Maize area mapping using multi-temporal Sentinel 1A SAR data in the Belagavi district of Karnataka, India
Abstract
The study explores the integration of remote sensing technologies with ground truth data for precise estimation
of maize cultivation areas in the Indian Belagavi district, Karnataka, during the rabi season of 2022-23. Leveraging
Sentinel-1A satellite data and advanced processing techniques, the study provides insights into crop dynamics,
phenology, and spatial distribution. Ground truth data collection involved 369 points covering diverse land use
and land cover types. The multi-temporal Synthetic Aperture Radar (SAR) imagery underwent automated pro
cessing, extracting features crucial for maize classification. Classification accuracy assessment revealed robust
performance, with 92.4% accuracy for maize and 91.1% for non-maize locations, supported by a Kappa index of
0.83. Taluk (sub- district) wise maize area estimation highlighted spatial variations, with Saudatti emerging as the
leading taluk, contributing 25.74% of the total maize cultivation area. The study underscores the importance of
localized agricultural planning strategies tailored to each region's agricultural landscape. Through comprehensive
analysis and accurate area estimation, policymakers and stakeholders gain valuable insights for informed deci
sion-making, ranging from optimizing input distribution to formulating targeted policies for rural development.