validation value of 85% allows for predictions of LULC changes in 2042,
using 2022 as the base input. Evaluation of the 2042 prediction model using the accuracy listed in the
MOLUSCE plugin data processing. The overall process of transforming land use change data into predictive
models is illustrated in Figure 2.

This journal is © Katili et al. 2025

JPSL, 15(5) | 808

Figure 2. Data processing of land use and land cover change in 2002, 2012, 2022, and land use prediction in 2024.

Results and Discussion
Results
Land Use and Land Cover (LULC) Identification in 2002, 2012, and 2022
LULC identification was conducted through image classification from Google Earth Engine (GEE) for 2002 and
2012, utilizing a combination of Landsat 5 and 7 imagery and Sentinel-2 imagery for 2022. The classification
included eight land use types: forests, gardens, settlements, open land, shrubs, rice fields, mangroves, and
water bodies. The overall accuracy of the RF classification for 2002, 2012, and 2022 was found to be 84%,
90%, and 89%, respectively. According to Sivanpillai et al. [48], Hidayah et al. [49], these accuracy values fall
within the excellent category, exceeding the 80% threshold. This indicates that the land use data for the
Banggai Islands Regency meets the validation requirements, making the resulting land use map suitable for
this study. The area proportions and LULC maps for the Banggai Islands Regency for 2002, 2012, and 2022
are presented in Table 3 and Figures 3 and 4, while the patterns of land use/cover change from 2002 to 2022
are detailed in Table 4.
Table 3. Land use and land cover area in 2002, 2012, and 2022, in Banggai Islands Regency.

No

Land use and land cover

1
Forest
2
Plantation
3
Built-up
4
Open land
5
Rice fields
6
Shrubs
7
Mangroves
8
Waterbody
Total

http://dx.doi.org/10.29244/jpsl.15.5.804

2002
ha
132,786.9
70,809.5
529.7
3,908.3
142.7
26,230.1
1,588.9
1,674.9
237,671

%
55.9
29.8
0.2
1.6
0.1
11.0
0.7
0.7
100

Land area
2012
ha
%
94,780.2 39.9
130,911.5 55.1
1,014.2
0.4
2,121.3
0.9
313.2
0.1
6,158.2
2.6
1,446.4
0.6
926.0
0.4
237,671 100

2022
ha
%
97,148.4 40.9
104,194.0 43.8
1,795.3
0.8
10,451.4
4.4
315.2
0.1
19,430.9
8.2
1,636.8
0.7
2,671.1
1.1
237,671 100

JPSL, 15(5) | 809

Figure 3. LULC map of 2002, 2012, and 2022 in the Banggai Islands Regency.

Precentage cover (%)

60.0
50.0
40.0
30.0
20.0
10.0
0.0

Forest

Plantation

Built-Up

Open land

Rice fields

Shrubs

Mangroves

Waterbody

2002

55.9

29.8

0.2

1.6

0.1

11.0

0.7

0.7

2012

39.9

55.1

0.4

0.9

0.1

2.6

0.6

0.4

2022

40.9

43.8

0.8

4.4

0.1

8.2

0.7

1.1

Figure 4. Distribution of LULC patterns 2002–2022, the Banggai Islands Regency.
Table 4. LULC from 2002 to 2022 period.
Extent
No

Land use

1
2
3
4
5
6
7
8

Forest
Plantation
Built-up
Open land
Rice fields
Shrubs
Mangroves
Waterbody

This journal is © Katili et al. 2025

2002–2012
ha
%
-38,006.7
60,102.0
484.6
-1,787.0
170.5
-20,071.9
-142.5
-749.0

-16.0
25.3
0.2
-0.8
0.1
-8.4
-0.1
-0.3

2012–2022
ha
%
2,368.2
-26,717.5
781.1
8,330.1
30.0
13,272.7
190.4
1,745.1

1.0
-11.2
0.3
3.5
0.0
5.6
0.1
0.7

2002–2022
ha
%
-35,638.5
33,384.5
1,265.6
6,543.1
200.5
-6,799.2
47.9
996.1

-15.0
14.0
0.5
2.8
0.1
-2.9
0.0
0.4

JPSL, 15(5) | 810

LULC in the Banggai Islands Regency was assessed based on classifications from 2002, 2012, and 2022,
highlighting the extent of changes in LULC. Table 4 illustrates these changes, revealing that the decline in land
use was inconsistent. This is because farmers clear agricultural land in a mobile manner, and after harvesting,
they will leave the land for eight years before reopening it. Naturally, the abandoned land will become
overgrown with wild plants and will be detected as forest, as well as other land uses. For instance, the forest
area decreased by 16% from 2002 to 2012, followed by a slight increase of 1% from 2012 to 2022. Similarly,
the plantation area increased by 25.3% during 2002–2012 but decreased by 11.2% in the subsequent period.
The reduction in plantation area can be attributed to population growth, which frequently leads to the
conversion of plantation land to non-plantation uses. Furthermore, local government policies have not
sufficiently supported the plantation sector in Banggai Islands Regency. Juniyanti et al. [66] notes that a
significant factor contributing to the decline in plantation land is the government's lack of emphasis on
plantations and agriculture, as policies prioritize other sectors. This shift can diminish farmers' motivation to
sustain their plantations or agricultural land.
The transitions in open land and shrubs also exhibited variability, attributed to shifting agricultural practices,
with changes of 0.8% and 8.4% for open land and shrubs in 2002–2012, and 3.5% and 5.6% in 2012–2022,
respectively. In contrast, residential land steadily increased over the 20 years, growing from 529.7 ha (0.2%)
in 2002 to 1,795.3 ha (0.8%) by 2022. Paddy fields experienced significant growth as well, with an increase of
170.5 ha from 2002 to 2012 and an additional 30 ha from 2012 to 2022. Mangrove areas, however, displayed
unstable changes, decreasing by 142.5 ha between 2002 and 2012 due to unsustainable harvesting for
firewood, but rebounding with an increase of 190.4 ha from 2012 to 2022. Water bodies in the study area
experienced a decrease of 0.3% from 2002 to 2012, followed by an increase of 0.7% between 2012 and 2022.
This fluctuation is attributed to seasonal droughts lasting from ten months to a year, during which the lake's
water levels drop significantly. However, the lake replenishes during the rainy season, leading to an increase
in water levels.
Land Use and Land Cover (LULC) Change in Banggai Islands Regency from 2002 to 2022
This study analyzed LULC changes over two years, using 2002 as the baseline and 2022 as the endpoint. The
initial analysis of land use changes from 2002 to 2022 produced a matrix detailing the increases and decreases
in the area of each land use type, illustrating the transitions between different land uses. The land use change
matrix is presented in Table 5.
Table 5. LULC transition matrix 2002 to 2022.
LULC
2002
1
2

