2026 年 14 巻 2 号 p. 117-134
Small islands in Indonesia still lack of spatial information on land resources to support food security related decisions. While the quality of land management decisions depends on the availability of sufficient, precise, accurate spatial information on land quality, use of a geographic information system (GIS) is crucial. The aim of this study was using a GIS to identify available land areas suitable for maize cultivation in Wakatobi Regency, Indonesia. In this study, land suitability classification was carried out based on the Framework for Land Evaluation produced by the Food and Agriculture Organization (FAO) of the United Nations. A total of 127 sampling sites within 21 land units were used for the identification of soil properties; the characteristics of these soils were then tested in the laboratory. Our analysis indicate that, land in suitability classes S2 (moderately suitable) and S3 (marginally suitable) covers a total area of 4,581 ha (44.3% of the island). However, the available areas that are also highly suitable for maize cultivation cover only 1,906.8 ha or 18.4% of the total area. The main limiting factors are the climate, especially in relation to how the amount of rainfall affects the availability of water, surface rockiness, and soil depth. The net income of farm households per hectare in the areas of suitable available land is in the range of IDR 4,844,167 to IDR 6,770,000. The results of this study can serve as guide for future land use management on this small island and help to improve local maize yields.
Food security of the whole regions in Indonesia has todate become one of the crucial issues in coping with rapid population growth, not only in main islands, but also in a considerable number of small islands. However, in small island regions, there are numerous factors that limit the amount of land that is available for the cultivation of food crops and thus affect food security; these include the rockiness of the soil surface, topographic conditions, the availability of fertile soils, and the increasing unpredictability of weather patterns. These factors tend to force islanders to rely on traditional knowledge regarding the selection of suitable crops, how to manage the land, and when best to plant crops. These have always become the reasons for low productivity of agriculture in the small islands. Therefore, a technological approach that can assist in dealing with such problems, especially in making appropriate decisions, is needed.
The use of a GIS coupled with careful ground surveys and laboratory analysis constitutes a powerful tool that can be used in the classification of land suitability, especially when dealing with large volumes of spatially georeferenced multidimensional land data. GISs can also be used to facilitate various mathematical operations and statistical functions (Han, Chen et al., 2021; Moisa, Tiye et al., 2022; Tashayo, Honarbakhsh et al., 2020) and to ensure that complex spatial problems encountered during the assessment of land suitability when processing large databases can be dealt with (Tashayo, Honarbakhsh et al., 2020).
Traditionally, the evaluation of the overall potential of a particular area of land requires a comparison between the qualities of the land and the requirements or limitations of a specific land use. A wide range of analytical methods is used for such comparisons depending on the specific purposes of the analysis as well as the availability of data sets and analytical tools (Gasser, Smith et al., 2016; Habibie, Noguchi et al., 2021; Kogo, Kumar et al., 2019; Salvacion, 2017; Wanyama, Mighty et al., 2021). The results are usually expressed as a spatial distribution of suitability classes (FAO, 1979) or land suitability indices (Baja, Chapman et al., 2002); they may be expressed in terms of economic feasibility if yields and associated economic factors were considered in the analysis (Ghazanchaii and Fariabi, 2014; Ngoune Tandzi and Mutengwa, 2019; Peng, Guan et al., 2018).
The primary aim of this study was to perform a land suitability and availability assessment in the relation to the cultivation of food crops on a small island. The study focused on Kaledupa Island, one of the main islands in the Wakatobi Regency in the province of Southeast Sulawesi, Indonesia. Among the food crops grown there, maize (Zea mays, L.) was selected for study as it has long been a staple food of the island community.
This study was carried out on Kaledupa Island (Figure 1) in Wakatobi Regency, Indonesia, one of the autonomous regions of Southeast Sulawesi Province. Tourism is the major industry in this region. The island is located about 220 km southeast of Kendari, the provincial capital. This regency is composed of many small islands, and Kaledupa Island is well known in the region as a crop-growing area. In terms of administration, Kaledupa Island is divided into two sub-districts: Kaledupa and South Kaledupa. Together, the two sub-districts consist of 28 villages and have a total population of 20,221. The total area of the island is 10,350 ha, which accounts for about 12.6% of the area of Wakatobi Regency.
