Estimation Models of Above-ground Dry Matter Accumulation of Summer Maize Based on Hyperspectral Remote Sensing Vegetation Indexes

Abstract:

An onsite field experiment, which includes five nitrogen fertilizer application rate treatments, four phosphorus fertilizer application rate treatments and two summer maize cultivars treatments, was conducted at agricultural experimental station of Northwest A&F University during 2011—2014. Summer maize canopy spectral reflectance and aboveground dry matter accumulation (ADMA) were measured at the huge bellbottom stage, silking stage, filling stage and maturity stage of summer maize. 21 canopy vegetation indexes of hyperspectral remote sensing in 2011 and 2013 were chosen to establish liner, logarithmic, quadratic and exponential regression relationship between ADMA and canopy spectral parameters for each cultivar. Different regression models were applied to establish the relationship between spectrum vegetation indexes and summer maize ADMA. Three models with high coefficients and F values at each growth stage were chosen to verify root mean square error and relative error with data of canopy spectral reflectance and ADMA in 2012 and 2014 separately. The smallest root mean square error and relative error models were chosen as the best models for estimation ADMA of maize. The results show that, at the huge bellbottom stage, filling stage and maturity stage of maize, spectrum vegetation indexes for the best fitting regression relationship models with ADMA were GNDVI, PSSRc, NDVI4 and DI. These models could be used as the best models for the estimation of summer maize aboveground ADMA.