2022
1
75,174.8
(31.6%)
17,817.8
(7.49%)

2
47,411.81
(19.9%)
40,008.36
(16.8%)

443.3
(0.19%)
0.36
(0.00%)
3,677.81
(1.55%)
11.45
(0.00%)
17.21
(0.01%)

1,620.93
(0.68%)
12.31
(0.01%)
14,636.22
(6.16%)
289.15
(0.12%)
214.12
(0.09%)

3
4
5
6
7
8

3
167.35
(0.07%)
469.96
(0.2%)
519.66
(0.22%)
330.28
(0.14%)
0.69
(0.00%)
330.6
(0.14%)
7.72
(0.00%)
173.45
(0.07%)

4
2,586.79
(1.09%)
4,274.16
(1.8%)

5
84.06
(0.04%)
119.32
(0.05%)

6
6,193.16
(2.6%)
7,856.97
(3.3%)

7
403.86
(0.17%)
45.58
(0.02%)

8
850.78
(0.35%)
261.22
(0.1%)

644.56
(0.27%)
3.61
(0.00%)
2,891.11
(1.22%)
1.54
(0.00%)
5.71
(0.00%)

8.42
(0.00%)
84.3
(0.04%)
19.1
(0.01%)
0.09
(0.00%)

694.36
(0.29%)
28.64
(0.01%)
4,623.9
(1.95%)
5.43
(0.00%)
28.42
(0.01%)

13.16
(0.01%)
0.65
(0.00%)
1.31
(0.00%)
1,095.78
(0.46%)
76.5
(0.03%)

155.53
(0.07%)
1.46
(0.00%)
60.21
(0.03%)
177.79
(0.07%)
1,159.83
(0.49%)

Notes: LULC = land use and land cover; 1 = forest; 2 = plantation; 3 = Built-up; 4 = open land; 5 = paddy field; 6 = shrubs; 7 = mangrove; 8 = water body.

Overall, the Banggai Islands Regency experienced a decline in forest and shrub areas from 2002 to 2022, with
reductions of 15% and 2.9%, respectively. This loss of forest land was converted into various land uses,
including water bodies, open land, mangroves, Built-up areas, plantations, rice fields, and shrubs. Forests that
have been detected to have changed into built-up land are located in the capital development expansion
area, while open land and shrubs are found in shifting plantations. In addition, the opening of new rice fields
has been linked to the conversion of forests into rice fields. Meanwhile, forest conversion to water bodies is

http://dx.doi.org/10.29244/jpsl.15.5.804

JPSL, 15(5) | 811

observed in coastal lake areas, where land transitions from forested to waterlogged regions—a phenomenon
more prevalent in island environments. In the study area, this conversion from forest to aquatic zones, such
as lakes, occurs naturally.
In line with the results of research by Juniyanti et al. [66], changes in mixed gardens, open land, and forests
can occur, and experience changes to water bodies. These areas will be detected on the coastal part of the
island around the mangrove area. As detailed in Table 4, forest land was transformed into plantations
covering 47,411.8 ha, Built-up totaling 167.4 ha, mangroves at 403.9 ha, and rice fields at 84.1 ha.
Additionally, 6,193.2 ha of forest were converted to shrubs, and 2,586.8 ha to open land. Consequently, LULC
changes in the Banggai Islands Regency varied, with areas classified as open land and shrubs in 2002 reverting
to forest by 2012, while forested areas transitioned to open land by 2022. An example of land use change in
the study area can be seen in Figure 5.

Figure 5. (a) LULC change from forest to open land; (b) LULC change from forest and plantation to developed land; (c)
LULC change from forest and open land to plantation; and (d) LULC change from mangrove and water body to
developed land.
Predicted LULC in Banggai Islands Regency in 2042
The modeling process will initially assess the model's capability to predict LULC for the target year. The
prediction for land use in 2022 was derived from the trends observed in 2002 and 2012 using the CA-Markov
simulation model. The accuracy of this prediction for the Banggai Islands Regency in 2022 was determined to
be 93%, allowing for further predictions of land use in 2042. The proportions of land use and cover areas are
presented in Figure 6 and Table 6. In contrast, the distribution map illustrating predicted land use changes
for 2042 in the Banggai Islands Regency is shown in Figure 7.

Figure 6. Projected pattern of LULC in 2042.

This journal is © Katili et al. 2025

JPSL, 15(5) | 812

Table 6. Projected land use and land cover change in Banggai Island Regency in 2042.