The local community uses small boats for transportation from the island to the capital of the region, Wangi-Wangi City. It is comparatively difficult to reach Kaledupa Island by sea during the periods June to August and December to February because of the large ocean waves that originate in the Banda Sea to the west and the Flores Sea to the east. The most favorable seasons for reaching Wakatobi by sea and for travel within the regency are from September to November and March to May.
Wakatobi Regency is recognized as having one of the driest climate regimes in Southeast Sulawesi. The average annual rainfall for the 10-year period 2011 to 2021 was 2,179 mm (Figure 2), and the average temperature was in the range 21°C – 33°C. The island is covered by shrubs, forests, agriculture, mangroves, and settlements, with agriculture—dry-land agriculture—still being the dominant land use type due to a lack of water and the main crops being maize, cassava, and rainfed rice (Hanani, Riniwati et al., 2024). Local varieties of maize together with cassava have long been recognized as the island’s main agricultural commodities and constitute the staple foods of the local community.
The topography of Kaledupa Island varies from flat or gently sloping along the coast (slopes in the categories 0%–3% and 3%–15%) to moderately steep (15%–25% slopes) in the central, upland part of the island (Figure 3). The elevation of the island is between 0 and 150 m above sea level, with the area having an elevation of between 20 and 30 m being the largest. The geology consists of young coral rocks and alluvium originating from marine deposits. The soil types found in the study area belong to the great groups Eutropepts and Troporthents (US Soil Taxonomy classification) (USDA, 2006). The soil has the texture of a clay loam with some rocks and has a depth of 40–60 cm.
Land unit in this study was derived from GIS data layers, which consist of geology, topography, soils, and land use/land cover information. As many as 21 land units were generated in this study. Soil observations and sampling in the field were carried out by means of a flexible grid. Observations were made at predetermined points on the work map (with a scale of 1 : 10,000) and the soil sample was taken as a representative of the land unit concerned. The more variation, the more sample observations. As many as 127 samples were taken from various forms of land unit. Direct observations in the field include soil horizon division, solum depth, limiting factors of depth of plant roots, soil texture, surface and underground drainage, rocks on the surface and in the soil, slopes and land use.
Climate data was obtained from nearest weather stations, and this is considered as the limitation of this study where, only limited rainfall data available for the Kapota Island. Therefore, the data sets were carefully tabulated and spatially interpolated to fill in the gaps and so that the whole area of the study area (Kaledupa island) can be mapped for the importance of land suitability analysis.
Soil analysis was carried out both on physical and chemical characteristics from samples at 0-20 cm and 20-35 cm depth. Soil physical characteristics were directly analysed and measured in the field, except for soil texture that was analysed with the hydrometer method in the laboratory. Soil chemical properties was analysed on soil samples in the laboratory, including soil pH, availability of nutrients (N, P and K), organic matter content, and soil cation exchange capacity.
Land suitability classificationAs there are no fixed, standard criteria for land suitability assessment, the selection of criteria for a given purpose may depend on the level of detail required, the specific land use requirements, the availability of data, and factors specific to the physical environment in the study region. The qualities on which land suitability assessments are commonly based include the climate, moisture availability, terrain factors (topography), rooting conditions, nutrient availability, and soil toxicity (FAO, 1979; Peng, Guan et al., 2018; Sumani, Mujiyo et al., 2018; Sys, 1985). In Sys (1985), criteria related to the climate, topography, wetness, physical soil characteristics, soil fertility, and soil toxicity were used as suitability criteria for a variety of crop species. In Davidson, Theocharopoulos et al. (1994), a large number of land characteristics were used, but essentially these belonged to four categories: topography, nutrient retention, resistance to erosion, and toxicity. In this study, the assessment of the land suitability was based on the following equation:
LSC = f{CLI, TOP, SOC} (1)
Here LSC denotes the land suitability class, CLI denotes climatic factors (represented by the annual rainfall), TOP denotes topographic factors (represented by the slope), and SOC denotes factors related to the soil characteristics (represented by a number of physical and chemical characteristics that are described later).
In the present study, the suitability classification was undertaken at class and sub-class levels based on the FAO Framework for Land Evaluation (FAO, 1979). The class-level classification that was adopted can be described as follows.
Matching of the soil attributes and crop requirements was carried out using a simple GIS overlay and spatial query procedure (Akıncı, Özalp et al., 2013; El Baroudy, 2016). This was done to identify the spatial suitability classes and sub-classes as well as the types and extent of limiting factors for each soil mapping unit. The final suitability maps for the mapping units were then produced.
