The Potential of Rice Field Development as a Hinterland for The New Capital City

Rice fields in North Penajam Paser Regency continue to experience conversion. The plan to move the capital city to North Penajam Paser Regency can lead to an expansion of paddy field conversion. Babulu Subdistrict is a sub-district in North Penajam Paser Regency that has the largest area of rice fields, so it has considerable potential for providing food for the new capital city. In order to forecast rice fields in 2031 using cellular automata-markov chain modeling and to examine the potential of rice fields in the hinterland area of the future capital city, this study will look at changes in land use from 2009 to 2020. Land use changes were interpreted using 2009 Landsat 5 TM and 2014 and 2020 Maxar Technologies imagery. The driving factors used in this research are distance from road, distance from settlement, distance from river, slope, and elevation. The findings revealed that the number of rice fields in the Babulu Subdistrict raised in 2014 and declined in 2020. The rice fields’ transformations took place in places with easy access and level terrain. Rice fields are predicted to decline in 2031 because they have turned into oil palm plantations. Priority III dominates the potential for rice fields in Babulu Subdistrict with regard to the suitability of land development directives. Meanwhile, Priority I has the smallest area and is in the lowlands, which tend to be flat, far from roads, close to rivers, and immediate to irrigation networks.


1.
Introduction The key engine for the nation's economy and development is the agricultural sector.According to Widyastuti (2017), this industry plays a vital role in the Indonesian economy.Rice fields are one of the land uses for the agriculture industry.The paddy fields are dwindling over time.The necessity for builtup land can increase with an increase in people, which can result in land conversion (Putra & Ismail, 2017).Land values increase annually as a result of the confluence of limited land resources and rapid population growth (Kusumastuti et al., 2018).Unchecked conversion of agricultural land can endanger the ability of the food supply to meet demand and, over time, result in social losses (Iqbal & Pusat, 2016).Kusumastuti et al. (2018) assert that the conversion of agricultural land may also be a result of the land's increased profitability.Due to the assurance of having a consistent monthly income with lower production expenses than cultivating rice, oil palm is thought to benefit farmers more than rice (Fitriyana, 2018).The declining availability of agricultural land, particularly for food crops like those 1291 (2024) 012005 IOP Publishing doi:10.1088/1755-1315/1291/1/012005 2 in North Penajam Paser Regency, is impacted by changes in land use to oil palm plantations (Widjayatnika et al., 2018).
North Penajam Paser Regency is the area designated as the new capital city.This Regency consists of four sub-districts, and Babulu Subdistrict is one of the rice granary areas in East Kalimantan.Babulu Subdistrict has considerable potential for providing the residents' food for prospective new capital city because it produces more than 50% of this Regency (BPS Kabupaten Penajam Paser Utara, 2019).Babulu Subdistrict has to be preserved because of its big land area and strong rice field output.In order to maintain food availability and satisfy the demands of the population, rice fields are crucial for the future capital city as a food source.
The Geographic Information System (GIS) can be used to determine if paddy fields will be available in the future.Utilizing different spatial and temporal scales from diverse sources, geographic information systems are useful for analysis, saving time and funds (Mugiyo, 2021).One of the GIS models that can be used to forecast future situations is the Cellular Automata-Markov Chain combination model (Quintero et Munthali et al., 2020).In its most basic form, a cellular automaton is a collection of arbitrary-shaped cells arranged in a grid-like layout, where each cell is capable of storing a variety of values over time (Singh et al., 2015;Ghosh et al., 2017).A stochastic model that depicts the potential for changes from one stage to a different one is used in conjunction with Markov Chain as a simulation tool to predict (Quintero et al., 2016;Rahnama et al., 2021).Research on landuse change and prediction using spatial modeling based on Cellular Automata-Markov chains has been carried out with various variables and other combined methods (Quintero et

2.
Material and Methods

Study Area
This study was carried out in the 45,178 ha Babulu Subdistrict, which is located in the southern region of North Penajam Paser Regency.Babulu Subdistrict is located at coordinates 116˚26'53" East Longitude and 1˚27'29" South Latitude.As shown in Figure 1, this subdistrict is made up of the Waru subdistrict to the north, the Makassar Strait to the east, the Long Kali subdistrict to the south, the Paser regency to the south, and the Long Kali and Waru subdistricts to the west.Figure 2 depicts this research workflow.ArcGIS, Google Earth Pro, IDRISI Selva, and Microsoft Excel were used to process the data.Other data needed in this study will be described in Table 1.

