Future Scenario for Farmland Abandonment in Karawang Regency

Abstract

The land plays a significant role in world life to continue its journey and fulfill living things’ needs. It becomes unarguable if the land use is then defined as the varying activities executed by humans to exploit the landscape. However, the dynamic of land use has recently caused irregularities in its original behavior. A noticeable increasing trend in farmland abandonment, directly affecting farmers’ livelihood and food security. However, there is lack study in Indonesia that describe, diagnostic, and floor the prescription to encounter the issue. Essentially, the Indonesian Central Statistics Agency has shown numbers to be able to describe farmland abandoned. However, the data does not have a spatial attribute, which in the future can result in errors in analyzing especially related to the distance. Therefore, the objectives of the study is to map farmland abandonment spatial distribution, followed by classifying the driving factors, and then the last objective is to endorse the scenario in Karawang as the second largest rice-producing district in Indonesia. Remote sensing technique by the maximum likelihood method worked to map the farmland abandoned. In addition, besides employing high resolution satellite imagery as an example data, study also manages a pattern of normalize different built-up index values to validate the quality of the sample data. Random forest analysis, as a part of machine learning, is used to sort out the importance degree of driving factors. Finally, the narrative scenario is employed to floor the prescriptive to prevent farmland abandonment issue in the future. As the result, all the districts and subdistricts in Karawang Regency have abandoned land with varying numbers and more distributed more intensely in the hub of the region. The study has discovered that the total number of farmland abandonment is 28 times of the Central Bureau Statistics has claimed. 75% producer accuracy, minimum unit data collected, terminology, land ownership, and the ability of remote sensing to read all earth reflections are the possible answer to the disparity. Machine learning informed that the top three essential variables are the irrigation service, the hillside, and the spatial plan that drives farmland abandonment. Thus, developing a new irrigation service to access the remote farmland and reviewing the spatial with more flavor to the agriculture is the scenario proposed by this study, while the hillside remains a constraint.