Almond tree training

Two outcomes are sought. Firstly, an assessment of the 2-dimensional fruiting space concept, and what may be required mechanically to remove nuts. Secondly, more insight into the balance between each variety’s natural growth and fruiting habit and the vigour imparted by each rootstock.

Radical tree spacings and training configurations are being trialled for other Prunus crops such as peaches and plums.

While the drivers for those trials in those crops are different to the drivers the almond industry contends with, some of the issues that need to be solved or managed are the same — viz. maximising reproductive capacity and managing tree shape — although harvesting methods are likely to be very different.  Irrespective of the way nuts are removed from the tree, the industry’s aim to avoid dropping the nuts on the ground will probably necessitate some form of catching system, most likely catcher plates deflecting around tree trunks. Minimising the number of times that those plates need to be deflected would improve the efficiency of detached nut capture.  An added benefit would be that fewer nuts would be left on the orchard floor, and so less over-winter food for insect pests.  So, the goal here then is to try to reconcile the need to maximise yield potential with the need to minimise potential inefficiencies in the nut catching system.  The aim is to have the same number of vertical shoots per metre of row to maximise the number of potential fruiting sites, but to vary the number of leaders.  This means that each vertical shoot may arise from a single tree (i.e. a central leader), or from a horizontally trained shoot.  The ability of each rootstock to supply enough water and mineral nutrients will be critical here because demand for water and nutrients will increase as the number of vertical shoots each root system is trying to support increases.  It goes without saying that row and tree shape will be maintained by hedging.  The natural tendency of each variety to throw new shoots following pruning, and for those shoots to produce spurs or terminal flowers will be important responses.

Video transcript: Tree training introduction. Dr Micheal Treeby, Irymple, September 2018

We're looking at all sorts of fairly radical tree training ideas and that's a trial that is very much, you know right out on the fringe, if you like. Remember that the Orchard of the future may well be a two dimensional fruiting space. By that we mean, probably more or less like a grape vine is at the moment. It is really just height and length, and at the moment almond orchards are more or less a three dimensional fruiting space. They've got height certainly, but they've got width and length as well and some depth, if you like. So, we're moving away from that in order ultimately to shake and catch. The industry wants to move away from shake onto the ground, which is an uncontrolled space that potentially produces some, or results in some issues for them, to a situation where they shake and catch the fruit, so it doesn't go onto the ground. It removes that problem. So, we've been looking at a row crop, very much a row crop but the dimensions or the layout of the orchard is the bit we don't know. And the the light experiment, if you like, trial, is trying to provide the design specifications, for want of another term.

Transcript: Update on research into Almond tree trellising with Dr Michael Treeby, December 2019

OK, we’re calling this a tree trellising trial, but basically, the story behind it is that we know how productive the most productive orchards are, currently. We know where the fruit or what sort of structures the fruit are produced on. And what we've got to try and do is find a way to maximize the number of those structures on trees for the future. So, what we're trying to do is explore ways of increasing the number of spurs that, and that's the structure we're talking about by investigating different training techniques. And one way we can do that is to try and alter the natural structure of the tree, such that it produces multiple leaders.

A tree's natural structure is just to send a shoot skyward and then everything else sort of hangs off that. We’re trying to send multiple shoots, skyward and all those shoots coming off the sides will be producing spurs, and that's what we need to do. We need to produce about 9,000 spurs per meter of row, tree row, if you like, on a two-dimensional fruiting space and that's the kind of structure we're looking at for the future.

So, as we do that, as we go into a two-dimensional fruiting space, and we have multiple leaders, we might have some idea of how many spurs are actually possible.

The other bit of the equation is how close can we put those rows together? And at the moment, these rows are about five meters apart. That's based on the equipment that we've got, the ease with which we can use that equipment between trees.  In the future, that consideration won't be there. There'll be newer machines invented and designed so forth to cater for those closer plantings. But for the moment for us, it's really ‘how many spurs can we get on a tree? What's biologically possible, if you like, using alternative training techniques? And those alternative training techniques will alter the relationship between the, the shoot tip, which is what happens when you have a central leader just going skyward. We will alter that by pushing that central leader down and let multiple leaders come off that. So, we have a horizontal cordon on if you like, and then multiple leaders coming off. And with that, we expect that we'll be able to improve the number of spurs on a tree.

We don't know if that's going to meet the performance criteria we've identified: 9,000 per lineal meter of row. But this is what this trial will actually establish. We've also done multiple, multiple densities, so we don't know whether, if we improve the number of spurs per meter of row, whether that a single root system or multiple root systems are going to be required to drive all of that. And that's where the density comes into it.

So, in the background here, we see trees that are half a meter apart and they will have two vertical leaders coming off a horizontal cordon. There are other ones where we are two meters apart and they will have eight coming off. So that gets kind of busy and we don't know whether that root system will be able to alter that. We also know that those roots systems that are underneath, the rootstock and there are multiple rootstocks in here, imparting different levels of vigour, whether they will actually alter that relationship between the growing tip and the rest of the tree or not. And that's the tree architecture, essentially, and that’s another question we're trying to answer here.

Now, we have all the posts in here because we know it's going to be a two-dimensional fruiting structure. The surface area that that will present to wind will be quite large, and the possibility of wind just simply knocking that over means that we're going to have to train these things onto wires simply to hold that structure up. And that's why there's some fairly heavy engineering, over-engineered certainly. And that's what you see here, over engineered. We do that because it's pretty hard to go back and retrofit. Let's over engineer now and we can come back and say in the future, you, we were a bit heavy handed, we could have saved a bit of money. But if it falls over, it's too late.