The Feasibility of Using Vegetation Indices and Soil Texture to Predict Rice Yield
More details
Hide details
Department of Soil Science, Faculty of Agriculture and Natural Resources, Science and Research Branch, Islamic Azad University, Tehran, Iran
Department of Water Engineering, Lahijan Branch, Islamic Azad University, Lahijan, Iran
Soil and Water Research Institute, Tehran, Iran
Online publish date: 2019-01-18
Submission date: 2017-10-04
Final revision date: 2017-11-28
Acceptance date: 2017-12-08
Identifying plant-environment interactions along with remote sensing provides grounds for designing management methods as well as predicting rice yield in different conditions; accordingly, it is very helpful to use vegetation indices for identifying the vegetation and greenness of farms. The regression between the local and high-yield varieties of rice in 2012 and the NDVI, SAVI, LAI, DVI, and RVI indices derived from Landsat 7 in northern Iran indicate the superiority of the NDVI index in the flowering stage of rice. Results show that the coefficient of determination of the fitted model for local and high-yielding varieties is 0.71 and 0.70, respectively, which indicates the good consistency of the results with the regional data. We evaluated the models for the local and high-yielding varieties in crop year 2013 with RMSE of 406 and 272 kg ha-1 and NRMSE of 12% and 6%, respectively. Moreover, the simulation results show that the yield of the models is well fitted with the observed values; besides, there is high correlation (R>0.80) between the real and predicted yield values. As shown by the investigation of the region’s soil texture, the fine-texture paddy fields have better yield.
Ebrahim Amiri   
IAU, Lahijan, 4481873959 Lahijan, Iran