SUNIL KUMAR MEDIDA*, P. PRASUNA RANI, P. V. GEETHA SIREESHA, NARESH KUMAR AND DIKSHIKA JAKHAR
Geospatial Technology Centre, RARS, Acharya N. G. Ranga Agricultural University, Guntur-522 034 (Andhra
Pradesh), India
*(e-mail : m.sunilkumar@angrau.ac.in; Mobile : 99080 37743)
(Received : May 8, 2022; Accepted : June 03, 2022)
ABSTRACT
Spatio-temporal distribution of seasonal fallow on a real-time basis helps to understand the resources
and to develop suitable alternate crop plans to increase the cropping intensity. The extent and distribution
of seasonal fallows in Guntur and Prakasam districts of Andhra Pradesh were assessed during
the kharif season of 2020 and 2021 using multi-temporal MODIS satellite data at Geospatial Technology
Centre, ANGRAU, Guntur, Andhra Pradesh. The satellite was analyzed in a GIS environment to develop
threshold-based algorithms for identifying seasonal fallows using spectral indices and band combinations.
With the increase in soil moisture content, the reflectance of the red band decreased and reflectance of
NIR was increased. The presence of surface soil moisture negatively influenced the Modified Bare Soil
Index (MBSI). Using the combined index of NDVI and MBSI, fallow with a thick mat of grass was inseparable
from the cropped land which negatively influenced the accuracy. The saturation of red band reflectance
and NDVI in fallow with a thick mat of weeds was observed and the relation was non-linear. Out of the
tested indices, the combined Normalized Difference Vegetation Index (NDVI) and MBSI recorded the
highest overall accuracy with moderate agreement followed by the combined index of NDVI and red.
Key words : Seasonal fallows, spatio-temporal, MBSI, NDVI, threshold values