Spatial analysis of land availability for food cropsIn this study, a spatial analysis was undertaken to identify the spatial matching between the land suitability classes and spatial planning regulation layers in the study area using a simple GIS overlay technique. This was done to ensure that areas of land available for maize cultivation could be identified along with the total area and distribution of these areas. This simple overlay technique also took into account existing land use types and units where there were severe limitations on land use due to the geomorphology. The areas that were both suitable and available for maize cultivation were then generated in the GIS based on the function:
MDA = f{SPE, LUT, GEO, LSC} (2)
Here MDA denotes the area suitable for maize cultivation, SPE = spatial planning entity (formally designated by the local government), LUT = existing land use type, GEO = geomorphology, and LSC = land suitability class.
To produce a map of the land areas suitable and available for maize cultivation in a GIS thematic layer based on Equation (2), the following procedure was followed.
Step 1: overlay the LSC, SPE, LUT, and GEO thematic layers in the GIS.
Step 2: eliminate protected forest and conservation areas in the local government spatial planning GIS layer (SPE) but preserve all management zones.
Step 3: eliminate mangrove areas along the coast in the LUT layer; keep other land cover types such as mixed agriculture, shrubs, and grass.
Step 4: in the GEO layer, eliminate land units that have severe limitations due to the geomorphology, especially dense rock outcrops and units with steep slopes.
Step 5: eliminate land units in suitability classes S3 and N in the LSC layer.
Analysis of farm incomeThe primary agricultural activity data were obtained from household heads and/or members in the villages that lay within the land units that had been categorized as suitable and available for maize cultivation. A simple questionnaire was used to obtain this information in addition to direct interviews. The gross farm income (GFI) and net farm income (NFI) that would be earned from the cultivation of maize were then estimated based on the information gathered, concerning with the cultivation of maize as staple crops from selected households in the study area.
Activities related to farm production range from soil preparation to harvesting. Therefore, calculation of costs in agricultural practices consisted of the costs for seeds, fertilizer, pesticide, and the labor required for soil preparation, as well as the cost of planting, adding fertilizer, weed control, pesticide spraying, and crop harvesting. In estimating these costs, two scenarios were considered. In the first scenario (SCENARIO 1), households are assumed to follow the local government recommendations for maize cultivation; the second scenario (SCENARIO 2) is based on actual or more realistic practice. Selling price for maize was obtained from local farmers, in form of farm gate price. Market process was used as basis for comparison with various farm gate prices.
It is important to note that the GFI reflected the total value of the agricultural output plus farm program payments made by the local government. The NFI was calculated by subtracting the farm expenses from the GFI (Rossiter, 1995). Both cash and noncash income were included in the NFI from the cultivation of maize (Maroeto, Suntoro et al., 2017).
As mentioned in section 2.2, the land units used in this study were based on the interpretation of geology, soils, and land use present within the study area (RePPProT, 1988). This resulted in 21 different land units being defined (see Figure 4). Most of these units were in slightly sloping areas of karst above metamorphic rocks (marble and limestone) where slopes of 8% to 15% are dominant (see Table 1). The dominant soil types are Eutropepts and Troporthents, which have a relatively fine texture. The soil depth is in the range 25–120 cm, and the soil is mostly well drained. These areas do not experience flooding, and the amount of surface rock cover ranges from 10% to 50%.
The soil on Kaledupa Island is generally classed as neutral (pH 6.2 to 6.8), the organic carbon content is moderate, the total nitrogen content is relatively low, the amount of available phosphorus is high, the amount of potassium exchanged is relatively low, and the cation exchange capacity (CEC) is high (Table 2).
