ORIGINAL RESEARCH
The Feasibility of Using Vegetation Indices and Soil Texture to Predict Rice Yield
 
More details
Hide details
1
Department of Soil Science, Faculty of Agriculture and Natural Resources, Science and Research Branch, Islamic Azad University, Tehran, Iran
2
Department of Water Engineering, Lahijan Branch, Islamic Azad University, Lahijan, Iran
3
Soil and Water Research Institute, Tehran, Iran
CORRESPONDING AUTHOR
Ebrahim Amiri   

IAU, Lahijan, 4481873959 Lahijan, Iran
Online publish date: 2019-01-18
Publish date: 2019-03-01
Submission date: 2017-10-04
Final revision date: 2017-11-28
Acceptance date: 2017-12-08
 
Pol. J. Environ. Stud. 2019;28(4):2473–2481
KEYWORDS
TOPICS
ABSTRACT
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.
eISSN:2083-5906
ISSN:1230-1485