2.4.
Accuracy Test of Land Use Classification It is required to do the accuracy-test method after classifying land use.The accuracy test was carried out with comparative data.To compare the results of image interpretation, the accuracy test can only evaluate the most recent or real-world data (Susanti et al., 2020).High-resolution of imagery is used to compare the data, which is based on the findings of field surveys.The confusion matrix is generated to calculate the land use classification findings accuracy.(Wulansari, 2017).Fuzzy Logic for Driving Factor Processing In general, natural environmental factors, socio-economic environment, and location affect changes in paddy fields.The quality and geographic distribution of paddy fields are directly impacted by natural environmental conditions.Their area and structure alter as a result of social, economic, and ecological variables.Location influences site selection and use modes (Xia et al., 2020).The essential element that can influence land-use change is proximity or distance.The proximity of physical aspects to land use classes is a driver of land change and can be used to estimate the future (Gidey et al., 2017).The processing of driving factors is done by giving a score or weight to each variable.Each variable can affect changes in land use in a particular location.
Fuzzy logic is used to process the driving factors in the assessment of the suitability of the land, with each variable receiving a score between 0 and 1 (extremely suitable) (Salvacion, 2019;Mostafiz et al., 2021;Wang et al., 2021).Euclidean Distance tools are used to process each driving factor in accordance with its classification.After performing fuzzy membership for each variable, all processed driving factors are blended with fuzzy overlays.A good and widely accepted method can increase prediction accuracy since fuzzy membership can be determined either subjectively or objectively (Kumar & Anbalagan, 2015).According to the chosen driving factor, fuzzy overlay employs a gamma value of 0.9 to signal the proper area to alter (Ilanlou et al., 2016).The procedure of modeling using cellular automata and markov chains will make use of the outcomes of the fuzzy overlay.

2.6.
Cellular Automata-Markov Chain Processing Markov Chain is processed on IDRISI Selva software to generate transition probability matrix data.Transition matrices, transition area matrices, and conditional probabilities are produced by Markov chains.The probability of each land-use type changing to another is represented by the transition matrix.The number of pixels anticipated for the transition makes up the transition area matrix.Conditional probability reflects the probability that a specific type of land use occurs in each pixel after a specified period (Viana, 2018; Rafaai et al., 2020).The probability matrix generated in this study is the transition probability matrix for 2020 and 2031.They were next, processing the Cellular Automata-Markov Chain proposed to land use in 2020.The 2020 prediction model will be validated using the existing 2020 land use map.Validation is carried out with the Kappa accuracy test, which is needed to ensure that the model that has been made has a good accuracy value or not.The most adopted quantitative technique for evaluating the reliability of data is the kappa test.(Kang et al., 2019).Validation tests were conducted to evaluate the degree of map accuracy resulting from the land use classification method (Wulansari, 2017).The land-use changes model can be utilized for predicting land-use transformation if the model validation results are more than 70% (Nurmiaty et al., 2014;Wang et al., 2021).

Processing Directions for Development of Rice Field Potential
The degree that indicates a piece of land is suitable for a certain purpose is known as its land suitability and can be assessed according to current conditions (Nugraha et al., 2019).The suitability of the direction for developing the potential of paddy fields in Babulu Subdistrict depends on five variables: the availability of water or irrigation, elevation, slopes, river networks, and road networks.Then the method of Spatial Multicriteria Analysis (MCA) with a weighted overlay technique was carried out to combine all variables.The output of the MCA shows the extent to which the criteria are met or not in different areas, so that decisions can be made (Umar et al., 2017).In doing the weighted overlay, the scores are 1 -4, 1 for N (Not Appropriate), 2 for S3 (Marginal Appropriate), 3 for S2 (Quite Appropriate), and 4 for S1 (Very Appropriate).Each is given the same frequency, which is 20%, to produce a total frequency of 100% for the five variables.The suitability of the paddy fields that have been made is then overlaid with protected areas and predictions of paddy fields in 2031.The result is a direction for developing the potential of paddy fields in Babulu District with the classification of Priority I, Priority II, Priority Region III, and Non-Priority Area.Priority areas are created from the suitability classifications that have been made previously.3.

3.1.
Land Use Change in Babulu Subdistrict in 2009-2020 Figures 3a -3e are land use maps for 2009-2020.Land changes in Babulu Subdistrict vary; some are increasing, some are decreasing, and some are increasing and then decreasing every year.In 2009, mixed plantations dominated Babulu Subdistrict, while oil palm plantations dominated it in 2014 and 2020.
Land uses that also experience an increase in each period are settlements, other uses, and barren land.Meanwhile, barren land also experienced a considerable increase in 2020.Another use, which is newly cleared land, indicates a decrease compared to other landuses.The land use can be in mixed gardens or groves, which have decreased in each period.Meanwhile, the land uses that continued to decline in each period were groves, mixed gardens, bushland, and ponds.Unlike previous land use, paddy fields, grasslands, dry fields, and similar forests have increased and decreased.Rice fields, the third-largest area in Babulu Subdistrict, were raised in 2014 and fell in 2020.In 2009, the rice field area of 5,808 Ha (12.86%) increased to 7,056 Ha (15.62%) and again decreased to 6,844 Ha (15.15%).The decline that occurs can be due to various factors, one of which is the problem of water availability.There are still a lot of paddy fields in the Babulu Subdistrict that only rely on rainwater, so when there is a long dry season, not many rice fields can be planted.Therefore, farmers choose to convert the land into oil palm plantations or dry fields.Figure 4 shows a transformation in the area of each land use in Babulu District, except for lakes.Based on Figure 6, it can be seen that oil palm plantations will experience a rapid increase in 2031.In 2031, there will be a decrease in paddy fields along with an expansion of oil palm plantations.In 2031 paddy fields have the potential to be converted into oil palm plantations.Oil Palm plantations that have grown in number will be accompanied by a reduction in paddy fields in 2031.By 2031, 805 hectares of willow fields will have the potential to be converted to oil palm plantations.In addition, rice fields can turn into dry fields.Ponds, mixed gardens, similar forests, and shrubs will also continue to decrease in 2031.Ponds have decreased by 431 hectares, mixed gardens have covered an area of 2416 hectares, similar forebys to 2 hectares, and shrub forests by 203 hectares.A comparison of a land-use site in 2009, 2014, 2020, and model year 2031 is shown in Table 3.

