Jump to Main Content
Yield estimation in commercial cranberry systems using physiological, environmental, and genetic variables
- DeVetter, Lisa, Colquhoun, Jed, Zalapa, Juan, Harbut, Rebecca
- Scientia horticulturae 2015 v.190 pp. 83-93
- Vaccinium macrocarpon, cranberries, crop yield, cultivars, environmental factors, models, plant physiology, prediction, regression analysis, Wisconsin
- Improved methods of yield prediction are desired by the cranberry industry for developing early crop pricing forecasts. However, yield is a complex trait that is influenced by multiple interacting factors involving plant physiology, the environment, and cultivar genetics. These factors and their interactions are poorly understood and this ambiguity complicates yield prediction. This study sought to improve the current understanding of yield by measuring the effects of physiological, environmental, and genetics variables on yield. Sixty-six variables were evaluated on ‘Stevens’ and ‘Ben Lear’ samples collected from eight commercial cranberry marshes located in Wisconsin during the 2011 and 2012 growing cycles. Regression analysis revealed berry number alone explained 84.5% and 91.3% of the variation associated with yield of ‘Stevens’ and ‘Ben Lear’, respectively. Models incorporating berry number and size were accurate at predicting yield (R2=0.99 for ‘Stevens’ and 0.92 for ‘Ben Lear’), yet are not useful for early crop forecasting purposes as desired by the industry. Additional regressions done to identify factors that influence berry number revealed large amounts of unexplained variation are associated with this trait. Intracultivar heterogeneity was also found to be substantial across sites and may contribute to the observed variation that could not be explained by the models. Differences in management practices could also contribute to this unexplained variation and is seldom accounted for in yield prediction studies. The main implications of our study suggests that improving the current understanding of variables influencing berry number could have positive implications on future efforts to develop more accurate methods of yield prediction for cranberry.