Sukma Adi Darmawan 

Magister of Physics, Faculty of Mathematics and Natural Sciences, University of Jember, Jember, Indonesia

Bowo Eko Cahyono

Department of Physics, Faculty of Mathematics and Natural Sciences, University of Jember, Jember, Indonesia

Agus Suprianto 

Department of Physics, Faculty of Mathematics and Natural Sciences, University of Jember, Jember, Indonesia

Inas Alfiyatul Umniyah

Magister of Physics, Faculty of Mathematics and Natural Sciences, University of Jember, Jember, Indonesia

DOI: https://doi.org/10.19184/jid.v26i2.53691

ABSTRACT 

This study aims to classify crops on fragmented agricultural land by integrating radar (Sentinel-1) and optical (Sentinel-2) satellite remote sensing data. The research responds to the pressing issue of decreasing agricultural land in Jember Regency due to land conversion, which threatens food security. Feature-level fusion is applied to combine spectral indices (NDVI, NDWI, NDBI) from Sentinel-2 and radar backscatter characteristics (VV, VH) from Sentinel-1. Classification was performed using the Random Forest algorithm in the Google Earth Engine (GEE) platform. The results showed that the combination of both datasets provided high overall accuracy (81.58%) in classifying eight land cover types including agricultural crops such as paddy, corn, sugarcane, and citrus. This integration enables better monitoring of complex agricultural landscapes, offering a practical tool for sustainable land management.

Keywords: Crop classification; Fragmented agricultural land, Radar-optical fusion, Remote sensing, Google Earth Engine.