No

Land use and land cover

1
Forest
2
Plantation
3
Built-up
4
Open land
5
Rice fields
6
Shrubs
7
Mangroves
8
Waterbody
Total

2022
ha
%
97,148.4 40.9
104,194.0 43.8
1,795.3
0.8
10,451.4
4.4
315.2
0.1
19,430.9
8.2
1,636.8
0.7
2,671.1
1.1
237,671 100

Land area
2042
ha
%
90,570.9 38.1
120,294.0 50.6
2,910.9
1.2
10,526.6
4.4
215.7
0.1
10,239.3
4.3
1,601.8
0.7
1,312.1
0.6
237,671 100

Land change
ha
%
-6,577.5 -2.77
16,100.0
6.77
1,115.6
0.47
75.2
0.03
-99.5 -0.05
-9,191.6 -3.87
-35.0 -0.01
-1,359.0 -0.57
237,671
100

Figure 7. Predicted LULC in Banggai Islands Regency in 2042.
According to LULC predictions for 2042, the area of forest and shrubs is estimated to decrease by 6,577.5 ha
and 9,191.6 ha, respectively. Conversely, the plantation area is projected to increase by 16,100 ha due to the
conversion of forest and shrub lands. Additionally, residential land in the Banggai Islands Regency is
anticipated to grow by 1,115.6 ha, driven by population growth, with the population growth rate in Banggai
Islands Regency reaching 0.96% per year. However, paddy fields are expected to decrease by 127.6 ha, as
local farmers face challenges in rice cultivation due to high costs and low yields. Furthermore, water body
areas will decline by 1,359 ha, with much of this land transitioning to open land and shrubs. This shift can be
attributed to seasonal variations in the region; during the rainy season, lake areas may expand, causing
adjacent open land and shrubs to be classified as water bodies.
Discussion
The identification of LULC changes in the Banggai Islands Regency over the past 20 years (2002 –2022) reveals
a clear decline in forest and shrub areas, while other land uses have increased. A similar study on land use
change conducted by Yulianto et al. [7] in Tondano, North Sulawesi, found a decrease in forest and shrubland
area over 20 years. Meanwhile, research identifying land use changes conducted over 15 years found that
forest area continued to increase. This increase in forest area was due to conversion to plantations and rice
fields [9]. The most significant decline in forest area in the Banggai Islands occurred between 2002 and 2012,
primarily attributed to human activities driven by population growth, which increased land demand for

http://dx.doi.org/10.29244/jpsl.15.5.804

JPSL, 15(5) | 813

purposes such as agriculture and plantations (notably coconut and clove cultivation) and illegal logging. This
decline was exacerbated by inadequate local government oversight of forest management and land use
practices. However, from 2012 to 2022, there was a slight increase in forest area, albeit only by 1%. This
uptick was facilitated by more active government oversight and improved planning, though it still lagged
other regions. The decline in forest cover during the earlier period was largely a result of extensive
development initiatives by both central and local governments, as the region had established itself as a new
district capital. Consequently, the central government developed office facilities in the district's urban center,
leading to the conversion of land into built-up areas within Banggai Islands Regency. Additionally, the rapid
conversion of forests to plantations is driven by farmers' needs for agricultural land to support their families
[67]. The substantial decrease in forest areas is largely due to land being repurposed for agriculture, with
plantations accounting for 47,411.81 ha. This is because the indigenous people of Banggai do not hesitate to
convert forests if their income increases, resulting in a very significant decline of 15% of the total forest area.
This study is similar to research conducted by Juniyanti et al. [66] in analyzing LULC changes on Bengkalis
Island, Riau, where the sharp decline in forest cover converted to plantations reached 10% of the total.
According to Meyer and Turner [68], Halim et al. [69], Dibaba et al. [11], population growth is a critical factor
influencing this conversion, as it increases the demand for housing and agricultural activities. While the
expansion of plantations may economically benefit farmers by potentially increasing their income, the loss
of forest cover poses significant environmental risks, including damage to ecosystems and biodiversity [70]
the conversion of forested areas into other land uses can lead to detrimental impacts on the environment
and the land's ecological potential [71].
Plantations in the Banggai Islands Regency saw significant growth from 2002 to 2022, largely due to the use
of agricultural land for mixed plantations that include food and horticultural crops such as pepper, tomatoes,
taro, and Banggai yam, which are cultivated alongside clove, coconut, and nutmeg trees (Agriculture Office
of Banggai Islands Regency 2022). This expansion is driven by the annual increase in the local population,
necessitating that farmers enlarge their agricultural and plantation areas to meet their needs. The community
employs a rotational agricultural system, resulting in a relatively high presence of open land and shrubs in
the region. According to Mulyani and Agus [72], Jin et al. [73], this open land and shrub area holds
considerable economic potential if developed for agricultural purposes, which could enhance local incomes.
According to McNicol et al. [74], the shifting cultivation practices are vital for the regeneration of native
plants, contributing to the abundance of open land and shrubs. In this context, based on information from
Banggai yam farmers and agricultural extension workers, that Banggai yam farmers in Banggai Kepulauan
Regency usually return to using previously cultivated land after about ten land rotation cycles, reflecting the
tradition of shifting cultivation that persists today.
In the Banggai Islands, farmers practicing shifting agriculture believe that land left fallow for several years
can regain fertility without the need for fertilizers upon reuse. Consistent rotation and resting of land can
promote ecosystem maturity and enhance land stability over time [75]. This agricultural pattern is crucial as
it facilitates soil recovery [74]. Similarly, rice fields in the study area are mostly rainfed, which are usually
suited to lowland conditions. However, their management is often ineffective due to individual land
ownership, leading to cultivation at specific times and leaving fields idle for extended periods without
alternative crops [76]. Utilizing rain-fed paddy fields for replacement crops, such as Banggai yam, could
benefit farmers; otherwise, neglect may lead to the conversion of these fields. Additionally, mangrove areas
in the region have significantly increased from 2002 to 2022, largely due to local government initiatives since
2016 aimed at monitoring and replanting mangroves to protect the ecosystem. According to Marois and
Mitsch [77], mangroves are very important because they can withstand storm surges and wind, thus
protecting beaches, housing, and other infrastructure, such as those in the Banggai Islands Regency.
LULC in Banggai Islands Regency in 2042 is predicted to decrease in forest land use by 2.77%. This underscores
the need for effective regulation and monitoring of forest areas to prevent disturbances, crimes, and threats
[78]. Forests are crucial for oxygen production and play a vital role in maintaining water systems, preventing
floods, controlling erosion, and sequestering carbon. Additionally, locally managed forests provide significant
economic and social benefits to communities [79]. Given the critical importance of forests for the residents
of Banggai Islands, enhancing oversight of these areas is essential for ensuring their integrity and
sustainability, particularly regarding ecosystem services and biodiversity [80]. Considering the current
degradation of forest areas, such as illegal logging and land conversion, one viable alternative is to designate
these lands for agroforestry, which can create jobs, optimize land use, and increase farmers' income in the
Banggai Islands. This approach is supported by the Regulation of the Minister of Environment and Forestry
of the Republic of Indonesia Number 4 of 2023 concerning the management of social forestry in forest areas

