A mathematical model for predicting annual fertiliser requirements of kiwifruit vines

Abstract

A mathematical model for predicting fertiliser requirements in commercial kiwifruit orchards is presented. The model assumes allometric distribution of dry weight amongst vine components for all levels of productivity, and uses a prediction of fruit yield to estimate vine dry weight, annual growth and hence nutrient uptake. The efficiency of nutrient recovery is used to estimate the total nutrient inputs to the soil required to maintain an equilibrium of the soil nutrient pools. Atmospheric inputs and nutrients cycled in senesced or pruned plant material are then deducted from the total nutrient requirement to calculate a nutrient deficit or fertiliser requirement. Leaf analysis is used to detect nutrient deficiencies or excesses that can be addressed in corrective fertiliser dressings. Sensitivity analysis showed that estimates of fruit yield and the efficiency of nutrient recovery had the greatest impact on predictions of fertiliser requirement. The validity of initial estimates of these parameters is discussed. Application of the model to a mature orchard and a developing orchard revealed that nitrogen fertiliser rates much less than those recommended did not lead to evidence of nitrogen deficiency, suggesting contributions to uptake from soil nitrogen reserves and/or incorrect assessment of the efficiency of nitrogen recovery. Potassium deficiency occurred in the mature orchard following low potassium fertiliser rates, and also in the developing orchard in spite of apparently high potassium fertiliser rates, highlighting the importance of this disorder for kiwifruit. For other nutrients, deficiencies or excesses followed fertiliser rates respectively lower or higher than those recommended.