Crop Monitoring and Biomass Estimation Based on Downscaled Remote Sensing Data in AquaCrop model (Case Study: Qazvin plain, Iran)

In order to ensure food security, it is necessary to be noticed of crop conditions before harvest time. In this study, the fusion of satellite images was considered to monitor the fodder corn growth trend in three study areas. This research was carried out in three parts: remote sensing, crop modeling, and creating a connection between these two parts. In the remote sensing phase, after implementing the downscaling algorithm and producing the LAI time series, results were compared with the values estimated from Landsat 8 and MODIS images, which were overestimated in all cases and also showed a high correlation of 95%. In the crop modeling section, AquaCrop model was first calibrated and implemented in each growth stage based on the measured observation data in each field, the accuracy of the simulated model was checked, according to the results of the Statistical indicators. The model was calibrated with high accuracy (NRMSE=10% and RMSE=0.03 (ton/ha)) at a significant level of 95% and was associated with underestimation. To relay on 70% of data relationship between the downscaled LAI and the calibrated CC (Crop Canopy) was estimated, using the SVM decision support algorithm and then validated by the other 30% remaining data (R2=0.99, NRMSE=0.01). Consequently, CC was predicted. Finally, biomass values ​​were compared with the observed biomass values. According to the results of statistical indicators (RMSE=0.19 (Ton/ha), NRMSE=0.01, R2=0.96), the accuracy in biomass estimation was high, and there was a high correlation between observed and remote sensing biomass values. Therefore, the accuracy of the investigated model and method is reliable based on statistical results and can be used to simulate and estimate biomass before harvesting.