TY - JOUR
T1 - Growth and yield monitoring of potato crop using Sentinel‑1 data through cloud computing
AU - Singha, Chiranjit
AU - Swain, Kishore
AU - Jayasuriya, Hemanatha
PY - 2022/9/26
Y1 - 2022/9/26
N2 - Agricultural crops required continuous monitoring, to protect them from climatic hazards and unusual stresses ensuring better crop yield. Potato crop is very sensitive to water stress which affects both tuber quality and yield. The cloud-free, high-resolution, freely distributed Sentinel 1 SAR data was used for near real-time monitoring of potato (Solanum tuberosum L.) crop in West Bengal, India. The image analysis was carried out for various phonological stages of potato crop to evaluate VH and VV backscattering values in cloud-based Google Earth Engine platform (GEE) in 50 farm plots. The VH values were found more suitable for monitoring the crop at different growth stages soundly aligned with groundtruthing based field photographs. Sentinel 1 data was validated against the optical sensor-based Sentinel 2-NDVI values along with time series model, showing positivity for field crop applications. Receiver operating characteristic (AUROC) curves along with the potato yield maps further enhanced the suitability of backscattered images for continuous monitoring of the crop even under overcast weather in the study area. In the process, better tuber quality and crop yield can be ensured, providing higher returns to the farmers.
AB - Agricultural crops required continuous monitoring, to protect them from climatic hazards and unusual stresses ensuring better crop yield. Potato crop is very sensitive to water stress which affects both tuber quality and yield. The cloud-free, high-resolution, freely distributed Sentinel 1 SAR data was used for near real-time monitoring of potato (Solanum tuberosum L.) crop in West Bengal, India. The image analysis was carried out for various phonological stages of potato crop to evaluate VH and VV backscattering values in cloud-based Google Earth Engine platform (GEE) in 50 farm plots. The VH values were found more suitable for monitoring the crop at different growth stages soundly aligned with groundtruthing based field photographs. Sentinel 1 data was validated against the optical sensor-based Sentinel 2-NDVI values along with time series model, showing positivity for field crop applications. Receiver operating characteristic (AUROC) curves along with the potato yield maps further enhanced the suitability of backscattered images for continuous monitoring of the crop even under overcast weather in the study area. In the process, better tuber quality and crop yield can be ensured, providing higher returns to the farmers.
KW - Sentinel 1 · Cloud computing · NDVI · VV and VH · GEE · Potato crop
UR - https://link.springer.com/article/10.1007/s12517-022-10844-6
M3 - Article
SN - 1866-7511
VL - 15
JO - Arabian Journal of Geosciences
JF - Arabian Journal of Geosciences
IS - 1567
M1 - 1567
ER -