| Land unit | Average slope (%) | Soil drainage | Sand (%) | Loam (%) | Clay (%) | Texture category | Soil depth (cm) | Surface rock (%) | Area (ha) |
|---|---|---|---|---|---|---|---|---|---|
| 1 | 15–25 | well drained | 26.25 | 33.5 | 40.25 | clay | 83 | 10 | 316.0 |
| 2 | 3–8 | well drained | 31.00 | 28.00 | 42.00 | clay | 87 | 10 | 66.1 |
| 3 | 8–15 | well drained | 28.33 | 25.75 | 45.92 | clay | 83 | 18 | 398.7 |
| 4 | 3–8 | well drained | 33.25 | 21.75 | 45.00 | clay | 26 | 38 | 380.1 |
| 5 | 8–15 | well drained | 18.50 | 32.83 | 48.67 | clay | 33 | 23 | 134.9 |
| 6 | 12–15 | well drained | 17.25 | 27.00 | 55.75 | clay | 60 | 13 | 160.6 |
| 7 | 3–8 | well drained | 21.00 | 29.50 | 49.50 | clay | 43 | 20 | 156.1 |
| 8 | 30–45 | well drained | 25.00 | 30.00 | 45.00 | clay | 59 | 50 | 11.8 |
| 9 | 8–15 | well drained | 24.00 | 20.00 | 56.00 | clay | 54 | 10 | 191.4 |
| 10 | 3– 8 | well drained | 21.00 | 23.00 | 56.00 | clay | 25 | 30 | 850.4 |
| 11 | 0–3 | imperfectly drained | 33.50 | 30.00 | 36.50 | clay loam | 120 | 10 | 34.8 |
| 12 | 3–8 | well drained | 21.67 | 29.67 | 48.67 | clay | 84 | 10 | 217.0 |
| 13 | 8–15 | well drained | 25.50 | 34.00 | 40.50 | clay | 99 | 10 | 86.0 |
| 14 | 3–8 | well drained | 23.33 | 21.33 | 55.33 | clay | 27 | 47 | 496.2 |
| 15 | 8–15 | well drained | 29.25 | 27.75 | 43.00 | clay | 36 | 10 | 87.7 |
| 16 | 0–3 | moderately well drained | 22.00 | 19.00 | 59.00 | clay | 60 | 10 | 99.9 |
| 17 | 3–8 | moderately well drained | 22.00 | 26.00 | 52.00 | clay | 111 | 10 | 68.9 |
| 18 | 15–24 | well drained | 31.00 | 30.50 | 38.50 | clay loam | 57 | 20 | 115.3 |
| 19 | 15–25 | imperfectly drained | 31.00 | 29.50 | 39.50 | clay loam | 85 | 10 | 346.9 |
| 20 | 8–15 | well drained | 8.00 | 32.00 | 60.00 | clay | 31 | 20 | 281.1 |
| 21 | 15–27 | well drained | 29.00 | 25.50 | 45.50 | clay | 80 | 15 | 81.1 |
| Land unit | pH |
C (organic) (%) |
N (total) (%) |
P (available) (ppm) |
K (exchangeable) (Cmol/kg) |
CEC (Cmol/kg) |
|---|---|---|---|---|---|---|
| 1 | 6.75 | 2.50 | 0.19 | 16.83 | 0.20 | 23.50 |
| 2 | 6.88 | 2.46 | 0.15 | 21.16 | 0.29 | 25.34 |
| 3 | 6.65 | 2.56 | 0.19 | 17.91 | 0.20 | 23.84 |
| 4 | 6.76 | 2.64 | 0.20 | 16.18 | 0.23 | 24.51 |
| 5 | 6.75 | 2.67 | 0.20 | 18.52 | 0.20 | 24.02 |
| 6 | 6.79 | 2.44 | 0.18 | 18.07 | 0.28 | 24.78 |
| 7 | 6.73 | 2.75 | 0.19 | 17.83 | 0.19 | 22.91 |
| 8 | 6.81 | 2.50 | 0.20 | 14.43 | 0.24 | 25.34 |
| 9 | 6.68 | 2.80 | 0.22 | 20.14 | 0.19 | 23.54 |
| 10 | 6.48 | 2.36 | 0.21 | 20.65 | 0.21 | 23.54 |
| 11 | 6.81 | 2.54 | 0.17 | 19.99 | 0.28 | 22.54 |
| 12 | 6.56 | 2.45 | 0.19 | 17.98 | 0.19 | 25.79 |
| 13 | 6.79 | 2.36 | 0.19 | 18.33 | 0.20 | 22.54 |
| 14 | 6.76 | 2.63 | 0.20 | 19.23 | 0.25 | 25.10 |
| 15 | 6.56 | 2.64 | 0.21 | 19.21 | 0.19 | 22.83 |
| 16 | 6.49 | 2.72 | 0.22 | 20.18 | 0.29 | 25.43 |
| 17 | 6.78 | 2.26 | 0.22 | 20.02 | 0.20 | 24.76 |
| 18 | 6.75 | 2.68 | 0.20 | 20.09 | 0.24 | 25.32 |
| 19 | 6.81 | 2.66 | 0.18 | 17.25 | 0.19 | 22.43 |
| 20 | 6.75 | 2.76 | 0.20 | 22.32 | 0.22 | 23.43 |
| 21 | 6.75 | 2.69 | 0.20 | 14.83 | 0.20 | 23.38 |
The GIS database that was generated had the form of a relational database (Farbo, Sarvia et al., 2024; Quintero-Dominguez, Morell et al., 2021; Soltani, Gu et al., 2020; Zhang, X., Tian et al., 2021). In this type of database, different spatial features are decomposed into several types of basic spatial entities in the GIS data layers. Corresponding tables within the system are designed to hold spatial data according to the relational form and characteristics of soils (see Figure 5). Soil data in relational databases are stored in tables, which are topologically connected to the map layers. Each table consists of a set of rows and columns that contain sets of soil characteristics. Each column corresponds to a data type, and many databases support one or more native spatial data types.