This journal is © Katili et al. 2025

JPSL, 15(5) | 814

with special management, through social forestry with special management, including areas with potential
for agrosilvopastoral activities. Although this regulation does not specifically promote agrosilvopastoralism,
it regulates social forestry more broadly. Furthermore, changes in plantation land use are projected to
increase by 120,294 hectares (6.77%) by 2042.
Based on the Spatial Plan for the Banggai Islands Regency, agricultural areas generally cover 136,900.02 ha
of the total land area. This growth is driven by economic and social factors, including limited land resources,
population growth, and economic development [81]. In the Banggai Islands Regency, the rising population
exacerbates land scarcity, while economic growth encourages plantation activities at a pace that often
exceeds land availability. The heightened demand for plantation land, fueled by socio-economic pressures,
has led to rapid expansion in this sector [67]. Therefore, the implementation of sustainable agricultural
systems, such as agroforestry (integration of trees and agriculture), organic farming, and LEISA (Low External
Input Sustainable Agriculture), can support sustainable land use. Plantation expansion can be carried out in
tandem with biodiversity conservation through the implementation of local agroforestry practices.
Agroforestry is a system that integrates plantation crops with trees and other natural vegetation, thereby
helping to reduce deforestation, increase community resilience, and preserve biodiversity. This approach
utilises traditional practices that have long been applied by local farmers in the Banggai Islands, such as
planting various species of plants in the same field.
Additionally, the increase in settlements in the Banggai Islands is expected to reach 0.47% by 2042. Although
this growth is relatively slow compared to other regions, proactive measures are necessary to manage it
effectively. Converting land to settlements can negatively impact soil properties, particularly in terms of
water absorption. According to Kurowska et al. [82], changes from green spaces to built-up environments
can disrupt runoff and water infiltration, resulting in the loss of plantation, forest, and agricultural lands. To
mitigate land conversion, particularly of underutilized rice fields, farmers could cultivate more profitable
crops such as horticultural products, taro, and Banggai yam. Rotating crops on the same land can optimise its
use, reduce costs, and increase revenue. Furthermore, several previous studies, such as [83–85], have stated
that the application of progressive taxes on land use can encourage the maximum utilization of existing land
without the need to develop new land, thereby minimizing environmental damage. Reducing settlement
rates can also be achieved through land-saving principles, such as promoting the construction of multi-story
buildings. However, these policies require revitalization of land conversion regulations to support sustainable
development. This can be accomplished by reformulating policy directions, objectives, and instruments,
ensuring alignment between land use and spatial patterns as established by regional regulations in the
Banggai Islands Regency.

Conclusions
The results of this study achieved a kappa value of 89% for LULC in 2022, with model validation yielding a
kappa of 93%, indicating its suitability for land change predictions through 2042. Over the 20 years from 2002
to 2022, the Banggai Islands Regency experienced a 15% decrease in forest area, primarily due to human
activities such as agricultural expansion and illegal logging, particularly between 2002 and 2012, when
47,411.81 hectares were converted to plantations. Projections for 2042 indicate further declines in forest,
shrubs, rice fields, mangrove, and water body areas by 7.27%, while plantation land is expected to increase
by 6.77% due to socio-economic factors. Built-up growth is anticipated at 0.47%, necessitating proactive
management to mitigate negative impacts on soil and water absorption. Strategies to optimize land use
include planting profitable crops on land that has not been optimally utilized. The involvement of
stakeholders in implementing agroforestry systems can lead to a sustainable land management system,
providing economic, social, and environmental benefits to the community. Further research on the alignment
between LULC and the Regional Spatial Plan is essential to evaluate the effectiveness of the planning efforts
of the Banggai Islands Regency.

Author Contributions
HAK: Conceptualization, Software, Formal Analysis, Resources, Initial Drafting, Visualization; S:
Conceptualization, Methodology, Validation, Investigation, Resources, Data Curation, Writing - Review &
Editing, Supervision; FI: Conceptualization, Validation, Investigation, Writing - Review & Editing, Supervision;
W: Conceptualization, Methodology, Validation, Formal Analysis, Investigation, Resources, Data Curation,
Writing - Review & Editing, Supervision.

http://dx.doi.org/10.29244/jpsl.15.5.804

JPSL, 15(5) | 815

Conflicts of Interest
There are no conflicts to declare.

Acknowledgments
We would like to thank the Indonesian Education Scholarship (BPI) of the Ministry of Education, Culture,
Research and Technology of the Republic of Indonesia, through funding from the Education Fund
Management Agency (LPDP) and the Higher Education Financing Centre (BPPT) for providing funds for this
research (No: 1924/J5.2.3/BPI.06/10/2021).

References
1.

Suryono, H.; Kuswanto, H.; Iriawan, N. Two-phase stratified random forest for paddy growth phase
classification: a case of imbalanced data. Sustainability 2022, 14, 1–13, doi:10.3390/su142215252.

2.

Nedd, R.; Light, K.; Owens, M.; James, N.; Johnson, E.; Anandhi, A. A synthesis of land use/land cover
studies: definitions, classification systems, meta-studies, challenges and knowledge gaps on a global
landscape. Land 2021, 10, 1–30, doi:10.3390/land10090994.

3.

Anandhi, A.; Douglas-Mankin, K.R.; Srivastava, P.; Aiken, R.M.; Senay, G.; Leung, L.R.; Chaubey, I. DPSIRESA Vulnerability Assessment (DEVA) framework: synthesis, foundational overview, and expert case
studies. Trans. ASABE 2020, 63, 741–752, doi:10.13031/trans.13516.

4.

