Crop Diversification Assessment in Tank Ayacut Areas of Lower Palar Sub-Basin of Chengalpattu District, Tamil Nadu, India Using Geo-Spatial Techniques
M. Vairavamani
Department of Remote Sensing and GIS, Tamil Nadu Agricultural University, Coimbatore, India.
D. Muthumanickam
Department of Remote Sensing and GIS, Tamil Nadu Agricultural University, Coimbatore, India.
S. Pazhanivelan
Centre for Water and Geospatial Studies, Tamil Nadu Agricultural University, Coimbatore, India.
R. Kumaraperumal
Department of Remote Sensing and GIS, Tamil Nadu Agricultural University, Coimbatore, India.
K. P. Ragunath
Centre for Water and Geospatial Studies, Tamil Nadu Agricultural University, Coimbatore, India.
*Author to whom correspondence should be addressed.
Abstract
For the assessment of crop diversification in the major tank Ayacut area of the Lower Palar sub-basin in Chengalpattu district of Tamil Nadu, research works were carried out using Sentinel 2 optical data by relating with ground truth data, to identify the crops in pixel-based classification and further classified the crops using Random Forest machine learning algorithms. The total area estimated under crop classification was 15767.97 and 28818.17 ha respectively for the summer seasons of 2018 and 2021. Since, the summer season experiences high crop diversification. The water spread area and water volume of tanks estimated were 612.31 and 1177.89 ha and 6,39,248 and 14,06,056 m3 respectively for 2018 and 2021. The accuracy assessment of ground truth points by confusion matrix reveals an overall classification accuracy of 96.8% (2018) and 94.9 % (2021) with kappa scores of 0.96 and 0.94 respectively. The crop diversification assessments were estimated using the Simpson Index of Diversity and values of 0.63 and 0.68 were accounted for in 2018 and 2021 respectively. The diversified pattern of crops is significantly correlated with tank water availability which increased the cropping area in 2021 as confirmed by the Crop Diversification factor.
Keywords: Crop diversification, SAR data, random forest classification, water spread, Simpson index of diversity