The Columnar Apple Tree System (CATS) combines columnar growth with higher tree densities compared to the spindle apple tree system. The former system is not yet commercially introduced, but it potentially could achieve higher fruit yields per ha compared to the currently used spindle tree apple orchard system. The most important reason for this is that the columnar shape allows for a better, more evenly distributed light interception. Still, light interception is influenced by tree spacing. Furthermore, in order to achieve high fruit yields, the intercepted light should be efficiently used to produce fruit.
This study aims to obtain the optimal plant spacing of columnar- versus spindle apple tree systems by measuring and modelling their light interception and fruit production efficiency (dry weight yield per unit of photosynthetic active radiation intercepted). Optimal plant spacing is defined as the plant spacing at which 70% of the available amount of PAR is intercepted by the canopy, since higher interception values were previously found to decrease fruit quality (especially fruit colouration) by increasing internal canopy shading.
From bloom till directly after harvest (for the columnar as well as the spindle apple trees) monthly light measurements were taken using the SunScan Canopy Analysis System underneath and within the canopy at different times during the day and two sides of the orchard row. Furthermore, leaf area index and fruit weight (fresh and dry) were measured monthly, whereas fruit colour of both tree types was determined only at harvest.
The optimal plant spacing was found to be rectangular for both apple tree systems (0.80x0.80m for the SATS and 0.3x0.3m for the CATS) according to the model ROWCROP, which resulted in a fraction PAR intercepted of 0.46 and 0.41 for the SATS and the CATS respectively (compared to 0.38 and 0.34 when using the default tree spacing (path width of 3.25 and 2.50m and row width of 0.80 and 0.30m for the SATS and the CATS respectively). However this tree spacing is not realistic considering the size of the machinery currently used in the orchards; the path width should be minimally 2.00m in order to allow this machinery to drive through. According to ROWCROP using this more realistic tree spacing (2.00x2.00m) the fraction PAR intercepted becomes 0.40 and 0.36 for the SATS and the CATS respectively. Assuming that the LAI increases linearly with the tree height, the 70% PAR interception aimed at could only be achieved at 5.00m tree height using the optimal plant spacing for both apple tree systems (path width of 0.80 and 0.3m and row width of 0.80 and 0.3m for the SATS and the CATS respectively).
The apple production efficiency was higher for the SATS than the CATS over time.