Evaluation of
optical multi-spectral satellite data for crop type and land cover identification
in Marathwada, India: a disaster management perspective
Kale Shashikant Rangnathrao, Holambe Raghunath Sambhaji and Chile Rajan Hari
Disaster Advances; Vol. 16(12); 42-54;
doi: https://doi.org/10.25303/1612da042054; (2023)
Abstract
This study evaluates the use of optical multi-spectral satellite data for crop type
and land cover identification in Marathwada, India, with a specific focus on disaster
management. The region is highly susceptible to various disasters including droughts
and other climate-related events that significantly impact agricultural productivity.
The study involves analyzing both single-date and multi-temporal satellite imagery
to develop composite images using different band combinations, aiming to identify
the most accurate combination for crop and land cover identification. A multi-class
classification approach based on random forest is employed for feature extraction
and the significance of different bands in the imagery is assessed.
The results demonstrate that a composite image composed of Red, Green, Blue, Near
Infrared and Shortwave Infrared bands yields the highest accuracy with an overall
accuracy (OA) of up to 93.69% for all land cover classes and 91.18% for crop classes
alone, using six-date multi-temporal imagery. The findings highlight the potential
of optical multi-spectral satellite data as an effective tool for crop type and
land cover identification in Marathwada, India, particularly in the context of disaster
i.e. agricultural draught management. The methodologies and results presented in
this study can serve as a valuable reference for similar research endeavors in other
agricultural draught prone regions of India and beyond.