Ambarwulan, W.; Yulianto, F.; Widiatmaka, W.; Rahadiati, A.; Tarigan, S.D.; Firmansyah, I.; Hasibuan,
M.A.S. Modelling land use/land cover projection using different scenarios in the Cisadane Watershed,
Indonesia: implication on deforestation and food security. Egypt. J. Remote Sens. Sp. Sci. 2023, 26, 273–
283, doi:10.1016/j.ejrs.2023.04.002.

5.

Alam, A.; Bhat, M.S.; Maheen, M. Using landsat satellite data for assessing the land use and land cover
change in Kashmir valley. Geojournal 2020, 85, 1529–1543, doi:10.1007/s10708-019-10037-x.

6.

Yusoff, M.M. Improving the quality of life for sustainable development. IOP Conf. Ser. Earth Environ. Sci.
2020, 561, 012020, doi:10.1088/1755-1315/561/1/012020.

7.

Yulianto, F.; Prasasti, I.; Pasaribu, J.M.; Fitriana, H.L.; Zylshal; Haryani, N.S.; Sofan, P. The dynamics of
land use/land cover change modeling and their implication for the flood damage assessment in the
Tondano Watershed, North Sulawesi, Indonesia. Model. Earth Syst. Environ. 2016, 2, 1–20,
doi:10.1007/s40808-016-0100-3.

8.

BPS (Badan Pusat Statistik). Luas penutupan lahan Indonesa di dalam dan di luar kawasan hutan menurut
provinsi (ribu hektar), 2023. 2025. Available online: https://www.bps.go.id/id/statisticstable/2/MTMwNCMy/luas-penutupan-lahan-indonesa-di-dalam-dan-di-luar-kawasan-hutan-menurutprovinsi.html (accessed on 19 April 2025).

9.

Nahib, I.; Turmudi; Windiastuti, R.; Suryanta, J.; Dewi, R.S.; Lestari, S. Comparing of land change modeler
and geomod modeling for the assessment of deforestation (case study: forest area at Poso Regency,
Central Sulawesi Province). Int. J. Adv. Eng. Manag. Sci. 2018, 4, 597–607, doi:10.22161/Ijaems.4.8.4.

10. Saing, Z.; Djainal, H.; Deni, S. Land use balance determination using satellite imagery and geographic
information system: case study in South Sulawesi Province, Indonesia. Geod. Geodyn. 2021, 12, 133–
147, doi:10.1016/j.geog.2020.11.006.
11. Dibaba, W.T.; Demissie, T.A.; Miegel, K. Drivers and implications of land use/land cover dynamics in
Finchaa Catchment, Northwestern Ethiopia. Land 2020, 9, 1–20, doi:10.3390/land9040113.
12. Wang, J.; Bretz, M.; Dewan, M.A.A.; Delavar, M.A. Machine learning in modelling land-use and land
cover-change (LULCC): current status, challenges and prospects. Sci. Total Environ. 2022, 822, 153559,
doi:10.1016/j.scitotenv.2022.153559.
13. Ren, Y.; Lü, Y.; Comber, A.; Fu, B.; Harris, P.; Wu, L. Spatially explicit simulation of land use/land cover
changes: current coverage and future prospects. Earth-Science Rev. 2019, 190, 398–415,
doi:10.1016/j.earscirev.2019.01.001.