The main benefit of the relational database model is that it provides an intuitive way of representing data and allows easy access to related data points (Quintero-Dominguez, Morell et al., 2021). As a result, relational databases are most commonly used for inventories as processes including inventory tracking to the processing of transactional data and application logging involve large amounts of structured data. However, the disadvantage of relational models is that the querying process is sometimes time-consuming and may be inefficient because multiple tables are used for specific operations (Yue and Tan, 2018). In addition to soil information such as soil characteristics and crop suitability data, the relation database developed in this study could also provide ancillary information on socio-economic conditions as well as spatial planning data for Kaledupa Island. Maps showing the lowest level of administrative boundaries were generated. Figure 6 shows a map of the 28 villages on Kaledupa Island.
The results of the land suitability analysis for Kaledupa Island indicated that all of the land belonged to just two classes—S2 (moderately suitable) and S3 (marginally suitable)—in terms of its suitability for maize growing. Twelve units were found to belong to class S2 and nine to class S3 (Table 3 and Figure 7). The main limiting factors were water availability (or climate) (wa), soil depth (rc), surface rock (lp), and slope (eh). Table 3 shows the land suitability class for each land unit along with the main limiting factors.
| Land unit | Suitability class | Suitability sub-class | Main limiting factor | Area (ha) |
|---|---|---|---|---|
| 1 | S2 | S2wa, lp | climate, surface rock | 316.0 |
| 2 | S2 | S2wa, lp | climate, surface rock | 66.1 |
| 3 | S2 | S2wa, lp | climate, surface rock | 398.7 |
| 4 | S3 | S3lp, rc | surface rock, soil depth | 380.1 |
| 5 | S3 | S3lp, rc | surface rock, soil depth | 134.9 |
| 6 | S2 | S2wa, lp | climate, surface rock | 160.6 |
| 7 | S3 | S3lp | surface rock | 156.1 |
| 8 | S3 | S3lp | surface rock | 11.8 |
| 9 | S2 | S2wa, lp | climate, surface rock | 191.4 |
| 10 | S3 | S3rc, lp | surface rock, soil depth | 850.4 |
| 11 | S2 | S2wa, lp | climate, surface rock | 34.8 |
| 12 | S2 | S2wa, lp | climate, surface rock | 217.0 |
| 13 | S2 | S2wa, lp | climate, surface rock | 86.1 |
| 14 | S3 | S3lp, rc | surface rock, soil depth | 496.2 |
| 15 | S3 | S3rc | soil depth | 87.7 |
| 16 | S3 | S2lp, r3 | surface rock, soil depth | 99.9 |
| 17 | S2 | S2wa, lp | climate, surface rock | 68.9 |
| 18 | S3 | S3eh, lp | climate, surface rock | 115.3 |
| 19 | S2 | S2wa, lp | climate, surface rock | 346.9 |
| 20 | S3 | S3lp, rc | surface rock, soil depth | 281.0 |
| 21 | S2 | S2wa, lp | climate, surface rock | 81.1 |
| Total area (ha)*) | 4,580.9 |
*) Note: other areas including land covered by large amounts of surface rock, as well as protected forests, mangroves, and residential areas were excluded from this table.
The results of the land suitability analysis indicated the “moderately suitable” class (S2) covered a total area of 1,906.8 ha and the “marginally suitable” class (S3) an area of 2,674.1 ha (see Table 3). Table 3 also shows more detailed statistics about the areas that are available and suitable for maize cultivation in the study area: it can be seen that most of the island is covered by the S2wa and S2lp sub-classes. Based on our ground truth surveys, we found that most of these areas are currently under different types of croping.
