Analysis Of Canopy Characteristics About Different Genotypes Of Proso Millet

Proso millet is one of the main food crops in arid and semi-arid regions of northern China. It’s nutritious, rich in dietary fiber and minerals, it is an important resource of new health food with a broad market prospect. Reasonable canopy structure is beneficial to build high-yield crop group, improve group canopy structure, promote photosynthesis, and increase yield. Therefore, proso millet is careen crops and weak in cultivation of high resilience, especially in high-yield group canopy structure and metabolic mechanism have not been reported. It is important to explore canopy characteristics of different drought-resistant proso millet.Eight different types of drought-resistant varieties of proso millet were set as materials. The canopy temperature of different genotype proso millet, canopy chlorophyll content, PAR vertical distribution and canopy spectral reflectance characteristics under different canopy structures were discussed in this study. A proso millet canopy chlorophyll content estimation model based vegetation index was set up as an application of hyperspectral remote sensing technology on cereal crops. The study reached the following main results.(1) Under drought stress conditions, there are significant differences in different genotypes of proso millet canopy temperature. The canopy temperature of drought-resistant varieties were 0.89℃ lower than that of drought-sensitive varieties. The chlorophyll content in different leaf position were ordered as follow, the top second leaf> the top third leaf> the flag leaf. The average chlorophyll content(SPAD value) in the flag leaf, the top second leaf and the top third leaf of drought-resistant varieties were 31.61, 32.05 and 31.77, which is higher than that of drought-sensitive varieties. Compared with drought-sensitive varieties, the drought-resistant varieties intercepted more photosynthetic active radiation, increased transparency, delay the leaf senescence, increase the leaf material production time and improve the photosynthetic capacity, it helps to get a higher yield.(2) The spectral reflectance of different proso millet varieties presented a consistent trend, there were significant differences in each wavelength. In visible light, the drought-sensitive variety Nianfeng No.5(M08) showed the lowest reflectance, other varieties had no significant difference. In the near infrared, reflectance between different varieties were more significant. The reflectance of drought-resistant variety Longmi No.21(M02) was the lowest while the drought-sensitive variety Jinmi No.5(M06) was the highest. The drought-resistant variety Longmi No.8(M04) showed the highest reflectance in shortwave infrared, while the droughtsensitive variety Nianfeng No.5(M08) was the lowest. Under treatment of different densities, canopy reflectance in each treatments have no significant difference. With the growth progress going on, the red edge position of different proso millet varieties moved to shorter wavelength, which is named "blue shift", and the red edge amplitude decreases.(3)The trend of correlation coefficient curve were basically the same and exhibited relatively stable positive correlation in jointing stage, the flowering stage and the early filling stage. The correlation coefficient curve of the other three growth stages also showed relatively similar trend. In the near-infrared band(743-950nm), the correlation coefficient for each growth period showed stable positive correlation, while the flowering was the highest. It is feasible to estimation proso millet canopy chlorophyll content using traditional vegetation index variables. The estimation model based on RVI had the highest precision. The regression equation is y=38.331+73.677x-1500.996x2+3689.252x3(y is canopy chlorophyll content, x is RVI value). By precision test, the root mean square error was 0.045, the relative percentage error was 9.58 percent. The model can predict proso millet canopy chlorophyll content much more accurately. It can be used as fast, non-destructive estimation model to get chlorophyll content of proso millet canopy.