This journal is © Katili et al. 2025

JPSL, 15(5) | 816

14. Eberl, J.; Gordeeva, E.; Weber, N. The policy coherence framework approach in a multi-level analysis of
European, German and Thuringian climate policy with a special focus on land use, land-use change and
forestry (LULUCF). World 2021, 2, 415–424, doi:10.3390/world2030026.
15. Munthali, M.G.; Davis, N.; Adeola, A.M.; Botai, J.O.; Kamwi, J.M.; Chisale, H.L.W.; Orimoogunje, O.O.I.
Local perception of drivers of land-use and land-cover change dynamics across Dedza District, Central
Malawi Region. Sustainability 2019, 11, 1–25, doi:10.3390/su11030832.
16. Kullo, E.D.; Forkuo, E.K.; Biney, E.; Harris, E.; Quaye-Ballard, J.A. The impact of land use and land cover
changes on socioeconomic factors and livelihood in the Atwima Nwabiagya district of the Ashanti region,
Ghana. Environ. Challenges 2021, 5, 100226, doi:10.1016/j.envc.2021.100226.
17. Rigo, R.; Houet, T. Do land use and land cover scenarios support more integrated land use management?
Land 2023, 12, 1–19, doi:10.3390/land12071414.
18. Krawchenko, T.; Tomaney, J. The governance of land use: a conceptual framework. Land 2023, 12, 1–11,
doi:10.3390/land12030608.
19. Fardilla, A.; Septiavi, R.; Juwita, T.R.; Arba, R.M.; Sefmaliza, R.; Fanbudy, A.R.; Afdhal, T.; Aadrean;
Novarino, W. Analysis of land cover changes in Padang City using satellite imagery. J. Tanah Dan
Sumberd. Lahan 2023, 10, 353–367, doi:10.21776/ub.jtsl.2023.010.2.19.
20. Boreel, A.; Silaya, T.M.; Parera, L.R.; Latupeirissa, Y.A. Perubahan penutupan lahan spatio-temporal
landscape hutan dan dampaknya terhadap stock carbon sebagai aksi mitigasi menuju net sink 2030 di
Negeri Hutumuri. J. Hutan Pulau-Pulau Kecil 2024, 8, 61–71.
21. Olorunfemi, I.E.; Fasinmirin, J.T.; Olufayo, A.A.; Komolafe, A.A. GIS and remote sensing-based analysis of
the impacts of land use/land cover change (LULCC) on the environmental sustainability of Ekiti State,
southwestern Nigeria. Environ. Dev. Sustain. 2020, 22, 661–692, doi:10.1007/s10668-018-0214-z.
22. Twisa, S.; Buchroithner, M.F. Land-Use and land-cover (LULC) change detection in Wami River Basin,
Tanzania. Land 2019, 8, 1–15, doi:10.3390/land8090136.
23. Mutanga, O.; Kumar, L. Google Earth Engine applications. Remote Sens. 2019, 11, 1–4.
24. Mahdianpari, M.; Salehi, B.; Mohammadimanesh, F.; Homayouni, S.; Gill, E. The first wetland inventory
map of newfoundland at a spatial resolution of 10 m using Sentinel-1 and Sentinel-2 data on the Google
Earth Engine cloud computing platform. Remote Sens. 2019, 11, 1–27, doi:10.3390/rs11010043.
25. Jean, N.; Burke, M.; Xie, M.; Davis, W.M.A.; Lobell, D.B.; Ermon, S. Combining satellite imagery and
machine learning to predict poverty. Science 2016, 353, 790–794, doi:10.1126/science.aaf7894.
26. Congalton, R.G.; Green, K. Assessing the Accuracy of Remotely Sensed Data : Principles and Practices,
3rd ed.; CRC Press: Boca Raton, Florida, USA, 2019; ISBN 9780429052729.
27. Miettinen, J.; Shi, C.; Liew, S.C. Towards automated 10–30 m resolution land cover mapping in insular
South-East Asia. Geocarto Int. 2017, 34, 443–457, doi:10.1080/10106049.2017.1408700.
28. Nguyen, H.T.T.; Doan, T.M.; Tomppo, E.; Mcroberts, R.E. Land use/land cover mapping using
multitemporal Sentinel-2 imagery and four classification methods—a case study from Dak Nong,
Vietnam. Remote Sens. 2020, 12, 1–27, doi:10.3390/rs12091367.
29. Zhang, M.; Lin, H.; Wang, G.; Sun, H.; Fu, J. Mapping paddy rice using a convolutional Neural Network
(CNN) with Landsat 8 datasets in the Dongting Lake Area, China. Remote Sens. 2018, 10, 1–22,
doi:10.3390/rs10111840.
30. Ramadhani, F.; Pullanagari, R.; Kereszturi, G.; Procter, J. Mapping of rice growth phases and bare land
using Landsat-8 OLI with machine learning. Int. J. Remote Sens. 2020, 41, 8428–8452,
doi:10.1080/01431161.2020.1779378.
31. Shaharum, N.S.N.; Shafri, H.Z.M.; Ghani, W.A.W.A.K.; Samsatli, S.; Al-Habshi, M.M.A.; Yusuf, B. Oil palm
mapping over Peninsular Malaysia using Google Earth Engine and machine learning algorithms. Remote
Sens. Appl. Soc. Environ. 2020, 17, 100287, doi:10.1016/j.rsase.2020.100287.
32. Mulyaqin, T.; Kardiyono; Hidayah, I.; Ramadhani, F.; Yusron, M. Deteksi alih fungsi lahan padi sawah
menggunakan Sentinel-2 dan Google Earth Engine di Kota Serang, Provinsi Banten. J. Ilmu Pertan.
Indones. 2022, 27, 226–236, doi:10.18343/Jipi.27.2.226.

http://dx.doi.org/10.29244/jpsl.15.5.804

JPSL, 15(5) | 817

33. Elyan, E.; Gaber, M.M. A genetic algorithm approach to optimising random forests applied to class
engineered data. Inf. Sci. 2017, 384, 220–234, doi:10.1016/j.ins.2016.08.007.
34. Liu, C.; Gu, Z.; Wang, J. A hybrid intrusion detection system based on scalable K-Means+ random forest
and deep learning. IEEE Access 2021, 9, 75729–75740, doi:10.1109/ACCESS.2021.3082147.
35. Suliman, S.; Setiawan, Y.; Syartinilia. Assessing the paddy fields conversion using optical satellite
imageries: a case study in Karawang Regency, West Java. IOP Conf. Ser. Earth Environ. Sci. 2022, 950,
012092, doi:10.1088/1755-1315/950/1/012092.
36. Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32, doi:10.1023/A:1010933404324.
37. LaRocque, A.; Phiri, C.; Leblon, B.; Pirotti, F.; Connor, K.; Hanson, A. Wetland mapping with Landsat-8
OLI, Sentinel-1, ALOS-1 PALSAR, and LiDAR Data in Southern New Brunswick, Canada. Remote Sens.
2020, 12, 1–30, doi:10.3390/rs12132095.
38. Qi, N.; Yang, H.; Shao, G.; Chen, R.; Wu, B.; Xu, B.; Feng, H.; Yang, G.; Zhao, C. Mapping tea plantations
using multitemporal spectral features by harmonised Sentinel-2 and Landsat images in Yingde, China.
Comput. Electron. Agric. 2023, 212, 108108, doi:10.1016/j.compag.2023.108108.
39. Al-Shalabi, M.; Billa, L.; Pradhan, B.; Mansor, S.; Al-Sharif, A.A.A. Modelling urban growth evolution and
land-use changes using GIS based cellular automata and SLEUTH Models: the case of Sana’a
metropolitan city, Yemen. Environ. Earth Sci. 2013, 70, 425–437, doi:10.1007/s12665-012-2137-6.
40. Andries, A.; Murphy, R.J.; Morse, S.; Lynch, J. Earth observation for monitoring, reporting, and
verification within environmental land management policy. Sustainability 2021, 13, 1–34,
doi:10.3390/su13169105.
41. Wei, X.; Zhang, W.; Zhang, Z.; Huang, H.; Meng, L. Urban land use land cover classification based on GF6 satellite imagery and multi-feature optimization. Geocarto Int. 2023, 38, 1–24.
42. Pramulya, R.; Dahlan, D.; Asy’Ari, R.; Gustawan, A.H.; Dzulfigar, A.; Novita, E.; Sutrisno, A.; Rahmah, D.M.
Assessment of Gayo agroforestry coffee characteristics and carbon stock potential in Mumuger social
forestry area, Central Aceh Regency. Trees, For. People 2025, 20, 100818.
43. Teluguntla, P.; Thenkabail, P.S.; Oliphant, A.; Xiong, J.; Gumma, M.K.; Congalton, R.G.; Yadav, K.; Huete,
A. A 30-m Landsat-derived cropland extent product of Australia and China using random forest machine
learning algorithm on Google Earth Engine cloud computing platform. ISPRS J. Photogramm. Remote
Sens. 2018, 144, 325–340, doi:10.1016/J.ISPRSJPRS.2018.07.017.
44. Fariz, F.; Winarsih, T. Enhancing competitive advantage of higher education institutions through the
implementation of knowledge management and information technology. Innov. J. Soc. Sci. Res. 2024, 4,
2010–2026.
45. Simonetti, D.; Pimple, U.; Langner, A.; Marelli, A. Pan-tropical Sentinel-2 cloud-free annual composite
datasets. Data Br. 2021, 39, 107488, doi:10.1016/j.dib.2021.107488.
46. Surni; Baja, S.; Arsyad, U. Dynamics of changes in land use, land cover concerning to potential loss of
biodiversity in the downstream of Tallo watershed, South Sulawesi. In Proceedings of the Seminar
Nasional Masyarakat Biodiversitas Indonesia, Semarang, ID, 9 May 2015; pp. 1050–1055.
47. Jaya, B.; Rustiadi, E.; Fauzi, A.; Pravitasari, A.E. Application of land cover maps for comparison of
alignment of land capability between predictions of land use in 2035 based on trends and spatial
planning policies (RTRW). J. Ecol. Environ. Conserv. 2021, 27, 39–46.
48. Sivanpillai, R.; Jacobs, K.M.; Mattilio, C.M.; Piskorski, E.V. Rapid flood inundation mapping by differencing
water indices from pre- and post-flood Landsat images. Front. Earth Sci. 2021, 15, 1–11,
doi:10.1007/s11707-020-0818-0.
49. Hidayah, E.; Pranadiarso, T.; Halik, G.; Indarto, I.; Lee, W.K.; Maruf, M.F. Flood mapping based on opensource remote sensing data using an efficient band combination system. Acta Geogr. Slov. 2022, 62, 47–
62, doi:10.3986/AGS.10598.
50. Yulianto, F.; Maulana, T.; Khomarudin, M.R. Analysis of the dynamics of land use change and its
prediction based on the integration of remotely sensed data and CA-Markov model, in the upstream
Citarum Watershed, West Java, Indonesia. Int. J. Digit. Earth 2018, 12, 1151–1176.
51. Foody, G.M. Status of land cover classification accuracy assessment. Remote Sens. Environ. 2002, 80,
185–201, doi:10.1016/S0034-4257(01)00295-4.