As mentioned above, the main factors limiting maize planting for class S2 areas are water availability (climate) and surface rock; this is in addition to soil depth and erosion hazard (slope) for S3 areas. These limiting factors can be dealt with using advanced management techniques. With a large increase in management input, it may be possible to promote land from class S3 to S2; with a more modest increase, promotion from class S2 to S1 may be possible.
In terms of spatial distribution, the results show that about 23.5% of the total land area of the island is suitable and available for cropping. This land is distributed among land units 1, 2, 3, 6, 9, 11, 12, 13, 17, 19, and 21. From the GIS overlay procedure, it was found that these units cover 18 out of the 28 villages on the island. It is important to note that not all of the area is currently planted with staple crops.
Farm incomeFor the 1,906.8 ha of land that had been found to be suitable and available for maize cultivation, a simple method of estimating the farm income was developed. The results of these estimates give some insight into the potential development of maize crops in the study area.
Details of the costs incurred by farmers in the production of maize were described in Section 2.5. The total cost incurred varied due to fluctuations in prices and depends on the inputs used bt farmers. In SCENARIO 1, the government recommendation was used as the basis for calculation, and the following values were used to calculate the cost of planting one ha of maize (one period of planting).
SCENARIO 2 was based on the results of the interviews with farmers. In this scenario, most farmers do not use fertilizers or pesticide, and no proper land management is used in the cultivation of staple crops. This can result in a yield loss of up to 35%–40% compared to the average yield (Dent and Young, 1981). The parameters used in the calculations under this scenario were as follows (again on a 1 ha basis).
Based on these results, the total production costs for the cultivation of one ha of maize in the study area range from IDR 4,480,000 to IDR 8,940,000, giving a GFI of between IDR 19,012,500 and IDR 29,250,000, depending on the type of land management that is adopted. The NFI (per hectare) per month is then estimated to be between IDR 4,844,167 and IDR 6,770,000. (As of October 2022; note: 1 US dollar approximately equals IDR 15,200.).
As discussed earlier, the main factors limiting the suitability and availability of land for planting maize on Kaledupa Island are the climate—as, in particular, the amount of rainfall affects the availability of water—the surface rockiness, and the soil depth. The availability of water is very important to maize cultivation even though large amounts are not required (Peng, Guan et al., 2018) as maize is very sensitive to water stress from the start of flowering to the end of seed formation (50–100 days after planting) (Huang, Duiker et al., 2015; Liu, Yu et al., 2022; Maitah, Malec et al., 2021).
The optimum moisture at soil profile for maize is at field capacity. Where there is no irrigation, this can only be supplied by rainfall (Zhou, Ozaki et al., 2017). Maize can grow in areas that receive rainfall of between 500 and 5,000 mm/year, and an optimal supply of water can be obtained in areas with rainfall of 1,000–1,500 mm/growing season (Ali, Ahmed et al., 2022; Baskin, 2022; Huang, Duiker et al., 2015; Mereu, Gallo et al., 2021). On Kaledupa Island, which covers a small area, temperatures range from 21°C to 34°C, and the average annual rainfall is 2,179 mm. However, it needs to be remembered that in the four consecutive months of July to October the rainfall is low (see Figure 1), and this limits the growth of maize. This problem can be overcome by adjusting the planting schedule to fit the rainfall pattern.
Maize can be cultivated in a range of different climates as long as the temperature does not fall below 0°C during the growth cycle (Liao, Niu et al., 2024; Liu, Yu et al., 2022), and maize can grow at any temperature in the range 14°C–40°C. However, the optimal temperature range for germination is 18°C–21°C, and optimal growth occurs at temperatures of between 18°C and 32°C (Maitah, Malec et al., 2021), with average minimum temperatures of between 12°C and 24°C and average maximum temperatures between 26°C and 29°C. In terms of temperature, the study area is categorized as suitable for maize cultivation (class S1). It is important to note that in this study climate data was obtained from nearest weather stations, and this is considered as the limitation of this study. To fill in the gaps within the study area, climate data sets were carefully tabulated and spatially interpolated to assess climate suitability analysis and mapping for maize cultivation (Zhang, J. and Zhang, 2024).
