An innovative approach to estimate carbon status for improved crop load management in apple

Abstract

Inconsistent cropping is a major issue in apple fruit production. Consequently, crop load management is critical for growers. Unbalanced crop loads can lead to the establishment of biennial bearing with subsequent high economic losses. An international collaborative project between Australia and Germany was set up to investigate the various aspects of crop load management utilising standard (physiological and molecular) and innovative (image analysis and thinning predictors through modelling) methods. The modelling aspect of the project will investigate a simplified MaluSim model to determine tree carbon status with the aim of providing advice for application of chemical thinners. The focus of the modelling effort will be on early season fruit development as chemical thinners are typically applied up to 80 days after green tip. The first step in evaluating a simplified model is the conversion of key parts of MaluSim into a more flexible format using the software program R. Another aspect of the project is to efficiently measure the physiology of the apple tree through state-of-the-art stereo image reconstruction and improve the accuracy in the carbon balance status. Stereo image reconstruction involves multi-view geometry to detect a number of different variables automatically (including spurs, shoots, flowers, fruit and leaf area) and will considerably reduce manual measurements. This paper will discuss the innovative methods being used and some of the results generated so far.