This journal is © Katili et al. 2025

JPSL, 15(5) | 818

52. Watson, P.F.; Petrie, A. Method agreement analysis: a review of correct methodology. Theriogenology
2010, 73, 1167–1179, doi:10.1016/j.theriogenology.2010.01.003.
53. Mitsova, D.; Shuster, W.; Wang, X. A cellular automata model of land cover change to integrate urban
growth with open space conservation. Landsc. Urban Plan. 2011, 99, 141–153.
54. Keshtkar, H.; Voigt, W. A spatiotemporal analysis of landscape change using an integrated Markov chain
and cellular automata models. Model. Earth Syst. Environ. 2016, 2, 1–13, doi:10.1007/s40808-015-00684.
55. Bondansari; Widiatmaka; Machfud; Munibah, K.; Ambarwulan, W. The alternative policies to sustain the
level of sufficient domestic rice staple food supply for the population (a case study in Banyumas Regency,
Central Java, Indonesia). IOP Conf. Ser. Earth Environ. Sci. 2023, 1266, 012032, doi:10.1088/17551315/1266/1/012032.
56. Sejati, A.W.; Buchori, I.; Rudiarto, I. The spatio-temporal trends of urban growth and surface urban heat
islands over two decades in the Semarang Metropolitan Region. Sustain. Cities Soc. 2019, 46, 101432,
doi:10.1016/j.scs.2019.101432.
57. Bindajam, A.A.; Mallick, J.; Talukdar, S.; Islam, A.R.M.T.; Alqadhi, S. Integration of artificial intelligence–
based LULC mapping and prediction for estimating ecosystem services for urban sustainability: past to
future perspective. Arab. J. Geosci. 2021, 14, 1–23, doi:10.1007/s12517-021-08251-4.
58. Sang, L.; Zhang, C.; Yang, J.; Zhu, D.; Yun, W. Simulation of land use spatial pattern of towns and villages
based on CA–Markov model. Math. Comput. Model. 2011, 54, 938–943.
59. Liu, D.; Zheng, X.; Zhang, C.; Wang, H. A new temporal–spatial dynamics method of simulating land-use
change. Ecol. Modell. 2017, 350, 1–10, doi:10.1016/j.ecolmodel.2017.02.005.
60. Abbas, Z.; Yang, G.; Zhong, Y.; Zhao, Y. Spatiotemporal change analysis and future scenario of LULC using
the CA-ANN approach: a case study of the Greater Bay Area, China. Land 2021, 10, 1–26,
doi:10.3390/land10060584.
61. Indrawati, L.; Hartono; Sunarto. Klasifikasi pohon keputusan untuk kajian perubahan penggunaan lahan
Kota Semarang menggunakan citra Landsat TM/ETM+. Maj. Geogr. Indones. 2009, 23, 109–123.
62. Briassoulis, H. Analysis of land use change: Theoretical and modeling approaches. 2020. Available online:
https://researchrepository.wvu.edu/rri-web-book/3/ (accessed on 24 December 2024).
63. Jensen, J.R.; Lulla, K. Introductory digital image processing: a remote sensing perspective. Geocarto Int.
1987, 2, 65, doi:10.1080/10106048709354084.
64. Jensen, J.R. Introductory Digital Image Processing: A Remote Sensing Perspective, 2nd ed.; Prentice Hall:
Upper Saddle River, New Jersey, USA, 1995; ISBN 978-0-13-205840-7.
65. Artikanur, S.D.; Widiatmaka, W.; Setiawan, Y.; Marimin, M. Predicting sugar balance as the impact of
land-use/land-cover change dynamics in a sugarcane producing regency in East Java, Indonesia. Front.
Environ. Sci. 2022, 10, 1–10, doi:10.3389/fenvs.2022.787207.
66. Juniyanti, L.; Prasetyo, L.B.; Aprianto, D.P.; Purnomo, H.; Kartodihardjo, H. Land-use/land cover change
and its causes in Bengkalis Island, Riau Province (from 1990-2019). Journal Nat. Resour. Environ. Manag.
2020, 10, 419–435, doi:10.29244/Jpsl.10.3.419-435.
67. Kainyande, A.; Auch, E.F.; Okoni-Williams, A.D. The socio-economic contributions of large-scale
plantation forests: perceptions of adjacent rural communities in the Northern Province of Sierra Leone.
Trees, For. People 2022, 10, 100329, doi:10.1016/j.tfp.2022.100329.
68. Meyer, W.B.; Turner, B.L. Human population growth and global land-use/cover change. Annu. Rev. Ecol.
Syst. 1992, 23, 39–61.
69. Halim, M.A.; Rahman, M.M.; Hassan, M.Z. Agricultural land conversion in the sub-urban area: a case
study of Rajshahi Metropolitan City. J. Life Earth Sci. 2013, 8, 21–30.
70. Fürst, C.; Frank, S.; Witt, A.; Koschke, L.; Makeschin, F. Assessment of the effects of forest land use
strategies on the provision of ecosystem services at regional scale. J. Environ. Manage. 2013, 127, 96–
116, doi:10.1016/j.jenvman.2012.09.020.

