Coupling Effects Of Different Field Management And Shading On Plant Growth And Phenological Period In Rice-wheat Rotation System

The environmental problems caused by global climate change have received more attentions.The rainy weather as a result of solar radiation weakening is one of the hot issues.Rice and wheat are two major food crops in China.Rice and wheat production is affected by solar radiation weakening,and is also closely related to field management such as water management or sowing date.In this study,a field simulating experiment was conducted to explore the coupled effects of shading and field management on the growth,yield and phenological period in rice-wheat rotation system.The canopy hyperspectral data combined with growth indexes were used for building multivariate statistical models to estimate the main phenological period.Shading intensities were set at 3 levels(natural light without shading,CK;light shading,S1;heavy shading,S2);water management in rice field experiment was set at 2 levels(conventional irrigation F,water-saving irrigation W);different sowing dates in winter wheat field experiment was set at 2 levels(conventional sowing date N,late sowing L).The main results are listed as below:(1)Growth and yield of rice and winter wheatWith increasing shading intensity,leaf area index(LAI),chlorophyll content and yield were significantly decreased,and phenological period was significantly delayed in rice.Water-saving irrigation inhibited plant height and chlorophyll content in rice under shading condition,promoted LAI and yield,advanced phenological period and shortened total growth period in rice.Shading increased chlorophyll content,decreased plant height,LAI and yield,and delayed phenophase in winter wheat.With increasing shading intensity,LAI and chlorophyll content decreased under conventional sowing condition,while LAI and chlorophyll content increased under late sowing condition.Late sowing reduced plant height,increased chlorophyll content and yield,advanced phenological period and shortened whole growth period in winter wheat under shading condition.(2)Plant height estimation in rice and winter wheat with canopy hyperspectral dataThe sensitive bands to plant height in rice and winter wheat are both visible and near-infrared bands.In the model of uniformly estimating plant height of rice and winter wheat from jointing to maturity,rice plant height model based on NDVI index has prediction accuracy of 71.6%.The model estimation accuracy for plant height based on RVI index is 74.4% in winter wheat.Among the models for estimating rice plant height at different growth stages,the models based on EVI index at rice jointing and heading stages are better,with accuracy of 72.9% and 77.7%,respectively.The estimation accuracy of plant height based on RDVI index at maturity stage is 73.6%.Modeling results based on RVI index,RDVI index and RVI index at jointing stage,heading stage and maturity stage of winter wheat are better,with accuracy of 76.8%,81.2% and 77.3%,respectively.The segmentation simulation model at different growth stages is more accurate than the unified estimation model.(3)Estimation of main phenological periods in rice and winter wheatThe EVI time series after denoising by wavelet filtering are used to estimat the heading stages in rice and winter wheat,with the accuracy rates being 97% and 98%,respectively.In the model of estimating rice maturity,the three-variable non-linear model can accurately estimate water content in rice ears and thus rice maturity.The estimation accuracy of the three-variable quadratic non-linear model based on EVI,NDWI and LAI parameters is 74.2%.In the models for estimating the maturity,the two-variable or three-variable non-linear models can accurately estimate the maturity in winter wheat.The estimation accuracy of the binary quadratic non-linear model established by EVI and LAI is 93%,same as that of the ternary quadratic non-linear model established with EVI,NDWI and LAI.