Models For Estimating Foliar Mineral Nutrition Concentrations Of Juglans Regia ’Xinxin2’ Using Spectral Reflectance

Juglans regia ‘Xinxin2’ is one of our country of South Xinjiang basin, the main economic crops, but also the development of characteristic forestry and fruit industry of walnut cultivars. Due to a large consumption of traditional technology, low efficiency, destruction of vegetation, also cannot be used in diagnosis of large area tree body nutrition status, walnut tree body nutrition imbalance caused by, or even seriously affect fruit quality and yield. Therefore, J. regia ‘Xinxin2’ was taken as the research object in this article, by J. regia ‘Xinxin2’ at various phenological periods foliar spectral characteristics of fruit development, the blades of each mineral nutrition(N, P, K, Ca, Mg, Fe, Mn, Cu, Zn) content and foliar spectral reflectance of correlation analysis, the sensitive bands screening and mineral nutrition content in foliar of high precision spectral estimation model of construction of the relevant research, for people on the tree in a timely and efficient nutrient diagnosis and rational fertilization lay theoretical basis. The main research contents and results are as follows:By J. regia ‘Xinxin2’ at various phenological periods of spectral characteristics of the study and found that foliar spectral reflectance in fruit at different phenological periods of different spectral bands have some differences and presents the specific spectral phenomenon.In the process of growth and development of J. regia ‘Xinxin2’, the foliar spectra in red edge position showed " two peaks " phenomenon, but overall not very obvious; at the same time red edge position in 705-708 nm mobile, showing the " red shift " and "blue shift " spectral phenomena. Among them, fruit-set period "red shift" phenomenon is most obvious.Through J. regia ‘Xinxin2’ walnut foliar mineral nutrition content and spectral reflectance spectra derived variables and their correlation analysis, filter out different fruits bands sensitive periods, spectral estimation model to build a power function, exponential function and so on. The results showed that J.regia ‘Xinxin2’ at various phenological periods of fruit development of a large number of mineral nutrition content and spectral variables sensitive indicator band each are not identical. During the 5 phenological periods of fruit development, the N content sensitive indicator band of the leaf is respectively 706-760 nm,310-450 nm and 720-940 nm, 335-420 nm and 550-640 nm, 310-630 nm, 710-755 nm. The P content sensitive indicator band of the leaf is respectively 310- 370 nm and 750- 1 050 nm, 720- 1 020 nm,310- 520 nm, 310- 390 nm, 390- 460 nm. And the K content sensitive indicator band is respectively310-410 nm and 710-1 130 nm, 310-400 nm, 520-566 nm and 700-960 nm, 310-430 nm and 620-706 nm, 310-390 nm and 480-610 nm, 310-360 nm and 650-685 nm. At five phenological periods of fruit development of medium volume elements(Ca, Mg) and trace elements(Fe, Mn, Cu, Zn) content of sensitive bands disparity.Independent samples were used for J. regia ‘Xinxin2’ walnut at various phenological periods of fruit development for each estimation model to test, select the appropriate mathematical expression indicating leaf spectral bands variables as independent variables to each of the foliar mineral nutrition content as the dependent variable, can Construction of J. regia ‘Xinxin2’ foliar mineral nutrition content of walnut spectrum estimation model has practical application and promotion value.The results showed that the estimation model of constructing mathematical expression for the independent variables using variables derived from the spectral reflectance estimation model, of J. regia ‘Xinxin2’ mass elements content of leaves of walnut fruit each period estimation error is small, high precision. And estimation model and the dual band spectral variables ? ’(lambda R) constructed the mathematical expression of the independent variables, the deviation at various phenological periods of fruit development foliar mineral nutrition content of trace elements is relatively small, lower than the precision of model established a large number of elements.