Font Size: a A A

Models For Estimating Foliar Nutrition Elements Concentration Of Armeniaca Vulgaris Cv. Luntaibaixing Using Spectral Reflectance

Posted on:2014-10-23Degree:MasterType:Thesis
Country:ChinaCandidate:Z Z HuFull Text:PDF
GTID:2253330401454347Subject:Forest management
Abstract/Summary:PDF Full Text Request
Armeniaca vulgaris cv. Luntaibaixing is an ancient local fine variety of Luntai, native to the SouthXinjiang Basin, with dried, fresh and open apricot stone processing. It has become the main variety of thefruit industry in South Xinjiang Basin. In recent years, lacking in providing a approach for developingeffective, timely and nondestructive nutrition diagnosis for the development of A. vulgaris cv.Luntaibaixing caused generally low and unstable production and irregular quality. In this study, A. vulgariscv. Luntaibaixing was taken as the object during the flourishing time, this reach which taken the coefficientbetween foliar nutrient element concentration and spectral reflectance,either as the main line was made upof spectral characteristics, the spectral sensitive bands and models for estimating nutrient elementconcentration of A. vulgaris cv. Luntaibaixing using spectral reflectance using the method which combedwith spectral measurement and leaf nutrient element concentration measurement in door. The mainresearch results as follows:It showed that the folia spectral characteristics existed deviations which are different at variousphenological periods of fruit development of A. vulgaris cv. Luntaibaixing, also, it presented the red shiftand blue shift for the position of red and infrared edge through the study of spectral characteristics. Thereare “two peaks” phenomena for the red edg of the folia spectral reflectance of A. vulgaris cv. Luntaibaixingwhile it was not a significant deviation. The spectra slope presented the same pattern as the folia spectra. Itpresented the red shift and the red edge platform during704707nm along with growth.The spectral sensitive bands of foliar nutrient element and its concentration were different at variousphenological periods of fruit development, fruit setting period, fruit enlargement period, fruit coloringperiod, fruit post-harvest period of A. vulgaris cv. Luntaibaixing. The spectral sensitive bands of foliar Nelement at four phenological stages of fruit development were363720nm,564664nm,3101031nm,430698nm respectively. The spectral sensitive bands of foliar P element at four phenological stages offruit development were408503nm and7101055nm,424656nm,511617nm,403452nm. Thespectral sensitive bands of foliar K element at four phenological stages of fruit development were496730nm,425701nm and7211113nm,500721nm,524718nm and7491074. The spectral sensitivebands of medium elements and trace elements concentration had a deviation at various phenologicalperiods.Model test on independent samples showed that selecting the leaf sensitive bands in appropriatemathematical expression form as an independent variable, leaf nutrient element concentration as thedependent variable, with the use of power function or exponential function, which was possible to establishthe spectral estimation model of practical value for A. vulgaris cv. Luntaibaixing at different growth periods.The estimation error of leaf N, P, K concentration of nutrient elements at different growth periods wassmaller when spectral estimation model was established in the form of mathematical expressions, andderived variable of its sensitive band was selected as the independent variable. However, the estimationerror of leaf Ca, Mg, Fe, Mn, Cu and Zn concentration of nutrient elements at different growth periods wassmaller when spectral estimation model was established in the form of mathematical expressions, andfirst-order differential of double spectral sensitive band ’(R_λ) was selected as the independent variable.
Keywords/Search Tags:Armeniaca vulgaris cv. Luntaibaixing, spectral characteristics, nutrient concentration, correlation analysis, sensitive bands, regression analysis, estimate model
PDF Full Text Request
Related items