Maize can also grow in a variety of soil types, but the soil depth is a crucial criterion affecting its growth. The maximum root depth for maize is 0.60 m, which means that the soil should have an effective depth of at least 0.60 m (Djaenudin, Hidayat et al., 2003). For example, areas with impaired drainage—which can be indicated by mottling at a depth of 0.60 m—should not be used for maize cultivation. The groundwater table must be at a depth of more than 1.0 m, and any inundation of the soil surface that last for one to two days will severely limit the growth of maize (Müller, Kramer et al., 2016). The main limiting factors found in the study area were the soil depth and surface rock. The soil depth in the study area ranges from 0.25 to 1.11 m; six land units—4, 5, 10, 14, 16, and 20—were found to have the soil depth as a limiting factor.
Surface rock becomes a limiting factor for maize production if it covers a large area. The term “surface rock” covers loose surface rocks and stones as well as rock outcrops. If there is a large amount of surface rock, the movement of farmers can be affected and machinery may be damaged; in extreme cases the area available for plant growth can also be reduced. Surface rock was found to be a limiting factor over a considerable part of the study area, with 20% to 50% of several land units covered by surface rock. This amount of surface rock severely limits crop growth; it also affects tillage during the preparation of land for the cultivation of maize and other crops. Where rock outcrops are too dense, the land becomes effectively non-arable. The cut-off point between arable and non-arable rocky land is a practical issue and depends on access to machinery and how much of the land is needed for cultivation. However, in some places, surface rocks can reduce soil moisture loss and lessen the erosion risk, especially in areas prone to erosion.
The effective soil depth indicates the thickness of the soil layer that is bounded by a layer of parent material (alluvium originating from marine deposit), parent rock, sand, and/or shallow groundwater. The effective soil depth is important because it affects the ability of the soil to support plant growth and to provide nutrients and water. It is also related to the volume of soil that can be explored by plant roots (Djaenudin, Hidayat et al., 2003). Thus, the effective soil depth determines the carrying capacity of the soil for plants.
It is important to note that the GFI and NFI vary according to the land management practices used. Local government farm program payments were not mentioned during the interviews with farmers, which means that, in any development of the land for maize cultivation, it should be assumed that only limited local government payments or subsidies will be available. This will limit farmers’ ability to reduce the impact of factors limiting the land suitability and thus overall farm yields (Ampha and Wongbumru, 2025; Rossiter, 1995).
A land suitability and availability analysis was performed using a GIS to produce recommendations for local government that could be used in the planning of the development of maize cultivation on the small island of Kaledupa, Indonesia. The work included an intensive field survey, a laboratory analysis, soil characteristic and land suitability mapping, and an extensive spatial analysis using a GIS. Ground survey was also carried out to find the actual information on the costs in agriculture practices in different areas, as well as the selling prices (farm gate prices) of maize. The results of the land suitability analysis indicate that the land with suitability classes S2 (moderately suitable) and S3 (marginally suitable) coversa total area of 4,581.9 ha or 44.3% of the total area of the island. However, the area that is highly suitable for cultivation was found to cover only 1,906.8 ha, with the main limiting factors being the climate—in particular how it affects the availability of water—the surface rockiness, and the soil depth. Furthermore, based on the limitations and land suitability classes that exist in the study area, the net income per hectare of farming households in the areas of suitable and available land is in the range IDR 4,844,167 to IDR 6,770,000. It is believed that the more improvement made for the soil limiting factors, the more net income will be gained by the farmers in the Kaledupa Island. The results of this study could be used as a good guide for future land use management on this small island and should have a positive impact on local crop yields and those in other similar regions of Wakatobi Regency. Further work should focus on the analysis of the relation between land suitability and the limiting factors and crop yield in the study area, or other regions with similar land characteristics.
Conceptualization, S.B., M.J. and R. N.; methodology, S.A., and N.; software, R.M.; investigation, S.; resources, A.S.S and A. F.A.; data curation, H.P.; writing—review and editing, A.A. and H. K. B.,.; All authors have read and agreed to the published version of the manuscript.
The authors declare that they have no conflicts of interest regarding the publication of the paper.
The authors would like to thank the Directorate General for Higher Education, Ministry of Education, Culture, Research, and Technology, Republics of Indonesia for the research grant provided through a competitive scheme. Also thanks to the Wakatobi Regency government (through the Agriculture Service) for providing data and other supporting facilities for this study. Advice and support given by the staff from the Department of Soil Science, Faculty of Agriculture, Hasanuddin University, Indonesia, are highly appreciated.