http://dx.doi.org/10.29244/jpsl.15.5.804

JPSL, 15(5) | 819

71. Brockerhoff, E.G.; Barbaro, L.; Castagneyrol, B.; Forrester, D.I.; Gardiner, B.; González-Olabarria, J.R.;
Lyver, P.O.B.; Meurisse, N.; Oxbrough, A.; Taki, H.; et al. Forest biodiversity, ecosystem functioning and
the provision of ecosystem services. Biodivers. Conserv. 2017, 26, 3005–3035, doi:10.1007/s10531-0171453-2.
72. Mulyani, A.; Agus, F. Kebutuhan dan ketersediaan lahan cadangan untuk mewujudkan cita-cita Indonesia
sebagai lumbung pangan dunia tahun 2045. Anal. Kebijak. Pertan. 2017, 15, 1–17.
73. Jin, G.; Chen, K.; Wang, P.; Guo, B.; Dong, Y.; Yang, J. Trade-Offs in land-use competition and sustainable
land development in the North China Plain. Technol. Forecast. Soc. Change 2019, 141, 36–46,
doi:10.1016/j.techfore.2019.01.004.
74. Mcnicol, I.M.; Ryan, C.M.; Williams, M. How resilient are African Woodlands to disturbance from shifting
cultivation? Ecol. Appl. 2015, 25, 2320–2336, doi:10.1890/14-2165.1.
75. Passaris, N.; Flower, K.C.; Ward, P.R.; Cordingley, N. Effect of crop rotation diversity and windrow burning
of residue on soil chemical composition under long-term no-tillage. Soil Tillage Res. 2021, 213, 105153,
doi:10.1016/j.still.2021.105153.
76. Natuhara, Y. Ecosystem services by paddy fields as substitutes of natural wetlands in Japan. Ecol. Eng.
2013, 56, 97–106, doi:10.1016/j.ecoleng.2012.04.026.
77. Marois, D.E.; Mitsch, W.J. Coastal protection from tsunamis and cyclones provided by mangrove
wetlands – a review. Int. J. Biodivers. Sci. Ecosyst. Serv. Manag. 2015, 11, 71–83.
78. Iksanov, R.; Khalikova, O.; Vladimirov, I.; Gizzatullin, R.; Baiturina, R.; Kovshov, V.; Selezneva, A.;
Khasanova, S. Forestry violations as a global issue of legal regulation in the field of forest control and
supervision: the Krasnodar territory and the Republic of Bashkortostan case study. J. For. Sci. 2021, 67,
272–284, doi:10.17221/99/2020-JFS.
79. Ngo, T.T.H.; Nguyen, T.P.M.; Duong, T.H.; Ly, T.H. Forest - related culture and contribution to sustainable
development in the Northern Mountain Region in Vietnam. For. Soc. 2021, 5, 32–47.
80. Mori, A.S.; Lertzman, K.P.; Gustafsson, L. Biodiversity and ecosystem services in forest ecosystems: a
research agenda for applied forest ecology. J. Appl. Ecol. 2017, 54, 12–27, doi:10.1111/13652664.12669.
81. Syahza, A. The potential of environmental impact as a result of the development of palm oil plantation.
Manag. Environ. Qual. 2019, 30, 1072–1094, doi:10.1108/meq-11-2018-0190.
82. Kurowska, K.; Kryszk, H.; Marks-Bielska, R.; Mika, M.; Leń, P. Conversion of agricultural and forest land
to other purposes in the context of land protection: evidence from polish experience. Land Use Policy
2020, 95, 104614, doi:10.1016/j.landusepol.2020.104614.
83. Putra, F.; Wike; Nurani, F. Kebijakan Sosial Di Indonesia: Perspektif Dan Implementasinya; Madza Media:
Malang, ID, 2022; ISBN 978-623-377-799-5.
84. Apriani, R.N.; Inayati, I. Alternative policy on progressive tax rate on idle land in the special capital
territory of Jakarta. Jurnal Administrasi Publik 2020, 10, 127–137, doi:10.31289/jap.v10i2.3178.
85. Hidayat, W. Potensi pengenaan pajak progresif pada lahan kosong melalui pendekatan zona nilai tanah
di Kabupaten Bogor. Tunas Agraria 2025, 8, 176–191, doi:10.31292/jta.v8i2.427.

This journal is © Katili et al. 2025

JPSL, 15(5) | 820