The Difference Analysis Of Citrus On-year Or Off-year Fruiting And Estimation Of Biochemical Parameters

Citrus on-year or off-year fruiting is a very common phenomenon, which directly affect the growth of citrus, and also cause the yield of fruit trees unstable in the recent, the supply of on-year exceeds demand, off-year is in short supply, the forward will make fruit trees early aging, and is vulnerable to diseases and insect pests, which makes the fruit quality poor and the fruit production decline year by year. The core impact is economic income. Therefore, it must take effective measures to overcome the phenomenon of citrus on-year or off-year fruiting, to prompt citrus annual yield, and finally the good economic benefit can be achieved. However, the malnutrition of tree is the internal cause which forms citrus on-year or off-year fruiting. Analysis of leaf nutrition is one of the important technical means to citrus nutrition analysis, the efficient plant nutrition diagnosis methods is one of hot topic researched by the nutrition workers. It is possible to extract biochemical parameters from remote sensing data due to the development of hyperspectral remote sensing technology, taking spectral reflectance data to estimate concentration or content of leaf biochemical components, which provides important theoretical basis for precision agriculture in real-time, fast, non-destructive and accurate to require crop nutrition information.This study is based on the universal phenomenon of citrus on-year or off-year fruiting, taking Beibei Jincheng orange as the research object, Spectral reflectance of citrus tree canopy were measured by Handheld Spectrometer, meanwhile the biochemical parameter of the leaves were determined in the lab. Yield data was obtained during the mature time. Severity of citrus on-year or off-year fruiting was analyzed and biochemical parameters difference of citrus on-year or off-year fruiting was analyzed. Using hyperspectral remote sensing technology to process data, the spectral variables were obtained. Both biochemical parameters differences and spectral differences of citrus on-year or off-year fruiting in the bud stage were analyzed. Combined with the main reason for the formation of citrus on-year or off-year fruiting:malnutrition, using statistical correlation analysis to analyze the relationship between leaf biochemical parameters and spectral variables in the bud stage of citrus on-year or off-year fruiting. And the prediction model of leaf biochemical parameters were established. The main research results are as follows:(1)There is a severe phenomenon of citrus on-year or off-year fruiting, continuous yield data of three years indicated that there are more than80%citrus trees exist on-year or off-year fruiting phenomenon, more than60%of citrus trees have also very serious on-year or off-year fruiting phenomenon. It has a certain relationship with the management of the test area farmers.(2) The moisture content, nitrogen,potassium,phosphorus content has been reduced from April to December. Because the spring is the period of growth and development of various organs, citrus trees need a lot of nutrients during this period, coupled with the farmers used fertilizer in early spring,which resulte each biochemical parameters higher. In addition to the potassium content, the rest of biochemical parameters of off-year are higher than on-year, it is because fruit of off-year period less than the on-year period, nutrients of tree supply to vegetative growth, resulting in leaves nutrient content of off-year higher than on-year.(3)According to nutrition diagnosis grading standards of citrus leaves and Zhou Xuewu’s research:Jincheng mineral content index:the amount of phase is2.75-3.25for nitrogen content,nitrogen content of off-year period is in excessive state in April. Nitrogen content of the rest periods are at low volume phase,so yield are not high in on-year or off-year high. The amount of phase is0.7-1.5for Potassium content, Potassium content is low in on-year and off-year period, which is not conducive to the growth of fruit,it affected fruit production. Phosphorus content appropriate stage is0.14-0.17, phosphorus content of off-year is excess in April, there is a high volume state in July, the phosphorus content of on-year period is appropriate.(4) Through doing paired samples T-test of biochemical parameters content of leaves in on-year and off-year period, the moisture content in the whole growing period are significant differences except October. There is P<0.01significance level differences of Nitrogen content in the entire period. In October and August, potassium content is no significant difference and in other months there are significant differences. Phosphorus content has no significant differences in July, there are significant differences in other months. This shows that there is a certain relationship with citrus leaves nutrient and the phenomenon of citrus on-year or off-year fruiting.(5) Biochemical parameters were analyzed in bud stage, the biggest difference is that the nitrogen content and other biochemical parameters differences are small, in addition to the potassium content,biochemical parameters content of on-year is greater than off-year. According to Zhou Xuewu’s research:Jincheng mineral content index, nitrogen and potassium content of most of the trees are not suitable value index, the phosphorus content of on-year period reach a suitable value, and it is lower than the appropriate value for the off-year period. Combineing with the previous results of the analysis, according to the content of nutrient elements of citrus leaves, it can to determine on-year or off year. If nutrient elements of leaves is high or low, it will be off-year, only nutrient elements of leaves is appropriate, it is conducive to growth and development of citrus, it may be on-year. So it is very important to estimate nutrients of citrus leaves,it is very significant to overcome the phenomenon of citrus on-year or off-year fruiting.(6)According to spectral differences analyzed in the bud stage, it can indicate:canopy spectral reflectance curve trend is consistent, but there are some differences in the reflectivity. Reflectivity of the off-year period is higher than the on-year period in the visible and near infrared bands, which can be obtained from differences of green peak and red Valley. Canopy spectral reflectance curve of off-year period is smoother than on-year period. There is a reflection peak among700-800of on-year period. After processing by differential technology, beside there is a big difference in the inflection points region, the other bands is small differences. It can be inferred that there is a big difference in the magnitude of the red edge and the red edge. Using paired sample tests and analysis of variance to analysis canopy spectral parameters of citrus bud phase:red edge position, red edge amplitude, yellow edge position, yellow edge amplitude, green peak reflectivity, red valley, green/red reflectance peak valley, green peak red valley normalized value8spectral parameters have significant differences, the other spectral parameters have no significant difference, indicating that some citrus canopy spectral data can response the biochemical parameters differences of citrus leaves. It is possible to estimation biochemical parameters of citrus leaves by citrus canopy spectral information.(7) Through the correlation analysis of citrus leaves biochemical parameters and citrus canopy hyperspectral variables in bud stage, All biochemical parameters and spectral variables are significantly correlated, the correlation of nitrogen content and spectral variables is best, which reached0.885, moisture content is second, potassium content is the lowest. Using the spectral variables and biochemical parameters to establish model and analyze the accuracy, finally to get the best models.Nitrogen content:the spectral variable is Dλ403, estimating model is y=2.617-1315X-5.554*106X2-7.486*109X3R2=0.885。Moisture content:the spectral variable is Dλ591, estimating model is y=58.4-921.7X+4.97*109X3, R2=0.638Phosphorus content:the spectral variable is Dλ605, estimating model is y=-0.05945-1642X-4239000X2-3.52*109X3, R2=0.748.Potassium content and spectrum variables reached a significant level, but the model failed to reach the level of significance,it may have a certain relationship with the. method of experimental data processing and modeling method. Conclusion show that:using hyperspectral remote sensing to estimate leaves biochemical parameters is feasible, which provides a theoretical basis for the diagnosis of citrus leaves nutrition, it is a scientific method for scientific management of Citrus on-year or off-year fruiting and controlling the phenomenon of citrus on-year or off-year fruiting.