Sensitivity in the estimation of parameters fitted by simple linear regression models in the ratio of blueberry buds to fruits in Chile using percentage counting

Blueberry exporting is an important activity in Chile, with fresh blueberries commanding the highest prices and being among the most exported products to the European and North American markets. To maintain quality in the centres of consumption, farmers must continuously improve the logistics of harvesting and shipping the blueberries. Thus every year they must calculate the production of the orchard well in advance in order to hire staff and ensure the logistic cold chain. For this calculation they use a count of flower buds and a simple linear model of which the slope parameter represents the number of fruits per bud. However, due to the cost of the counting procedure, some producers count only a fraction of each plant (25%, 50% or 100%), and furthermore they do not know what effect the variety and productive age of the plants may have on the estimation. The objective of this work is to measure the impact of the cultivated variety, the age of the plant in productive years, and the percentage of fruits counted in estimating the parameter fruits per bud. The study involved monitoring 310 plants of different varieties and ages distributed in northern, central and southern Chile (over an area of approximately 700 km × 200 km). The parameter was estimated by fitting simple linear regression models (SLRM) as a function of the number of fruits and flower buds. To evaluate the impact on the parameter, the SLRM was fitted considering the variables observed in all the plants, by percentage counted, by variety and by variety-age of the plant. The major findings indicate significant differences in the estimation of the parameter, suggesting that in order to estimate fruits per bud the whole plant must be counted and its age and variety taken into account.

[1]  J. Tognetti,et al.  Analysis of highbush blueberry (Vaccinium corymbosum L.) fruit growth with exponential mixed models , 2008 .

[2]  J. Egea,et al.  Apricot flower bud development and abscission related to chilling, irrigation and type of shoots , 2003 .

[3]  L. Howard,et al.  Blueberry fruit response to postharvest application of ultraviolet radiation , 2008 .

[4]  D. Ruiz,et al.  Analysis of the variability and correlations of floral biology factors affecting fruit set in apricot in a Mediterranean climate , 2008 .

[5]  Bruce Harte,et al.  Postharvest shelf life extension of blueberries using a biodegradable package. , 2008, Food chemistry.

[6]  R. A. Lawrence,et al.  Lowbush blueberry quality changes in response to prepacking delays and holding temperatures , 1999 .

[7]  O. Kodad.  INFLUENCE OF GENOTYPE, YEAR AND TYPE OF FRUITING BRANCHES ON THE PRODUCTIVE BEHAVIOUR OF ALMOND , 2006 .

[8]  A. Molan,et al.  Controlled atmosphere storage of rabbiteye blueberries enhances postharvest quality aspects , 2007 .

[9]  M. P. Banados.  Expanding blueberry production into non-traditional production areas: Northern Chile and Argentina, Mexico and Spain , 2009 .

[10]  J. Egea,et al.  Influence of flower bud density, flower bud drop and fruit set on apricot productivity , 2004 .

[11]  M. P. Banados.  BLUEBERRY PRODUCTION IN SOUTH AMERICA , 2006 .

[12]  Qamar Uz Zaman,et al.  Computer vision system for wild blueberry fruit yield mapping , 2010 .

[13]  Changying Li,et al.  Gas sensor array for blueberry fruit disease detection and classification , 2010 .

[14]  Yonghua Zheng,et al.  Effect of high oxygen atmospheres on fruit decay and quality in Chinese bayberries, strawberries and blueberries , 2008 .