Multi-Temporal Analysis of Cropping Patterns and Intensity Using Optical and SAR Satellite Data for Sustaining Agricultural Production in Tamil Nadu, India

Analyzing the spatial and temporal trends in cropping patterns and intensity on a larger scale is essential for implementing timely policy decisions and strategies in response to climate change and variability. By converting cropping intensity estimates, we can compute net and gross production values, indirectly indicating food security status in the study region. This study compared the utility of optical (MOD13Q1) and SAR (Sentinel 1A) datasets for determining cropping patterns and associated intensity estimates across multiple agricultural seasons from 2019 to 2023, with spatial resolutions of 250 m and 20 m, respectively. The analysis revealed that the highest and lowest gross cropped areas using Sentinel 1A data were 55.85 lakh hectares (2022–2023) and 52.88 lakh hectares (2019–2020), respectively. For MODIS data, the highest and lowest gross cropped areas were 62.07 lakh hectares (2022–2023) and 56.87 lakh hectares (2019–2020). Similarly, the highest and lowest net sown areas using Sentinel 1A data were 43.71 lakh hectares (2022–2023) and 41.76 lakh hectares (2019–2020), and for MODIS data, the values were 48.81 lakh hectares (2022–2023) and 46.39 lakh hectares (2019–2020), respectively. Regardless of the datasets used, the highest gross and net cropped areas were reported in Tiruvannamalai district and the lowest in Kanchipuram district. Thiruvarur district reported the highest cropping intensity, while Sivagangai district had the lowest. Among all seasons, the rabi season accounted for the maximum area, followed by the kharif and summer seasons. The study concluded that single cropping (51%) was the dominant cropping pattern in Tamil Nadu, followed by double cropping (31%) and triple cropping (17%) in both datasets. Sentinel 1A data showed better performance in estimating gross and net cropped areas than optical data, with deviations ranging from 7.02% to 11.01%, regardless of the year and cropping estimates derived. The results indicated that the spatial resolution of the datasets was not a significant factor in determining cropping patterns and intensity on a larger scale. However, this may differ for smaller study areas.