A model to predict the developmental stage of Stenocereus thurberi (Cactaceae) fruit from pulp and skin colour

Abstract

The species Stenocereus thurberi grows in the Sonoran Desert and produces a fruit known as pitaya by the local people. Recently the economic benefits of exporting fruits of several cactus species have been highlighted, indicating the importance of further studies in the pre- and post-harvest physiology of pitaya fruit. The aim of this study was to develop a model to predict the stage of development in Stenocereus thurberi fruits (pitaya), in terms of days after flowering (DAF). A randomized sampling of fruits with 5, 10, 15, 20, 25, 30, 37 and 40 DAF was done during two consecutive years. A Minolta Chroma Meter CR-300 set in the L*, a* and b* colour space was used to quantify fruit skin colour (Ls, as, bs) and fruit pulp colour (Lp, ap, bp) in every sampled fruit. A basic multiple linear model was proposed. The predictive model was analyzed using the stepwise procedure in forward selection. Several regressors were considered to fit the best model, such as the variables Ls, Lp, as, ap, bs, bp, linear and squared, as well as the chroma and angle hue for both fruit skin and fruit pulp. The F statistic, mean square error, coefficient of determination (r2), Mallows coefficient (Cp) and distribution of residuals around zero were used as indicators of the model's predictive efficiency. The best predictive model included the variables Ls, bs, Lp and bp with r2=0.795 and Cp=4.99. We concluded that it is possible to predict the stage of ripening development in DAF of Stenocereus thurberi fruit by measuring the fruit pulp and fruit skin using the L*, a* and b* space colour.