Research On Digitizing In Original Site And Fractal Feature Of The Configuration Of Apple Trees

The configuration of fruit tree is very important to its growth and productivity, acts thekey factor of its growth and is decided by the number, length and spatial distribution ofbranches. These factors are closely related to the yield of orchard, have effect on the light in-terception of tree, hence they affact the yield and fruit quality of tree. However, there has beenfew evaluation parameters about tree configuration.13years old ‘Fuji’ apple trees trained totall spindle shape were used as experimental materials, and three dimensional digitizer asso-ciated with software PiafDigit and VegeSTAR was used to reconstruct digitizing models. Themodel was layer horizontally and cubically by50cm distance and side length respectively.Each layered canopy and each cubical canopy were added false colors. Meanwhile, the fractalfeature was studied by box-counting.2years old apple trees trained to V-shaped configurationwere used as experimental materials, and the fractal dimension numbers of two-dimensionalimages with different cultivars at different orientations were investigated employingbox-counting dimension and Photoshop image processing technology. The main conclusionsobtained were as follows:1Digital method on site could be employed to reconstruct3-dimensional model ofapple treesObtained coordinates of trees by3D digitizer and came into being3D model. The resultof comparison between the model and photo taken was satisfactory. The method of three di-mensional digitizer associated with software PiafDigit and VegeSTAR used to reconstructdigitizing mode was workable and effective. STAR of3D model was significantly positiverelative to fractal dimension and leaf area index (LAI), and negative relative to yield, hence,fractal dimension and LAI were smaller, the light interception was larger, and the yield in-creased with STAR.2STAR values can be employed to evaluate spatial light interception of apple treesThe order of STAR values of short bourse shoots, long bourse shoots, short vegetativeshoots and long vegetative shoots of different size trees was large size trees <middle trees <small trees. The order of STAR values of four types of shoots was long vegetative shoots> long bourse shoots> short vegetative shoots> short bourse shoots. It declared that STARvaried with the type and length of shoots. To the same length shoots, STAR of vegetativeshoots was larger than bourse shoots; To the same type of shoots, STAR of long shoots waslarger than short shoots. Except0～50cm layer canopy, leaf area, fruit number and PLA de-creased by the height of each layer canopy; Shaded STAR values of each layer canopy in-creased by the height of tree, which fit the increasing trends of light transmittance measuredby light quantum probes. STAR of unit cubes increased with the height of canopy. STARvalues of peripheral canopy of fruit trees were larger than that of crown.3STAR values can be employed to evaluate shaded condition of apple treesThe light interception of unshaded shoots by other types of shoots was larger than shadedones. Decreasing STAR by shading of other types of shoots of short bourse shoots and shortvegetative shoots were smaller than that of long bourse shoots and long vegetative shoots.Shaded STAR of each layered canopy was shaded by other layers. No shaded and shadedSTAR of each layer canopy varied between0.188～0.243and0.134～0.241respectively. Noshaded STAR was larger than shaded one, and decreasing STAR by shade decreased with theheight of tree, the value varied between0.097～0.001.4Fractal dimension can be employed to evaluate tree constrcture of apple treestrained to V–shaped configuratonOn the whole level, the fractal dimension numbers of trees trained to V-shaped configu-ration ranged from1.6494to1.6998and the descending order by cultivar was ‘Honggailu’>‘Fuhong Early-maturing Gala’>‘Yuhua Early-maturing Fuji’>‘Li Gala’>‘Yanfu No.6’>‘Nagafu No.2’. The fractal dimension numbers of front views ranged from1.5734to1.6312and the order by cultivar was ‘Honggailu’>‘Yuhua Early-maturing Fuji’>‘Li Gala’>‘Yanfu No.6’>‘Fuhong Early-maturing Gala’>‘Nagafu No.2’. Total fractal dimensionnumber of front views and side views of young apple trees ranged from1.6201to1.6600andthe descending order by cultivar was ‘Honggailu’>‘Yuhua Early-maturing Fuji’>‘Li Gala’>‘Fuhong Early-maturing Gala’>‘Nagafu No.2’>‘Yanfu No.6’. Tree constructure variedwith cultivars and orientations, fractal dimension numbers could be employed to quantizethese differences and to be an index of evaluating spatial characteristics of young trees.5Fractal dimension can be employed to guide training and pruning of apple treestrained to V–shaped configuratonThe fractal dimension numbers of side views with six cultivars of eastern and westernyoung apple trees trained to V-shaped configuration ranged from1.593to1.683,1.645to1.720respectively. The fractal dimension numbers of front views with the six cultivars ofeastern and western young apple trees trained to V-shaped configuration ranged from1.548to 1.627,1.568to1.657respectively. The fractal dimension numbers of front and side views ofeastern apple trees were both smaller than western ones. When conducting the training andpruning of fruit trees, the influences on the tree shape of this phenomenon should be consided,the level of pruning on western trees should be larger than eastern ones. The quantity ofshoots reserved was determined by fractal dimension. The fractal dimension could reflectthe effect of training and pruning by evaluating tree constructure. The reason why the fractaldimension can be employed to guide training and pruning of trees were that it could evaluatethe spatial distribution of trees beforehand in the condition of no-seeing by eyes or ineffec-tiveness of traditionas valuated method of tree constructure and then, conduct necessary, pur-posive and effective training and pruning on the basis of objective or optimal tree shape.