Directions for Development of Rice Field Potential in Babulu Subdistrict
The results of the direction of the potential development of Babulu Subdistrict's paddy fields can be seen in Figure 6.S3 dominates the suitability of Babulu Subdistrict's paddy fields, while S1 is the smallest area.For the direction of potential development of paddy fields, the suitability is overlaid with predictions of paddy fields in 2031 and protected areas as a barrier to producing priority and non-priority areas.Meanwhile, non-priority areas dominate Labangka with an area of 924.76 hectares or 2.05%.Overall, the size of land that is not prioritized is 3,245.3acres, 7.18%.Land that is not suitable is located in the Protected Area and is adjacent to the road network.The potential of paddy fields in the hinterland area of the new capital city is dominated by S3 (Marginal Appropriate).Meanwhile, S1 (Very Appropriate) has the smallest size compared to other classes.For the direction of potential development of paddy fields, Priority I has an area of 1,515.2.83%,49 Ha or 2.83% and is spread over eight villages, namely Sebakung Jaya, Gunung Intan, Mount Makmur, Babulu Laut, Labangka Barat, and Rintik.Priority Area I is located in a low area with a flat slope, far from the road network, close to the river netwimmediately adjacentork, and immediate to the irrigation network.

Figure 2 .
Figure 2. Workflow of research This study used images from Landsat 5 TM in 2009 and Maxar Technologies in 2014 and 2020.These images are used to interpret images and classify land use by digitizing.Classification of land use used based on Kartono et al. 1989 on 1:25,000 scale.There are 13 categories of land use: forest, rice fields, fields, oil palm plantations, mixed dry fields, homogeneous forest,forests, lakes, barren land, grassland, bushland, and other uses.

Figure .
Figure .(a) Fuzzy distance from the road, (b) fuzzy distance from the settlement, (c) fuzzy

Figure .
Figure .(d) fuzzy elevation, (e) Fuzzy slope 3.3.Prediction of Land Use in Babulu Subdistrict in 2031 The Kappa validation test was carried out to determine the agreement value of the 2020 model.Before conducting the kappa validation test among the model and the existing land use in 2020, an accuracy

Figure 7 .
Figure 7. Directions for Development of Rice Field Potential in Babulu Subdistrict Priority I has the smallest area of the other classes, with an area of 1,515.49Ha.The suitability is spread across eight villages in Babulu subdistrict, namely Sebakung Jaya, Gunung Intan, Mount Makmur, Babulu Laut, Labangka Barat, and Rintik.Priority I is located in low-lying areas with flat slopes, far from road networks, close to river networks, and close to irrigation networks.Meanwhile, suitability for paddy fields is dominated by Priority III, with an area of 25,539.7 Ha or 56.53%.Priority III is spread across all villages in Babulu Subdistrict.Priority III indicates that land has a severely limiting factor that will affect land productivity.Meanwhile, non-priority areas dominate Labangka with an area of 924.76 hectares or 2.05%.Overall, the size of land that is not prioritized is 3,245.3acres, 7.18%.Land that is not suitable is located in the Protected Area and is adjacent to the road network.

4 .
ConclusionAll land use in Babulu Subdistrict from 2009 to 2020 continues to change.Paddy fields, which is the third land-use area, also experienced changes in the area.In 2009, the size of rice fields was 5,808 Ha (12.86%), then increased in 2014 to 7,056 Ha (15.62%), then decreased in 2020 to 6,844 Ha (15.15%).The results of land use predictions in Babulu Subdistrict in 2031 show that rice fields have decreased in the area.Rice fields in 2031 are predicted to have an area of 6,505 Ha or 14.4%.In 2031, paddy fields have the potential becoming plantations for the production of oil palm.

Table 2 .
Land Suitability Criteria for Rice Field Development Potential

Table 4 .
Area and Percentage of Suitability of Rice Fields in Each Village Total 1,515.493.35 8,372.5118.53 25,539.7056.53 3,245.307.18