Analisis Pola Tanam Lahan Pertanian Berbasis Citra Satelit Sentinel 2 di Kabupaten Sukoharjo

Abstract

ABSTRAK MUHAMMAD ALFANI FAUZI. Analisis Pola Tanam Lahan Pertanian Berbasis Citra Satelit Sentinel 2 di Kabupaten Sukoharjo. Dibimbing oleh LIYANTONO. Pola tanam yang diterapkan dengan tepat dapat berdampak besar terhadap peningkatan produktivitas pertanian. Untuk itu, pemantauan lahan secara efisien dan akurat menjadi kebutuhan penting. Penelitian ini dilakukan untuk mengidentifikasi pola tanam pertanian di Kabupaten Sukoharjo, Jawa Tengah, dengan memanfaatkan citra satelit Sentinel-2 yang memiliki resolusi spasial 10 meter dan temporal 5–10 hari. Data diambil selama satu musim tanam tahun 2024, kemudian diproses melalui tahap pemotongan area, masking awan, serta perhitungan indeks vegetasi seperti NDVI (Normalized Difference Vegetation Index). Analisis dilakukan menggunakan algoritma K-means clustering untuk mengelompokkan lahan berdasarkan karakteristik vegetasi. Hasil penelitian memperlihatkan adanya variasi pola tanam antar desa di Kecamatan Bendosari. Peta tematik yang dihasilkan menggambarkan dinamika spasial dan temporal lahan pertanian. Hasil klasifikasi telah dikaji secara visual dan divalidasi terbatas dengan data referensi, sehingga dapat memberikan gambaran awal yang berguna bagi perencanaan pengelolaan pertanian yang lebih efektif dan berkelanjutan di wilayah Sukoharjo.

Kata kunci: NDVI, penginderaan jauh, pola tanam, Sentinel-2

ABSTRACT MUHAMMAD ALFANI FAUZI. Analysis of Agricultural Land Cropping Patterns Based on Sentinel 2 Satellite Imagery in Sukoharjo Regency. Supervised by LIYANTONO. The right cropping pattern can have a major impact on increasing agricultural productivity. Therefore, efficient and accurate land monitoring is an important need. This study was conducted to identify agricultural cropping patterns in Sukoharjo Regency, Central Java, by utilizing Sentinel-2 satellite imagery which has a spatial resolution of 10 meters and a temporal resolution of 5–10 days. Data was taken during one planting season in 2024, then processed through the stages of area cutting, cloud masking, and calculating vegetation indices such as NDVI (Normalized Difference Vegetation Index). The analysis was carried out using the K-means clustering algorithm to group land based on vegetation characteristics. The study's findings revealed variations in cropping patterns among villages in Bendosari District. The resulting thematic maps illustrated the spatial and temporal dynamics of the agricultural land. The classification results were visually reviewed and validated with limited reference data, thus providing a useful preliminary overview for planning more effective and sustainable agricultural management in the Sukoharjo region.

Keywords: NDVI, planting pattern, remote sensing, sentinel-2